- Email: [email protected]
Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Available online at www.sciencedirect.com
ScienceDirect http://www.elsevier.com/locate/biombioe
Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis Dania A. Fonseca a,1, Robert Lupitskyy a,1, David Timmons b, Mayank Gupta a, Jagannadh Satyavolu a,* a b
Conn Center for Renewable Energy Research, University of Louisville, Louisville, KY 40292, USA Brown-Forman Corporation, 850 Dixie Hwy, Louisville, KY 40210, USA
article info
abstract
Article history:
The abundant availability and high level of hemicellulose content make dried distillers
Received 3 March 2014
grains (DDG) an attractive feedstock for production of pentoses (C5) and conversion of C5 to
Received in revised form
bioproducts. One target of this work was to produce a C5 extract (hydrolyzate) with high
6 October 2014
yield and purity with a low concentration of C5 degradation products. A high selectivity
Accepted 9 October 2014
towards pentoses was achieved using dilute acid hydrolysis of DDG in a percolation reactor
Available online xxx
with liquid recirculation. Pretreatment of starting material using screening and ultrasonication resulted in fractional increase of the pentose yield by 42%. A 94% yield of
Keywords:
pentoses on the DDG (280.9 g kg1) was obtained. Selective extraction of individual pen-
Dried distiller's grains
toses has been achieved by using a 2-stage hydrolysis process, resulting in arabinose-rich
Selective hydrolysis
(arabinose 81.5%) and xylose-rich (xylose 85.2%) streams. A broader impact of this work is
Integrated bio-refinery
towards an Integrated Bio-Refinery based on DDG e for production of biofuels, biochemical
Hemicellulose
intermediates, and other bioproducts.
Acid hydrolysis
© 2014 Elsevier Ltd. All rights reserved.
Degradation products
1.
Introduction
In recent years, significant research effort has been dedicated to the development of economically viable methods for conversion of lignocellulosic biomass, one of the most abundant and largely unused renewable resource on our planet, into bioproducts such as fuels and chemicals [1e3]. The biomass can be residue from farming, byproducts from agricultural and industrial processing, or from crops grown
solely for the purpose of energy production. However, biomass inherently has some drawbacks for use as starting material for fuels, chemicals, and other biomaterials [4,5]. The drawbacks include: low bulk density, leading to high storage and transportation costs; lower shelf life compared to petroleum (the biomass is dried to improve shelf life, thus adding to its cost); and its processing to sugars leaves residue, which needs to be handled. All these add to the overall cost of biomass processing, as shown in Fig. 1. In order to reduce the impact of the cost of feedstock on the overall production cost,
* Corresponding author. Conn Center for Renewable Energy Research, University of Louisville, 216 Eastern Parkway, Ernst Hall, Louisville, KY 40292, USA. Tel.: þ1 502 852 3923. E-mail address: [email protected] (J. Satyavolu). 1 Equal contribution. http://dx.doi.org/10.1016/j.biombioe.2014.10.008 0961-9534/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
2
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Zhang [5] suggested some of the top priorities of biomaterials R&D to be: 1) cost-effective release of sugars from lignocellulose and 2) co-utilization of lignocellulose components for the production of value-added compounds that subsidize whole biorefineries. Our goal is to develop an integrated approach for making bioproducts from biomass in order to contain the impact of the high feedstock cost on the overall process economics (Fig. 2). In this approach, we chose dried distillers grains (DDG) as our low cost renewable raw material. With an increasing demand for ethanol-based fuels, DDG e which is the major byproduct of dry grind corn ethanol production e is a highly desirable biomass starting material. For an estimate, about 7% by weight of corn is seed hull (pericarp) fiber biomass. The amount of corn seed hulls available in the US is estimated to be about 20 million tons and growing e a low cost and large volume opportunity. The DDG are commercially available and are used in animal feed applications [6]. Their shipping, storage, and other logistics costs are more effectively manageable compared to other traditional biomass feedstocks, such as corn stover. In addition, corn fiber has a unique composition e higher hemicellulose compared to cellulose and lower lignin [7,8] which makes this feedstock particularly attractive for pentose extraction. As shown in Table 1, corn fiber has about 4 times the amount of hemicellulose compared to that of cellulose. In addition, the lignin is only 3 %e6 % (all the % values in this work will refer to mass fraction). Such a composition of corn fibers will allow us to develop a process with higher selectivity, fewer process steps, and more importantly with lower cost. The described approach can co-exist with existing grain processing operations with a minimal economic impact on food and feed markets. The appropriate utilization of DDG to biofuels, biochemicals, and carbon fibers using an integrated approach will not only make the existing processing facilities cost efficient, but will also create new business and employment opportunities. This will also promote new opportunities for local agriculture and agricultural products. As shown in Fig. 2, our approach consists of the following steps: 1) Corn fiber from DDG is used to produce a pentose-rich hydrolyzate, 2) The pentose-rich hydrolyzate is used to produce biofuels and biochemicals, and 3) The residual fiber after
Fig. 2 e An integrated bio-refinery based on dried distillers grains (DDG).
acid hydrolysis is used in animal feed application or high value applications such as to produce carbon fibers. Pentoses (e.g., xylose and arabinose) can be biologically converted to liquid fuels, such as ethanol and butanol [9,10]. Alternatively, pentoses can be a platform for synthesis of a variety of industrially important chemicals (e.g., cyclopentadiene, cyclopentane, etc.), which are currently derived from petroleum [11]. Pentoses were identified by the US Department of Energy in 2004 among the top candidates of valuable chemical precursors which could be produced from biomass [11]. Many studies have been performed to develop processes for extraction of pentoses from a wide range of feedstocks [12e14]. Dilute acid hydrolysis is considered to be the most efficient method for selective extraction of pentoses, because hemicellulose is more easily hydrolyzed than cellulose due to amorphous structure and lack of hydrogen bonding. The primary goal of this study and the first step in our DDG biorefinery approach is to develop a process for pentose extraction from DDG that will meet several requirements: (1) high yield of pentoses, (2) high selectivity towards pentoses (maximized percent of pentoses in total monomeric sugars), (3) minimized sugar degradation, and (4) minimized acid and energy consumption. A thorough analysis of the past work by several researchers suggests that this goal can be achieved through several process steps:
1.1. Increase of fiber content in starting material via mechanical pretreatment Increasing the fiber content of DDG is typically accomplished by reducing its proteins and oil content. In the past studies reported in the literature, screening or air aspiration were
Table 1 e The composition analysis of fiber from various types of biomass.a
Fig. 1 e Factors contributing to high processing cost for bioproducts from biomass.
Composition
Corn fiber from Kernels
Softwood
Hardwood
Hemicellulose % Cellulose % Lignin % Ash
39e40 11e13 3e6 <1
25e29 40e44 25e31 <1
25e35 43e47 16e24 <1
a
[7,8].
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
3
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
evaluated [15] to increase fiber content in DDG. The fraction removed by aspiration showed a slight increase in the neutral detergent fiber content while leaving a residual fraction enriched in proteins and oil. Ultrasound pretreatment was also used to increase accessibility of lignocellulose during enzymatic hydrolysis [16] and acid hydrolysis [17].
1.2.
Selection of appropriate reactor design
Selection of an appropriate reactor configuration can significantly improve efficiency of the hydrolysis process. For example, past studies using percolation reactors showed several advantages over stir-tank reactors, such as reduction in acid consumption, higher yield and lower degradation of hemicellulosic sugars [18,19].
1.3.
Optimization of hydrolysis conditions
Optimization of hydrolysis conditions is another key factor for economical pentose extraction. Energy input (temperature, pressure, reactor configuration) and acid concentration should be minimized. Xu and Hanna studied the effect of temperature, time, and acid concentration on the sugar yield from DDGS [20]. Using statistical approach, they found the optimal conditions for xylose and arabinose extraction at 112 C during 84.5 min, and sulfuric acid 3.1%. Noureddini and Byun [21] observed high yields of monomeric sugars at a reaction temperature of 140 C when the biomass loading was low (5%) and the acid concentration was high (2.7%). They, however, noted that the hydrolysis conditions provided low selectivity towards pentoses. The past work also showed that, as expected, higher acid concentration would lead to, higher rate of sugar degradation [22,23], higher costs of acid neutralization and/or recovery, as well as higher equipment maintenance costs due to corrosion [24]. In this work, we demonstrate that by using an appropriate process design (reactor type, solids loading, solids-to-liquid ratio, hydrolysis conditions, etc.) it is feasible to significantly minimize the acid consumption while maintaining high pentose yields.
2.
Materials and methods
2.1.
Materials
Samples of DDG were obtained from Brown-Forman Distillery located in Louisville, KY. High purity standards of D(þ)Xylose, D(þ)Arabinose, D(þ)Glucose), furfural, 5hydroxymethylfurfural, acetic acid, and glycerol were purchased from Sigma Aldrich (St. Louis, MO).
2.2.
Analysis of DDG
DDG samples were analyzed according to standard procedures for crude protein (method AOAC 2001.11) [25], fat (method AOAC 2003.06) [26], ash (method AOAC 942.05) [27], moisture (method AOAC 935.29) [28], neutral detergent fiber (NDF) (method AOAC 2002.04) [29], and acid detergent fiber (ADF) (method AOAC 973.18) [30]. Fat, protein, and moisture
were analyzed by Brown-Forman. The ash, NDF, and ADF results were obtained from Midwest laboratory (Omaha, NE). All samples were analyzed in triplicate and the results are shown in Table 2. Total carbohydrates were determined by subtracting the total percentage of fat, proteins, and ash from 100. Neutral detergent fiber was used as a measure of total cellulosic material (cellulose þ hemicellulose þ lignin). Acid detergent fiber was used to estimate the amount of cellulose and lignin. Hemicellulose was determined as a difference between NDF and ADF.
2.3.
Pretreatment of DDG
In this study, screening of the starting material was performed using U.S. standard sieve No. 20 (0.85 mm opening). Typically 150 g of DDG was screened at a time until visually no material was passing through the sieve. The fraction of DDG that did not pass through the sieve e the coarse fraction - was used in hydrolysis experiments. On an average 62% of the DDG was retained as coarse fraction. The coarse fraction was shown to have higher fiber content. Ultrasound treatment was performed as follows: 50 g of screened DDG was added to 1 L of water and sonicated for 30 min at 200 W power using an ultrasonic homogenizer UP200S (Hielscher Ultrasonics, Germany) with a 40 mm diameter probe. After sonication, the fiber fraction was dewatered using a 40 mesh screen and used in acid hydrolysis.
2.4.
Dilute acid hydrolysis
Hydrolysis of DDG was performed in a 6 L percolation reactor with liquid recirculation (M/K systems Inc, Peabody MA). Schematic representation of the reactor is shown in Fig. 3. As shown in the Figure, acid solution is percolated through a bed of the coarse fiber placed in a basket in the digester. The basket has a perforated bottom; the percolated liquid drains through the bottom and is recirculated to the top of the basket through a heater using a pump. The heater is controlled through a programmable controller. A set point for temperature, time to ramp the temperature of the acid solution from ambient to the set point and time at the set point are inputs to the controller.
Table 2 e Composition of original and pretreated DDG (dry basis). Original Screened Screened and DDG DDG ultrasonicated DDG Crude protein (%) Crude fat (%) Ash (%) Neutral detergent fiber (%)a Acid detergent fiber (%)b Total carbohydrates (%) Hemicellulose (%) a b
25.72 10.48 1.76 53.51
21.35 11.49 1.40 57.38
23.28 10.78 0.93 58.12
27.76 62.04 25.75
26.57 65.76 30.81
28.39 65.01 29.73
Mass fraction of total cellulosic material. Mass fraction of total cellulose and lignin.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
4
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
microscope (FEI Company; Hillsboro, OR). The samples were dried in air, attached onto a carbon tape and coated with 3 nm gold layer.
Fig. 3 e Schematic description of the percolation reactor with liquid recirculation used for biomass hydrolysis.
The amount of water was kept at 3 L and the DDG amount was varied from 200 g (6.25%) to 750 g (20.1%). DDG loading higher than 20% caused water holding and interruption in water circulation due to excessive swelling of DDG in the reactor. For most of the experiments, the reactor was loaded with 300 g solids. The samples were hydrolyzed at temperatures of 110 C, 120 C or 140 C using sulfuric acid at concentration of 0.2% and 0.4%, and reaction time of 30 min and 60 min. The heating profiles were as follows: 40 min, 45 min, and 50 min ramp up time to reach 110 C, 120 C, and 140 C, respectively. After the reaction time was over, the heating was stopped and the reactor was cooled using a cold water jacket on the recirculation line until 50 C, after which the hydrolyzate was drained. Hydrolyzate samples were neutralized and kept frozen until analysis.
3.
Results and discussion
3.1.
Pretreatment of DDG
The composition of the original and pretreated DDG is given in Table 2. Results clearly indicate the improvement of the DDG composition upon screening: fractional increase in protein content of 17% and fractional increase in fiber content (NDF) of 7%. However, fat content slightly increased upon screening. Ultrasonication of the screened DDG did not result in a noticeable change in the composition, except for small decrease in fat and nearly 34% fractional decrease in ash content. Composition and nutrition analyses of the resulting fines fraction were performed in a separate study [31] and showed that it is enriched in proteins compared to original DDG (fractional increase of crude protein content of 11%; fractional increase of 36% of lysine content) and can be used in animal feed applications.
3.2.
Selective hydrolysis of hemicelluloses
3.2.1.
Biomass loading
The hydrolysis of DDG was performed in a percolation reactor with liquid recirculation (Fig. 3) as these reactor types are easily scalable for commercial production. However, percolation reactors suffer from high water consumption due to the swelling of biomass during hydrolysis and shrinkage of flow path of water through the bed. High water consumption would lead to dilution of sugars in the hydrolyzate; but this can be addressed through appropriate recirculation of liquors. As discussed below, this type of reactor allowed the use of low
2.5. Analysis of sugars and degradation products in the hydrolyzate HPLC analysis of the hydrolyzate samples was performed using a Waters 600E HPLC system (Waters Corporation, Milford, MA) with an Agilent 1260 Infinity refractive index detector, and an Agilent Hi-Plex H column (300 mm 7.7 mm, 8 mm). Column temperature was set to 60 C with a refractive index detector temperature of 55 C. The mobile phase consisted of sulfuric acid solution with concentration of 5 mol m3. The flow rate was set to 700 mm3 min1. Hydrolyzate samples were filtered through a 0.45 mL syringe filter and 20 mL of each standard solution and the samples were injected in duplicated. Sample runtime was monitored for 45 min. The hydrolyzate samples were analyzed for monomeric sugars (glucose, xylose, arabinose), as well as for furfural.
2.6. Characterization of DDG fibers by scanning electron microscopy (SEM) The morphology of the DDG fibers before and after hydrolysis was studied using NOVA NanoSEM 600 scanning electron
Fig. 4 e Pentoses yield and concentration in the hydrolyzate as a function of biomass loading. DDG pretreatment: screening. Hydrolysis conditions: 140 C, 0.4% of H2SO4, 60 min.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
5
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Table 3 e Yield of pentoses and selectivity as a function of pretreatment methods and hydrolysis conditions.a Pretreatment None
Temperature ( C)
Sulfuric acid (%)
Pentoses (g kg1 of DDG)
Pentosesb (%)
120
0.2 0.4 0.2 0.4 0.2 0.4 0.2 0.4 0.2 0.4 0.2 0.4
84.7 157.4 120.5 197.8 74.7 166.4 153.1 243.8 97.7 191.7 162.1 280.9
95.3 96.1 95.5 91.4 94.5 96.5 95.1 91.2 96.6 97.1 95.9 91.7
140 Screening
120 140
Screening plus ultrasonication
120 140
a b
Time of hydrolysis: 60 min. Mass fraction of pentose in total sugars (selectivity).
concentrations (0.2 %e0.4 %) of sulfuric acid and achieved high hemicellulose conversion to sugars. In order to ensure the proper circulation, an adequate amount of liquid based on DDG loading should be maintained. As indicated earlier, swelling of biomass during hydrolysis leads to shrinkage of flow path of water through the bed of DDG. A sufficient flow of liquid must also be maintained to ensure efficient mass transfer inside the reactor. In this regard, the first parameter addressed was the DDG loading (DDG to water ratio). Fig. 4 shows total pentose yield obtained upon hydrolysis at different DDG loading (hydrolysis conditions: 140 C; 0.4% of H2SO4; 60 min). It can be seen that the pentose yield decreases with the increase of biomass loading throughout the entire range, however, the most pronounced decrease is observed
above 10% biomass loading. A similar trend was observed by Noureddini and Byun [21] for a stirred tank reactor, where the highest sugar yields were observed at the lowest DDG loading (5%). Although the pentose yields were a little higher at the lower biomass loading, the sugar concentration in the hydrolyzate was found to decrease significantly. For this reason, the rest of the experiments were fixed to biomass loading at 9.1% (1:10 DDG to water ratio).
3.2.2.
Optimization of hydrolysis conditions
Table 3 summarizes the effect of physical pretreatment, temperature, and acid concentration on pentose yield and percent of pentoses in total sugars (selectivity). The duration of hydrolysis was 60 min. In all the cases, the yield of the pentoses increased with the increase of temperature and acid
Fig. 5 e SEM images of DDG fibers (a) before and (bee) after hydrolysis at different conditions: (b) 120 C, 0.2% of H2SO4, (c) 140 C, 0.2% of H2SO4 (d) 120 C, 0.4% of H2SO4 and (e) 140 C, 0.4% of H2SO4. The scale bars in the figures correspond to 5 mm, while the scale bars in the insets correspond to 1 mm. Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
6
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
concentration. Based on the pentose yields, the hydrolysis conditions can be arranged in order of the increase in their yield as follows: (120 C and 0.2% of H2SO4) < (140 C and 0.2% of H2SO4) < (120 C and 0.4% of H2SO4) < (140 C and 0.4% of H2SO4). Thus, the increase in H2SO4 concentration from 0.2 % to 0.4 % showed a stronger effect on sugars yield than an increase in temperature from 120 C to 140 C. SEM micrographs (Fig. 5) visually demonstrate changes in the fiber surface morphology as the severity of the hydrolysis conditions (reaction temperature from 120 C to 140 C and sulfuric acid concentration from 0.2% to 0.4%) increases. A progressive exposure of the cellulose micro-fibrils can be observed, as the degree of the hemicellulose solubilization increases. It can be noticed that the hemicellulose extraction increases with the increase of temperature (Fig. 5 b and c), as well as with the increase of acid concentration (Fig. 5, d and e). The observed trend is in a good agreement with the pentose yields under the same conditions (Table 3). There is a noticeable effect of physical pretreatment on the pentose yield. Two types of DDG pretreatment were done: 1step pretreatment (screening only) and 2-step pretreatment (screening followed by ultrasonication). Both types of pretreatment improved the yield of pentoses with more pronounced results for hydrolysis conditions with higher temperature and acid concentration. The pentose yield at 140 C with 0.4% of H2SO4 increased from 197.8 to 243.8 g kg1 as a result of screening (Table 3). This increase can be attributed to the 20% fractional increase in hemicellulose content upon screening (Table 2). Ultrasonication of screened DDG resulted in further increase in pentose yields for each hydrolysis condition. In particular, ultrasonication as a second pretreatment step further increased the pentose yield from 243.8 to 280.9 g kg1 when the hydrolysis conditions were 140 C with 0.4% of H2SO4. This corresponds to 94% yield of available hemicellulose. Thus, the 2-step pretreatment procedure (ultrasonication of screened material) resulted in 42% overall fractional increase in pentose yield under the above mentioned hydrolysis conditions. The increase in sugar yield is most likely due to dislodging of starches, proteins, and fats from a fiber surface due to vibration resulting from ultrasonication. A high mass fraction of the pentoses in total sugars (95 %e 97 %) was achieved for most of the hydrolysis conditions used in this study (Table 3). It decreased, however, to around 91% for hydrolysis conditions with the highest temperature (140 C) and acid mass fraction (0.4%). This pentose mass fraction is still significantly higher than the mass fraction (64.5%) reported by Noureddini and Byun [21] for DDG hydrolysis conditions that gave similar pentose yield. It should be pointed out that while there was no apparent difference in
selectivity between original and screened samples, ultrasonicated samples showed on average higher selectivity for most of the hydrolysis conditions. The highest pentose mass fraction of 97.1% in total monomeric sugars was obtained for ultrasonicated sample hydrolyzed at 120 C with 0.4% of H2SO4. This shows that ultrasonication improved the efficiency of the dilute acid hydrolysis, which is consistent with earlier reports [17]. The improvement is most likely due to the surface erosion and fibrillation of lignocellulose matrix caused by ultrasound, thereby making it more accessible to acid [32]. Overall, compared to other studies, our process design resulted in similar or higher pentose yields and the highest hydrolysis selectivity while using significantly lower acid concentration (Table 4). In some processes, individual pentoses (xylose and arabinose) are required. Therefore it is important to determine the effect of DDG pretreatment and hydrolysis conditions on the yield of each of the pentoses. These data are summarized in Fig. 6. As expected, the arabinose requires much milder conditions than xylose: most of it was already formed at the lowest temperature and acid concentration (120 C, 0.2% of H2SO4). Under the most severe conditions (140 C, 0.4% of H2SO4) the amount of arabinose increased only slightly. These observations are consistent with literature data [21]. This is attributed to the fact that arabinose has higher structural angle strain in the furanosic sugar units [33]. On the other hand, the yield of xylose rises steadily with the increase of the severity of the conditions. Increase in temperature from 120 C to 140 C at 0.2% of H2SO4 led to 2.7e4.2 times increase in xylose yield, depending on pretreatment method. The same increase in temperature at 0.4% of H2SO4 has smaller effect on xylose yield (1.7e1.9 fold increase). Increase in acid mass fraction from 0.2 % to 0.4 % in the similar manner had stronger effect on xylose yield at lower temperature (3.8e5.1 times increase) compared to higher temperature (2.3e2.4 times increase). No apparent effect of mechanical pretreatment on arabinose yield was observed. This is likely due to the fact that under the hydrolysis conditions used in this study the higher limit of arabinose extraction has been reached. On the other hand, the yield of xylose consistently increases with each pretreatment step. Under all reaction conditions, there is a consistent 55 %e59 % fractional increase in xylose yield after the 2-step pretreatment. Depending on particular purpose and technology at hand, there may be a need to separate the DDG hydrolyzate into arabinose-rich and xylose-rich streams. For example, coconversion of both arabinose and xylose to bioethanol is a challenging task [34]. In such cases, separation of the pentose mixtures makes it easier to achieve the desired products such
Table 4 e Comparison of hydrolysis conditions, pentose yields and selectivity from different studies. Reactor type
Stir-tank Stir-tank Stir-tank Percolation
Hydrolysis conditions
H2SO4 (%)
Temp ( C)
Time (min)
3.1 2.7 3.27 0.4
112 140 140 140
84.5 60 20 30
C5 sugars (g kg1 of dry DDG)
Glucose in total sugars (%)
Reference
349.5 173.5 220.2 237
11.5 27.4 37.1 4.1
Xu and Hanna [20] Noureddini and Byun [21] Tucker et al. [36] This work
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
7
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 7 e Concentration of monomeric sugars from screened DDG hydrolyzed with 0.2% of H2SO4 under various conditions: a) fixed reaction time (60 min) and variable temperature (110C and 120 C), b) fixed temperature (120 C) and variable reaction time (30 min and 60 min).
collected, the residual fiber was washed with water and submitted to a second hydrolysis process at 140 C and sulfuric acid mass fraction of 0.4%. Table 5 shows that in the first hydrolyzate the weight fraction of arabinose in total pentoses is 81.5%, while in the second hydrolyzate xylose predominates (85.2% in total pentoses). While the proposed 2-stage hydrolysis obviously requires higher materials, time, and energy consumption to obtain the same total sugar amount, its usefulness will be determined by the particular end utilization of the individual pentoses, as well as method and cost of downstream processing.
3.2.3. Fig. 6 e Concentration of monomeric sugars obtained during DDG hydrolysis as a function of pretreatment method and temperature, using (a) 0.2% of H2SO4 and (b) 0.4% of H2SO4. Hydrolysis duration was 60 min for all experiments.
as ethanol or xylitol [22,35]. Under the mild conditions used (120 C, 0.2% of H2SO4), arabinose constitutes majority of the total pentoses (78.6% for screened DDG sample) (Fig. 6a). On the other hand, under the severe conditions (140 C, 0.4% of H2SO4) xylose constitutes the majority of the pentoses (64.5% for screened DDG sample) (Fig. 6b). Further testing indicated that the fraction of arabinose in total pentoses can be further increased by using even milder conditions. During hydrolysis at 0.2% of H2SO4 for 60 min, the mass fraction of arabinose in total pentoses increased from 78.6 % to 80.8 % as the reaction temperature is lowered from 120 to 110C; however, compared to that at 120C, the arabinose yield decreased by 12% at 110 C (Fig. 7a). Decrease in reaction time from 60 min to 30 min (120 C, 0.2% of H2SO4) did not show any improvement (Fig. 7b). Based on these findings, a two-stage hydrolysis process was carried out to favor the extraction of arabinose during the first stage and xylose during the second stage, while reusing the biomass (Table 5). The first stage was conducted at 110 C and acid mass fraction of 0.2%. After the first hydrolyzate was
Residual fiber after hydrolysis
Composition and nutrition analyses of the residual fibers after hydrolysis were performed in a separate study [31] and showed that the residual fiber after acid hydrolysis has significantly higher level of crude fat (a fractional increase of 113%), higher total digestible nutrients (a fractional increase of 15%), and higher digestible energy (fractional increase of 15%) compared to the original DDG as feed. The analysis also showed that lysine content in the residual fiber decreased by 69% compared to the original DDG. Overall chemical analysis indicates that this residual fiber has beneficial feed value and would be a useful and sustainable outlet for a co-product from this process which would have to be verified with additional testing and animal feeding studies.
Table 5 e Selective extraction of individual pentoses using 2-stage hydrolysis.a Compound Arabinose Xylose Glucose Glycerol Acetic acid 5-HMF Furfural a b c
First stageb (kg m3)
Second stagec (kg m3)
5.41 1.23 0.26 1.16 0.53 0.01 0.02
2.52 14.46 2.22 0.88 1.84 0.16 1.07
DDG pretreatment: screening. 110 C and 0.2% mass fraction of H2SO4. 140 C and 0.4% mass fraction of H2SO4.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
8
3.2.4.
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Pentose degradation
Hydrolysis conditions that provided high yield of pentoses usually lead to their partial conversion to furfural and other degradation products, resulting in a loss of potentially recoverable sugars. The degradation products are inhibitors to further conversion of these sugars to useful products. For instance, furfural is known to have inhibitory effect during microbial conversion of xylose to xylitol or ethanol [23]. Previous studies have shown that the production of furfural is strongly affected by temperature and reaction time [21e23]. On the other hand, hemicellulose hydrolysis is strongly affected by acid concentration. Thus, it may be possible to adjust these three parameters so that the rate of furfural formation can be decreased while maintaining high yield of C5 sugars. Fig. 8a shows that temperature has a major influence on the total amount of furfural formed. At low temperature (120 C), increase in the sulfuric acid mass fraction from 0.2 % to 0.4 % has a minimal effect on furfural formation. When the temperature was raised to 140 C, the furfural formation
Table 6 e Effect of time, acid concentration, and temperature on furfural formation.a Temperature ( C) 110 120 120 140 140 a
Sulfuric acid (%) 0.2 0.2 0.2 0.4 0.4
Time Furfural (min) (kg m3) 60 60 30 60 30
0.02 0.22 0.02 1.61 0.79
Pentoses (kg m3) 6.4 7.5 7.5 23.7 24.4
DDG pretreatment: screening.
increased significantly with increase in acid concentration. Fig. 8a also shows that furfural concentration in the hydrolyzate increased with pretreatment of DDG. This is due to the additional extraction of pentoses in the hydrolyzate as a result of pretreatment. This becomes clear in Fig. 8b, when the furfural formation is represented as the percent degradation of the available pentoses. Taking this into account, further testing was conducted varying temperature and time to see whether the furfural formation can be further decreased without negatively impacting pentose yield (Table 6). Indeed, when the temperature was decreased from 120 C to 110 C (0.2% of H2SO4, 60 min), the amount of furfural decreased by an order of magnitude, while the fractional decrease of pentose yield was 14%. When the reaction time decreased from 60 min to 30 min (120 C, 0.2% of H2SO4), the same order of magnitude decrease in furfural was observed, while the pentoses yield was not affected. At 140 C and 0.4% of H2SO4 concentration, decreasing reaction time to 30 min resulted in the reduction of furfural by 50%, while the pentoses yield was still unaffected. Thus, the reaction time seems to be more suitable parameter to be used as a tool for controlling sugar degradation. The above results indicate that it is possible to minimize the degradation of sugars while maintaining the same yield.
4.
Conclusions
In this study we showed that a pretreatment of DDG prior to dilute acid hydrolysis and the use of a percolation type reactor, can improve the selectivity towards the pentoses. Screening of DDG in order to remove fines followed by ultrasonication to roughen up the surface resulted in a fractional increase of pentose yield of 42%. This pretreatment also reduced the relative degradation of pentoses during the hydrolysis. It was also shown that using a two-stage hydrolysis process at different conditions it allows for predominant extraction of individual pentoses resulting in arabinose-rich (81.5% of arabinose in total pentoses) and xylose-rich (85.2% of xylose in total pentoses) streams. Fig. 8 e (a) Amount of furfural formed during DDG hydrolysis and (b) amount-of-substance fraction of pentoses converted to furfural as a function of pretreatment method, temperature, and acid weight fraction. Hydrolysis duration: 60 min. DDG pretreatment: screening.
Acknowledgments The authors thank Brown-Forman Corporation and Conn Center for their support.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
references [18] [1] Cherubini F, Stromman AH. Production of biofuels and biochemicals from lignocellulosic biomass: estimation of maximum theoretical yields and efficiencies using matrix algebra. Energy Fuels 2010;24(4):2657e66. [2] Perlack RD, Wright LL, Turhollow AF, Graham RI, Stokes BJ, Erbach DC. Biomass as feedstock for a bioenergy and bioproduct industry: the technical feasibility of a billion-ton annual supply. Oak Ridge TN: Oak Ridge National Laboratory; 2005 May. p. 78 DOE/GO-102995-2135, ORNL/TM-2005/66. [3] Xu Y, Hanna MA, Isom L. “Green” chemicals from renewable agricultural biomass e a mini review. Open Agr J 2008;2(1):54e61. [4] Kazi FK, Fortman JA, Anex RP, Hsu DD, Aden A, Dutta A, et al. Techno-economic comparison of process technologies for biochemical ethanol production from corn stover. Fuel 2010 November;89:S20e8. [5] Zhang YHP. What is vital (and not vital) to advance economically-competitive biofuels production. Process Biochem 2011;46(11):2091e110. [6] Powers WJ, Harris Jr B, Wilcox CJ. Effects of variable sources of distillers dried grains plus solubles on milk yield and composition. J Dairy Sci 1995;78(2):388e96. [7] Kim Y, Mosier NS, Hendrickson R, Ezeji T, Blaschek H, Dien B, et al. Composition of corn dry-grind ethanol by-products: DDGS, wet cake, and thin stillage. Bioresour Technol 2008;99(12):5165e76. [8] Liu KS. Chemical composition of distillers grains, a review. J Agric Food Chem 2011;59(5):1508e26. [9] Chandel AK, Chandrasekhar G, Radhika K, Ravinder R, Ravindra P. Bioconversion of pentose sugars into ethanol: a review and future directions. Biotechnol Mol Biol Rev 2011;6(1):008e20. [10] Raganati F, Curth S, Gotz P, Olivieri G, Marzocchella A. Butanol production from lignocellulosic-based hexoses and pentoses by fermentation of Clostridium acetobutylicum. Chem Eng Trans 2012;27(2):91e6. [11] Werpy T, Petersen G. Top value added chemicals from biomass: results of screening for potential candidates from sugars and synthesis gas, vol. 1. Washington DC: U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy; 2004 August. p. 76. DOE/GO-102004-1992. [12] Rafiqul ISM, Sakinah AMM. Design of process parameters for the production of xylose from wood sawdust. Chem Eng Res Des 2012;90(9):1307e12. [13] Tortosa CIG, Breijo FJG, Yufera EP. An economic-process for preparation of xylose and derivatives by hydrolysis of corn cobs. Biol Wastes 1990;33(4):275e86. [14] Herrera A, Tellez-Luis SJ, Ramirez JA, Vazquez M. Production of xylose from sorghum straw using hydrochloric acid. J Cereal Sci 2003;37(3):267e74. [15] Srinivasan R, Moreau RA, Rausch KD, Belyea RL, Tumbleson ME, Singh V. Separation of fiber from distillers dried grains with solubles (DDGS) using sieving and elutriation. Cereal Chem 2005;82(5):528e33. [16] Nitayavardhana S, Rakshit SK, Grewell D, van Leeuwen J, Khanal SK. Ultrasound pretreatment of cassava chip slurry to enhance sugar release for subsequent ethanol production. Biotechnol Bioeng 2008;101(3):487e96. [17] Yunus R, Salleh SF, Abdullah N, Biak DRA. Effect of ultrasonic pre-treatment on low temperature acid hydrolysis of oil
[19]
[20]
[21] [22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33] [34]
[35]
[36]
9
palm empty fruit bunch. Bioresour Technol 2010;101(24):9792e6. Yang B, Wyman CE. Effect of xylan and lignin removal by batch and flowthrough pretreatment on the enzymatic digestibility of corn stover cellulose. Biotechnol Bioeng 2004;86(1):88e95. Lee YY, Iyer P, Torget RW. Dilute-acid hydrolysis of lignocellulosic biomass. Recent progress in bioconversion of lignocellulosics. Springer; 1999. p. 93e115. Xu YX, Hanna MA. Optimum conditions for dilute acid hydrolysis of hemicellulose in dried distillers grains with solubles. Ind Crops Prod 2010;32(3):511e7. Noureddini H, Byun J. Dilute-acid pretreatment of distillers' grains and corn fiber. Bioresour Technol 2010;101(3):1060e7. Mussatto SI, Roberto IC. Acid hydrolysis and fermentation of brewer's spent grain to produce xylitol. J Sci Food Agric 2005;85(14):2453e60. Roberto IC, Mussatto SI, Rodrigues R. Dilute-acid hydrolysis for optimization of xylose recovery from rice straw in a semipilot reactor. Ind Crops Prod 2003;17(3):171e6. Balat M, Balat H, Oz C. Progress in bioethanol processing. Prog Energy Combust Sci 2008;34(5):551e73. AOAC Official Method 2001.11. Protein (crude) in animal feed, forage (plant tissue), grain, and oilseeds. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2002. AOAC 2003.06. Crude fat, hexanes extraction, in feed, cereal grain, and forage. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2003. AOAC 942.05. Determination of ash in animal feed. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2012. AOAC 935.29. Moisture content. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2005. AOAC 2002.04. Treated neutral detergent fiber in feeds using refluxing in beakers or crucibles. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2002. AOAC 973.18. Fiber (Acid detergent) and lignin in animal feed. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 1977. Lupitskyy R, Staff C, Satyavolu J. Evaluation of feed application for co-products from dilute acid hydrolysis of dried distillers grains (DDG), unpublished results. Li W, Wang R, Liu SX. Nanocrystalline cellulose prepared from softwood kraft pulp via ultrasonic-assisted acid hydrolysis. Bioresources 2011;6(4):4271e81. Sharples A. The hydrolysis of cellulose and its relation to structure. Trans Faraday Soc 1957;53:1003e13. Karhumaa K, Wiedemann B, Hahn-Hagerdal B, Boles E, Gorwa-Grauslund MF. Co-utilization of L-arabinose and Dxylose by laboratory and industrial Saccharomyces cerevisiae strains. Microb Cell Fact 2006;5(18). Chen YF, Dong BY, Qin WJ, Xiao DG. Xylose and cellulose fractionation from corncob with three different strategies and separate fermentation of them to bioethanol. Bioresour Technol 2010;101(18):6994e9. Tucker MP, Nagle NJ, Jennings EW, Ibsen KN, Aden A, Nguyen QA, et al. Conversion of distiller's grain into fuel alcohol and a higher-value animal feed by dilute-acid pretreatment. Appl Biochem Biotechnol 2004 Spring;115:1139e59.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
Available online at www.sciencedirect.com
ScienceDirect http://www.elsevier.com/locate/biombioe
Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis Dania A. Fonseca a,1, Robert Lupitskyy a,1, David Timmons b, Mayank Gupta a, Jagannadh Satyavolu a,* a b
Conn Center for Renewable Energy Research, University of Louisville, Louisville, KY 40292, USA Brown-Forman Corporation, 850 Dixie Hwy, Louisville, KY 40210, USA
article info
abstract
Article history:
The abundant availability and high level of hemicellulose content make dried distillers
Received 3 March 2014
grains (DDG) an attractive feedstock for production of pentoses (C5) and conversion of C5 to
Received in revised form
bioproducts. One target of this work was to produce a C5 extract (hydrolyzate) with high
6 October 2014
yield and purity with a low concentration of C5 degradation products. A high selectivity
Accepted 9 October 2014
towards pentoses was achieved using dilute acid hydrolysis of DDG in a percolation reactor
Available online xxx
with liquid recirculation. Pretreatment of starting material using screening and ultrasonication resulted in fractional increase of the pentose yield by 42%. A 94% yield of
Keywords:
pentoses on the DDG (280.9 g kg1) was obtained. Selective extraction of individual pen-
Dried distiller's grains
toses has been achieved by using a 2-stage hydrolysis process, resulting in arabinose-rich
Selective hydrolysis
(arabinose 81.5%) and xylose-rich (xylose 85.2%) streams. A broader impact of this work is
Integrated bio-refinery
towards an Integrated Bio-Refinery based on DDG e for production of biofuels, biochemical
Hemicellulose
intermediates, and other bioproducts.
Acid hydrolysis
© 2014 Elsevier Ltd. All rights reserved.
Degradation products
1.
Introduction
In recent years, significant research effort has been dedicated to the development of economically viable methods for conversion of lignocellulosic biomass, one of the most abundant and largely unused renewable resource on our planet, into bioproducts such as fuels and chemicals [1e3]. The biomass can be residue from farming, byproducts from agricultural and industrial processing, or from crops grown
solely for the purpose of energy production. However, biomass inherently has some drawbacks for use as starting material for fuels, chemicals, and other biomaterials [4,5]. The drawbacks include: low bulk density, leading to high storage and transportation costs; lower shelf life compared to petroleum (the biomass is dried to improve shelf life, thus adding to its cost); and its processing to sugars leaves residue, which needs to be handled. All these add to the overall cost of biomass processing, as shown in Fig. 1. In order to reduce the impact of the cost of feedstock on the overall production cost,
* Corresponding author. Conn Center for Renewable Energy Research, University of Louisville, 216 Eastern Parkway, Ernst Hall, Louisville, KY 40292, USA. Tel.: þ1 502 852 3923. E-mail address: [email protected] (J. Satyavolu). 1 Equal contribution. http://dx.doi.org/10.1016/j.biombioe.2014.10.008 0961-9534/© 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
2
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Zhang [5] suggested some of the top priorities of biomaterials R&D to be: 1) cost-effective release of sugars from lignocellulose and 2) co-utilization of lignocellulose components for the production of value-added compounds that subsidize whole biorefineries. Our goal is to develop an integrated approach for making bioproducts from biomass in order to contain the impact of the high feedstock cost on the overall process economics (Fig. 2). In this approach, we chose dried distillers grains (DDG) as our low cost renewable raw material. With an increasing demand for ethanol-based fuels, DDG e which is the major byproduct of dry grind corn ethanol production e is a highly desirable biomass starting material. For an estimate, about 7% by weight of corn is seed hull (pericarp) fiber biomass. The amount of corn seed hulls available in the US is estimated to be about 20 million tons and growing e a low cost and large volume opportunity. The DDG are commercially available and are used in animal feed applications [6]. Their shipping, storage, and other logistics costs are more effectively manageable compared to other traditional biomass feedstocks, such as corn stover. In addition, corn fiber has a unique composition e higher hemicellulose compared to cellulose and lower lignin [7,8] which makes this feedstock particularly attractive for pentose extraction. As shown in Table 1, corn fiber has about 4 times the amount of hemicellulose compared to that of cellulose. In addition, the lignin is only 3 %e6 % (all the % values in this work will refer to mass fraction). Such a composition of corn fibers will allow us to develop a process with higher selectivity, fewer process steps, and more importantly with lower cost. The described approach can co-exist with existing grain processing operations with a minimal economic impact on food and feed markets. The appropriate utilization of DDG to biofuels, biochemicals, and carbon fibers using an integrated approach will not only make the existing processing facilities cost efficient, but will also create new business and employment opportunities. This will also promote new opportunities for local agriculture and agricultural products. As shown in Fig. 2, our approach consists of the following steps: 1) Corn fiber from DDG is used to produce a pentose-rich hydrolyzate, 2) The pentose-rich hydrolyzate is used to produce biofuels and biochemicals, and 3) The residual fiber after
Fig. 2 e An integrated bio-refinery based on dried distillers grains (DDG).
acid hydrolysis is used in animal feed application or high value applications such as to produce carbon fibers. Pentoses (e.g., xylose and arabinose) can be biologically converted to liquid fuels, such as ethanol and butanol [9,10]. Alternatively, pentoses can be a platform for synthesis of a variety of industrially important chemicals (e.g., cyclopentadiene, cyclopentane, etc.), which are currently derived from petroleum [11]. Pentoses were identified by the US Department of Energy in 2004 among the top candidates of valuable chemical precursors which could be produced from biomass [11]. Many studies have been performed to develop processes for extraction of pentoses from a wide range of feedstocks [12e14]. Dilute acid hydrolysis is considered to be the most efficient method for selective extraction of pentoses, because hemicellulose is more easily hydrolyzed than cellulose due to amorphous structure and lack of hydrogen bonding. The primary goal of this study and the first step in our DDG biorefinery approach is to develop a process for pentose extraction from DDG that will meet several requirements: (1) high yield of pentoses, (2) high selectivity towards pentoses (maximized percent of pentoses in total monomeric sugars), (3) minimized sugar degradation, and (4) minimized acid and energy consumption. A thorough analysis of the past work by several researchers suggests that this goal can be achieved through several process steps:
1.1. Increase of fiber content in starting material via mechanical pretreatment Increasing the fiber content of DDG is typically accomplished by reducing its proteins and oil content. In the past studies reported in the literature, screening or air aspiration were
Table 1 e The composition analysis of fiber from various types of biomass.a
Fig. 1 e Factors contributing to high processing cost for bioproducts from biomass.
Composition
Corn fiber from Kernels
Softwood
Hardwood
Hemicellulose % Cellulose % Lignin % Ash
39e40 11e13 3e6 <1
25e29 40e44 25e31 <1
25e35 43e47 16e24 <1
a
[7,8].
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
3
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
evaluated [15] to increase fiber content in DDG. The fraction removed by aspiration showed a slight increase in the neutral detergent fiber content while leaving a residual fraction enriched in proteins and oil. Ultrasound pretreatment was also used to increase accessibility of lignocellulose during enzymatic hydrolysis [16] and acid hydrolysis [17].
1.2.
Selection of appropriate reactor design
Selection of an appropriate reactor configuration can significantly improve efficiency of the hydrolysis process. For example, past studies using percolation reactors showed several advantages over stir-tank reactors, such as reduction in acid consumption, higher yield and lower degradation of hemicellulosic sugars [18,19].
1.3.
Optimization of hydrolysis conditions
Optimization of hydrolysis conditions is another key factor for economical pentose extraction. Energy input (temperature, pressure, reactor configuration) and acid concentration should be minimized. Xu and Hanna studied the effect of temperature, time, and acid concentration on the sugar yield from DDGS [20]. Using statistical approach, they found the optimal conditions for xylose and arabinose extraction at 112 C during 84.5 min, and sulfuric acid 3.1%. Noureddini and Byun [21] observed high yields of monomeric sugars at a reaction temperature of 140 C when the biomass loading was low (5%) and the acid concentration was high (2.7%). They, however, noted that the hydrolysis conditions provided low selectivity towards pentoses. The past work also showed that, as expected, higher acid concentration would lead to, higher rate of sugar degradation [22,23], higher costs of acid neutralization and/or recovery, as well as higher equipment maintenance costs due to corrosion [24]. In this work, we demonstrate that by using an appropriate process design (reactor type, solids loading, solids-to-liquid ratio, hydrolysis conditions, etc.) it is feasible to significantly minimize the acid consumption while maintaining high pentose yields.
2.
Materials and methods
2.1.
Materials
Samples of DDG were obtained from Brown-Forman Distillery located in Louisville, KY. High purity standards of D(þ)Xylose, D(þ)Arabinose, D(þ)Glucose), furfural, 5hydroxymethylfurfural, acetic acid, and glycerol were purchased from Sigma Aldrich (St. Louis, MO).
2.2.
Analysis of DDG
DDG samples were analyzed according to standard procedures for crude protein (method AOAC 2001.11) [25], fat (method AOAC 2003.06) [26], ash (method AOAC 942.05) [27], moisture (method AOAC 935.29) [28], neutral detergent fiber (NDF) (method AOAC 2002.04) [29], and acid detergent fiber (ADF) (method AOAC 973.18) [30]. Fat, protein, and moisture
were analyzed by Brown-Forman. The ash, NDF, and ADF results were obtained from Midwest laboratory (Omaha, NE). All samples were analyzed in triplicate and the results are shown in Table 2. Total carbohydrates were determined by subtracting the total percentage of fat, proteins, and ash from 100. Neutral detergent fiber was used as a measure of total cellulosic material (cellulose þ hemicellulose þ lignin). Acid detergent fiber was used to estimate the amount of cellulose and lignin. Hemicellulose was determined as a difference between NDF and ADF.
2.3.
Pretreatment of DDG
In this study, screening of the starting material was performed using U.S. standard sieve No. 20 (0.85 mm opening). Typically 150 g of DDG was screened at a time until visually no material was passing through the sieve. The fraction of DDG that did not pass through the sieve e the coarse fraction - was used in hydrolysis experiments. On an average 62% of the DDG was retained as coarse fraction. The coarse fraction was shown to have higher fiber content. Ultrasound treatment was performed as follows: 50 g of screened DDG was added to 1 L of water and sonicated for 30 min at 200 W power using an ultrasonic homogenizer UP200S (Hielscher Ultrasonics, Germany) with a 40 mm diameter probe. After sonication, the fiber fraction was dewatered using a 40 mesh screen and used in acid hydrolysis.
2.4.
Dilute acid hydrolysis
Hydrolysis of DDG was performed in a 6 L percolation reactor with liquid recirculation (M/K systems Inc, Peabody MA). Schematic representation of the reactor is shown in Fig. 3. As shown in the Figure, acid solution is percolated through a bed of the coarse fiber placed in a basket in the digester. The basket has a perforated bottom; the percolated liquid drains through the bottom and is recirculated to the top of the basket through a heater using a pump. The heater is controlled through a programmable controller. A set point for temperature, time to ramp the temperature of the acid solution from ambient to the set point and time at the set point are inputs to the controller.
Table 2 e Composition of original and pretreated DDG (dry basis). Original Screened Screened and DDG DDG ultrasonicated DDG Crude protein (%) Crude fat (%) Ash (%) Neutral detergent fiber (%)a Acid detergent fiber (%)b Total carbohydrates (%) Hemicellulose (%) a b
25.72 10.48 1.76 53.51
21.35 11.49 1.40 57.38
23.28 10.78 0.93 58.12
27.76 62.04 25.75
26.57 65.76 30.81
28.39 65.01 29.73
Mass fraction of total cellulosic material. Mass fraction of total cellulose and lignin.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
4
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
microscope (FEI Company; Hillsboro, OR). The samples were dried in air, attached onto a carbon tape and coated with 3 nm gold layer.
Fig. 3 e Schematic description of the percolation reactor with liquid recirculation used for biomass hydrolysis.
The amount of water was kept at 3 L and the DDG amount was varied from 200 g (6.25%) to 750 g (20.1%). DDG loading higher than 20% caused water holding and interruption in water circulation due to excessive swelling of DDG in the reactor. For most of the experiments, the reactor was loaded with 300 g solids. The samples were hydrolyzed at temperatures of 110 C, 120 C or 140 C using sulfuric acid at concentration of 0.2% and 0.4%, and reaction time of 30 min and 60 min. The heating profiles were as follows: 40 min, 45 min, and 50 min ramp up time to reach 110 C, 120 C, and 140 C, respectively. After the reaction time was over, the heating was stopped and the reactor was cooled using a cold water jacket on the recirculation line until 50 C, after which the hydrolyzate was drained. Hydrolyzate samples were neutralized and kept frozen until analysis.
3.
Results and discussion
3.1.
Pretreatment of DDG
The composition of the original and pretreated DDG is given in Table 2. Results clearly indicate the improvement of the DDG composition upon screening: fractional increase in protein content of 17% and fractional increase in fiber content (NDF) of 7%. However, fat content slightly increased upon screening. Ultrasonication of the screened DDG did not result in a noticeable change in the composition, except for small decrease in fat and nearly 34% fractional decrease in ash content. Composition and nutrition analyses of the resulting fines fraction were performed in a separate study [31] and showed that it is enriched in proteins compared to original DDG (fractional increase of crude protein content of 11%; fractional increase of 36% of lysine content) and can be used in animal feed applications.
3.2.
Selective hydrolysis of hemicelluloses
3.2.1.
Biomass loading
The hydrolysis of DDG was performed in a percolation reactor with liquid recirculation (Fig. 3) as these reactor types are easily scalable for commercial production. However, percolation reactors suffer from high water consumption due to the swelling of biomass during hydrolysis and shrinkage of flow path of water through the bed. High water consumption would lead to dilution of sugars in the hydrolyzate; but this can be addressed through appropriate recirculation of liquors. As discussed below, this type of reactor allowed the use of low
2.5. Analysis of sugars and degradation products in the hydrolyzate HPLC analysis of the hydrolyzate samples was performed using a Waters 600E HPLC system (Waters Corporation, Milford, MA) with an Agilent 1260 Infinity refractive index detector, and an Agilent Hi-Plex H column (300 mm 7.7 mm, 8 mm). Column temperature was set to 60 C with a refractive index detector temperature of 55 C. The mobile phase consisted of sulfuric acid solution with concentration of 5 mol m3. The flow rate was set to 700 mm3 min1. Hydrolyzate samples were filtered through a 0.45 mL syringe filter and 20 mL of each standard solution and the samples were injected in duplicated. Sample runtime was monitored for 45 min. The hydrolyzate samples were analyzed for monomeric sugars (glucose, xylose, arabinose), as well as for furfural.
2.6. Characterization of DDG fibers by scanning electron microscopy (SEM) The morphology of the DDG fibers before and after hydrolysis was studied using NOVA NanoSEM 600 scanning electron
Fig. 4 e Pentoses yield and concentration in the hydrolyzate as a function of biomass loading. DDG pretreatment: screening. Hydrolysis conditions: 140 C, 0.4% of H2SO4, 60 min.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
5
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Table 3 e Yield of pentoses and selectivity as a function of pretreatment methods and hydrolysis conditions.a Pretreatment None
Temperature ( C)
Sulfuric acid (%)
Pentoses (g kg1 of DDG)
Pentosesb (%)
120
0.2 0.4 0.2 0.4 0.2 0.4 0.2 0.4 0.2 0.4 0.2 0.4
84.7 157.4 120.5 197.8 74.7 166.4 153.1 243.8 97.7 191.7 162.1 280.9
95.3 96.1 95.5 91.4 94.5 96.5 95.1 91.2 96.6 97.1 95.9 91.7
140 Screening
120 140
Screening plus ultrasonication
120 140
a b
Time of hydrolysis: 60 min. Mass fraction of pentose in total sugars (selectivity).
concentrations (0.2 %e0.4 %) of sulfuric acid and achieved high hemicellulose conversion to sugars. In order to ensure the proper circulation, an adequate amount of liquid based on DDG loading should be maintained. As indicated earlier, swelling of biomass during hydrolysis leads to shrinkage of flow path of water through the bed of DDG. A sufficient flow of liquid must also be maintained to ensure efficient mass transfer inside the reactor. In this regard, the first parameter addressed was the DDG loading (DDG to water ratio). Fig. 4 shows total pentose yield obtained upon hydrolysis at different DDG loading (hydrolysis conditions: 140 C; 0.4% of H2SO4; 60 min). It can be seen that the pentose yield decreases with the increase of biomass loading throughout the entire range, however, the most pronounced decrease is observed
above 10% biomass loading. A similar trend was observed by Noureddini and Byun [21] for a stirred tank reactor, where the highest sugar yields were observed at the lowest DDG loading (5%). Although the pentose yields were a little higher at the lower biomass loading, the sugar concentration in the hydrolyzate was found to decrease significantly. For this reason, the rest of the experiments were fixed to biomass loading at 9.1% (1:10 DDG to water ratio).
3.2.2.
Optimization of hydrolysis conditions
Table 3 summarizes the effect of physical pretreatment, temperature, and acid concentration on pentose yield and percent of pentoses in total sugars (selectivity). The duration of hydrolysis was 60 min. In all the cases, the yield of the pentoses increased with the increase of temperature and acid
Fig. 5 e SEM images of DDG fibers (a) before and (bee) after hydrolysis at different conditions: (b) 120 C, 0.2% of H2SO4, (c) 140 C, 0.2% of H2SO4 (d) 120 C, 0.4% of H2SO4 and (e) 140 C, 0.4% of H2SO4. The scale bars in the figures correspond to 5 mm, while the scale bars in the insets correspond to 1 mm. Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
6
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
concentration. Based on the pentose yields, the hydrolysis conditions can be arranged in order of the increase in their yield as follows: (120 C and 0.2% of H2SO4) < (140 C and 0.2% of H2SO4) < (120 C and 0.4% of H2SO4) < (140 C and 0.4% of H2SO4). Thus, the increase in H2SO4 concentration from 0.2 % to 0.4 % showed a stronger effect on sugars yield than an increase in temperature from 120 C to 140 C. SEM micrographs (Fig. 5) visually demonstrate changes in the fiber surface morphology as the severity of the hydrolysis conditions (reaction temperature from 120 C to 140 C and sulfuric acid concentration from 0.2% to 0.4%) increases. A progressive exposure of the cellulose micro-fibrils can be observed, as the degree of the hemicellulose solubilization increases. It can be noticed that the hemicellulose extraction increases with the increase of temperature (Fig. 5 b and c), as well as with the increase of acid concentration (Fig. 5, d and e). The observed trend is in a good agreement with the pentose yields under the same conditions (Table 3). There is a noticeable effect of physical pretreatment on the pentose yield. Two types of DDG pretreatment were done: 1step pretreatment (screening only) and 2-step pretreatment (screening followed by ultrasonication). Both types of pretreatment improved the yield of pentoses with more pronounced results for hydrolysis conditions with higher temperature and acid concentration. The pentose yield at 140 C with 0.4% of H2SO4 increased from 197.8 to 243.8 g kg1 as a result of screening (Table 3). This increase can be attributed to the 20% fractional increase in hemicellulose content upon screening (Table 2). Ultrasonication of screened DDG resulted in further increase in pentose yields for each hydrolysis condition. In particular, ultrasonication as a second pretreatment step further increased the pentose yield from 243.8 to 280.9 g kg1 when the hydrolysis conditions were 140 C with 0.4% of H2SO4. This corresponds to 94% yield of available hemicellulose. Thus, the 2-step pretreatment procedure (ultrasonication of screened material) resulted in 42% overall fractional increase in pentose yield under the above mentioned hydrolysis conditions. The increase in sugar yield is most likely due to dislodging of starches, proteins, and fats from a fiber surface due to vibration resulting from ultrasonication. A high mass fraction of the pentoses in total sugars (95 %e 97 %) was achieved for most of the hydrolysis conditions used in this study (Table 3). It decreased, however, to around 91% for hydrolysis conditions with the highest temperature (140 C) and acid mass fraction (0.4%). This pentose mass fraction is still significantly higher than the mass fraction (64.5%) reported by Noureddini and Byun [21] for DDG hydrolysis conditions that gave similar pentose yield. It should be pointed out that while there was no apparent difference in
selectivity between original and screened samples, ultrasonicated samples showed on average higher selectivity for most of the hydrolysis conditions. The highest pentose mass fraction of 97.1% in total monomeric sugars was obtained for ultrasonicated sample hydrolyzed at 120 C with 0.4% of H2SO4. This shows that ultrasonication improved the efficiency of the dilute acid hydrolysis, which is consistent with earlier reports [17]. The improvement is most likely due to the surface erosion and fibrillation of lignocellulose matrix caused by ultrasound, thereby making it more accessible to acid [32]. Overall, compared to other studies, our process design resulted in similar or higher pentose yields and the highest hydrolysis selectivity while using significantly lower acid concentration (Table 4). In some processes, individual pentoses (xylose and arabinose) are required. Therefore it is important to determine the effect of DDG pretreatment and hydrolysis conditions on the yield of each of the pentoses. These data are summarized in Fig. 6. As expected, the arabinose requires much milder conditions than xylose: most of it was already formed at the lowest temperature and acid concentration (120 C, 0.2% of H2SO4). Under the most severe conditions (140 C, 0.4% of H2SO4) the amount of arabinose increased only slightly. These observations are consistent with literature data [21]. This is attributed to the fact that arabinose has higher structural angle strain in the furanosic sugar units [33]. On the other hand, the yield of xylose rises steadily with the increase of the severity of the conditions. Increase in temperature from 120 C to 140 C at 0.2% of H2SO4 led to 2.7e4.2 times increase in xylose yield, depending on pretreatment method. The same increase in temperature at 0.4% of H2SO4 has smaller effect on xylose yield (1.7e1.9 fold increase). Increase in acid mass fraction from 0.2 % to 0.4 % in the similar manner had stronger effect on xylose yield at lower temperature (3.8e5.1 times increase) compared to higher temperature (2.3e2.4 times increase). No apparent effect of mechanical pretreatment on arabinose yield was observed. This is likely due to the fact that under the hydrolysis conditions used in this study the higher limit of arabinose extraction has been reached. On the other hand, the yield of xylose consistently increases with each pretreatment step. Under all reaction conditions, there is a consistent 55 %e59 % fractional increase in xylose yield after the 2-step pretreatment. Depending on particular purpose and technology at hand, there may be a need to separate the DDG hydrolyzate into arabinose-rich and xylose-rich streams. For example, coconversion of both arabinose and xylose to bioethanol is a challenging task [34]. In such cases, separation of the pentose mixtures makes it easier to achieve the desired products such
Table 4 e Comparison of hydrolysis conditions, pentose yields and selectivity from different studies. Reactor type
Stir-tank Stir-tank Stir-tank Percolation
Hydrolysis conditions
H2SO4 (%)
Temp ( C)
Time (min)
3.1 2.7 3.27 0.4
112 140 140 140
84.5 60 20 30
C5 sugars (g kg1 of dry DDG)
Glucose in total sugars (%)
Reference
349.5 173.5 220.2 237
11.5 27.4 37.1 4.1
Xu and Hanna [20] Noureddini and Byun [21] Tucker et al. [36] This work
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
7
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Fig. 7 e Concentration of monomeric sugars from screened DDG hydrolyzed with 0.2% of H2SO4 under various conditions: a) fixed reaction time (60 min) and variable temperature (110C and 120 C), b) fixed temperature (120 C) and variable reaction time (30 min and 60 min).
collected, the residual fiber was washed with water and submitted to a second hydrolysis process at 140 C and sulfuric acid mass fraction of 0.4%. Table 5 shows that in the first hydrolyzate the weight fraction of arabinose in total pentoses is 81.5%, while in the second hydrolyzate xylose predominates (85.2% in total pentoses). While the proposed 2-stage hydrolysis obviously requires higher materials, time, and energy consumption to obtain the same total sugar amount, its usefulness will be determined by the particular end utilization of the individual pentoses, as well as method and cost of downstream processing.
3.2.3. Fig. 6 e Concentration of monomeric sugars obtained during DDG hydrolysis as a function of pretreatment method and temperature, using (a) 0.2% of H2SO4 and (b) 0.4% of H2SO4. Hydrolysis duration was 60 min for all experiments.
as ethanol or xylitol [22,35]. Under the mild conditions used (120 C, 0.2% of H2SO4), arabinose constitutes majority of the total pentoses (78.6% for screened DDG sample) (Fig. 6a). On the other hand, under the severe conditions (140 C, 0.4% of H2SO4) xylose constitutes the majority of the pentoses (64.5% for screened DDG sample) (Fig. 6b). Further testing indicated that the fraction of arabinose in total pentoses can be further increased by using even milder conditions. During hydrolysis at 0.2% of H2SO4 for 60 min, the mass fraction of arabinose in total pentoses increased from 78.6 % to 80.8 % as the reaction temperature is lowered from 120 to 110C; however, compared to that at 120C, the arabinose yield decreased by 12% at 110 C (Fig. 7a). Decrease in reaction time from 60 min to 30 min (120 C, 0.2% of H2SO4) did not show any improvement (Fig. 7b). Based on these findings, a two-stage hydrolysis process was carried out to favor the extraction of arabinose during the first stage and xylose during the second stage, while reusing the biomass (Table 5). The first stage was conducted at 110 C and acid mass fraction of 0.2%. After the first hydrolyzate was
Residual fiber after hydrolysis
Composition and nutrition analyses of the residual fibers after hydrolysis were performed in a separate study [31] and showed that the residual fiber after acid hydrolysis has significantly higher level of crude fat (a fractional increase of 113%), higher total digestible nutrients (a fractional increase of 15%), and higher digestible energy (fractional increase of 15%) compared to the original DDG as feed. The analysis also showed that lysine content in the residual fiber decreased by 69% compared to the original DDG. Overall chemical analysis indicates that this residual fiber has beneficial feed value and would be a useful and sustainable outlet for a co-product from this process which would have to be verified with additional testing and animal feeding studies.
Table 5 e Selective extraction of individual pentoses using 2-stage hydrolysis.a Compound Arabinose Xylose Glucose Glycerol Acetic acid 5-HMF Furfural a b c
First stageb (kg m3)
Second stagec (kg m3)
5.41 1.23 0.26 1.16 0.53 0.01 0.02
2.52 14.46 2.22 0.88 1.84 0.16 1.07
DDG pretreatment: screening. 110 C and 0.2% mass fraction of H2SO4. 140 C and 0.4% mass fraction of H2SO4.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
8
3.2.4.
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
Pentose degradation
Hydrolysis conditions that provided high yield of pentoses usually lead to their partial conversion to furfural and other degradation products, resulting in a loss of potentially recoverable sugars. The degradation products are inhibitors to further conversion of these sugars to useful products. For instance, furfural is known to have inhibitory effect during microbial conversion of xylose to xylitol or ethanol [23]. Previous studies have shown that the production of furfural is strongly affected by temperature and reaction time [21e23]. On the other hand, hemicellulose hydrolysis is strongly affected by acid concentration. Thus, it may be possible to adjust these three parameters so that the rate of furfural formation can be decreased while maintaining high yield of C5 sugars. Fig. 8a shows that temperature has a major influence on the total amount of furfural formed. At low temperature (120 C), increase in the sulfuric acid mass fraction from 0.2 % to 0.4 % has a minimal effect on furfural formation. When the temperature was raised to 140 C, the furfural formation
Table 6 e Effect of time, acid concentration, and temperature on furfural formation.a Temperature ( C) 110 120 120 140 140 a
Sulfuric acid (%) 0.2 0.2 0.2 0.4 0.4
Time Furfural (min) (kg m3) 60 60 30 60 30
0.02 0.22 0.02 1.61 0.79
Pentoses (kg m3) 6.4 7.5 7.5 23.7 24.4
DDG pretreatment: screening.
increased significantly with increase in acid concentration. Fig. 8a also shows that furfural concentration in the hydrolyzate increased with pretreatment of DDG. This is due to the additional extraction of pentoses in the hydrolyzate as a result of pretreatment. This becomes clear in Fig. 8b, when the furfural formation is represented as the percent degradation of the available pentoses. Taking this into account, further testing was conducted varying temperature and time to see whether the furfural formation can be further decreased without negatively impacting pentose yield (Table 6). Indeed, when the temperature was decreased from 120 C to 110 C (0.2% of H2SO4, 60 min), the amount of furfural decreased by an order of magnitude, while the fractional decrease of pentose yield was 14%. When the reaction time decreased from 60 min to 30 min (120 C, 0.2% of H2SO4), the same order of magnitude decrease in furfural was observed, while the pentoses yield was not affected. At 140 C and 0.4% of H2SO4 concentration, decreasing reaction time to 30 min resulted in the reduction of furfural by 50%, while the pentoses yield was still unaffected. Thus, the reaction time seems to be more suitable parameter to be used as a tool for controlling sugar degradation. The above results indicate that it is possible to minimize the degradation of sugars while maintaining the same yield.
4.
Conclusions
In this study we showed that a pretreatment of DDG prior to dilute acid hydrolysis and the use of a percolation type reactor, can improve the selectivity towards the pentoses. Screening of DDG in order to remove fines followed by ultrasonication to roughen up the surface resulted in a fractional increase of pentose yield of 42%. This pretreatment also reduced the relative degradation of pentoses during the hydrolysis. It was also shown that using a two-stage hydrolysis process at different conditions it allows for predominant extraction of individual pentoses resulting in arabinose-rich (81.5% of arabinose in total pentoses) and xylose-rich (85.2% of xylose in total pentoses) streams. Fig. 8 e (a) Amount of furfural formed during DDG hydrolysis and (b) amount-of-substance fraction of pentoses converted to furfural as a function of pretreatment method, temperature, and acid weight fraction. Hydrolysis duration: 60 min. DDG pretreatment: screening.
Acknowledgments The authors thank Brown-Forman Corporation and Conn Center for their support.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008
b i o m a s s a n d b i o e n e r g y x x x ( 2 0 1 4 ) 1 e9
references [18] [1] Cherubini F, Stromman AH. Production of biofuels and biochemicals from lignocellulosic biomass: estimation of maximum theoretical yields and efficiencies using matrix algebra. Energy Fuels 2010;24(4):2657e66. [2] Perlack RD, Wright LL, Turhollow AF, Graham RI, Stokes BJ, Erbach DC. Biomass as feedstock for a bioenergy and bioproduct industry: the technical feasibility of a billion-ton annual supply. Oak Ridge TN: Oak Ridge National Laboratory; 2005 May. p. 78 DOE/GO-102995-2135, ORNL/TM-2005/66. [3] Xu Y, Hanna MA, Isom L. “Green” chemicals from renewable agricultural biomass e a mini review. Open Agr J 2008;2(1):54e61. [4] Kazi FK, Fortman JA, Anex RP, Hsu DD, Aden A, Dutta A, et al. Techno-economic comparison of process technologies for biochemical ethanol production from corn stover. Fuel 2010 November;89:S20e8. [5] Zhang YHP. What is vital (and not vital) to advance economically-competitive biofuels production. Process Biochem 2011;46(11):2091e110. [6] Powers WJ, Harris Jr B, Wilcox CJ. Effects of variable sources of distillers dried grains plus solubles on milk yield and composition. J Dairy Sci 1995;78(2):388e96. [7] Kim Y, Mosier NS, Hendrickson R, Ezeji T, Blaschek H, Dien B, et al. Composition of corn dry-grind ethanol by-products: DDGS, wet cake, and thin stillage. Bioresour Technol 2008;99(12):5165e76. [8] Liu KS. Chemical composition of distillers grains, a review. J Agric Food Chem 2011;59(5):1508e26. [9] Chandel AK, Chandrasekhar G, Radhika K, Ravinder R, Ravindra P. Bioconversion of pentose sugars into ethanol: a review and future directions. Biotechnol Mol Biol Rev 2011;6(1):008e20. [10] Raganati F, Curth S, Gotz P, Olivieri G, Marzocchella A. Butanol production from lignocellulosic-based hexoses and pentoses by fermentation of Clostridium acetobutylicum. Chem Eng Trans 2012;27(2):91e6. [11] Werpy T, Petersen G. Top value added chemicals from biomass: results of screening for potential candidates from sugars and synthesis gas, vol. 1. Washington DC: U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy; 2004 August. p. 76. DOE/GO-102004-1992. [12] Rafiqul ISM, Sakinah AMM. Design of process parameters for the production of xylose from wood sawdust. Chem Eng Res Des 2012;90(9):1307e12. [13] Tortosa CIG, Breijo FJG, Yufera EP. An economic-process for preparation of xylose and derivatives by hydrolysis of corn cobs. Biol Wastes 1990;33(4):275e86. [14] Herrera A, Tellez-Luis SJ, Ramirez JA, Vazquez M. Production of xylose from sorghum straw using hydrochloric acid. J Cereal Sci 2003;37(3):267e74. [15] Srinivasan R, Moreau RA, Rausch KD, Belyea RL, Tumbleson ME, Singh V. Separation of fiber from distillers dried grains with solubles (DDGS) using sieving and elutriation. Cereal Chem 2005;82(5):528e33. [16] Nitayavardhana S, Rakshit SK, Grewell D, van Leeuwen J, Khanal SK. Ultrasound pretreatment of cassava chip slurry to enhance sugar release for subsequent ethanol production. Biotechnol Bioeng 2008;101(3):487e96. [17] Yunus R, Salleh SF, Abdullah N, Biak DRA. Effect of ultrasonic pre-treatment on low temperature acid hydrolysis of oil
[19]
[20]
[21] [22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33] [34]
[35]
[36]
9
palm empty fruit bunch. Bioresour Technol 2010;101(24):9792e6. Yang B, Wyman CE. Effect of xylan and lignin removal by batch and flowthrough pretreatment on the enzymatic digestibility of corn stover cellulose. Biotechnol Bioeng 2004;86(1):88e95. Lee YY, Iyer P, Torget RW. Dilute-acid hydrolysis of lignocellulosic biomass. Recent progress in bioconversion of lignocellulosics. Springer; 1999. p. 93e115. Xu YX, Hanna MA. Optimum conditions for dilute acid hydrolysis of hemicellulose in dried distillers grains with solubles. Ind Crops Prod 2010;32(3):511e7. Noureddini H, Byun J. Dilute-acid pretreatment of distillers' grains and corn fiber. Bioresour Technol 2010;101(3):1060e7. Mussatto SI, Roberto IC. Acid hydrolysis and fermentation of brewer's spent grain to produce xylitol. J Sci Food Agric 2005;85(14):2453e60. Roberto IC, Mussatto SI, Rodrigues R. Dilute-acid hydrolysis for optimization of xylose recovery from rice straw in a semipilot reactor. Ind Crops Prod 2003;17(3):171e6. Balat M, Balat H, Oz C. Progress in bioethanol processing. Prog Energy Combust Sci 2008;34(5):551e73. AOAC Official Method 2001.11. Protein (crude) in animal feed, forage (plant tissue), grain, and oilseeds. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2002. AOAC 2003.06. Crude fat, hexanes extraction, in feed, cereal grain, and forage. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2003. AOAC 942.05. Determination of ash in animal feed. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2012. AOAC 935.29. Moisture content. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2005. AOAC 2002.04. Treated neutral detergent fiber in feeds using refluxing in beakers or crucibles. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 2002. AOAC 973.18. Fiber (Acid detergent) and lignin in animal feed. AOAC International. Gaithersburg, MD (USA): Association of Analytical Communities; 1977. Lupitskyy R, Staff C, Satyavolu J. Evaluation of feed application for co-products from dilute acid hydrolysis of dried distillers grains (DDG), unpublished results. Li W, Wang R, Liu SX. Nanocrystalline cellulose prepared from softwood kraft pulp via ultrasonic-assisted acid hydrolysis. Bioresources 2011;6(4):4271e81. Sharples A. The hydrolysis of cellulose and its relation to structure. Trans Faraday Soc 1957;53:1003e13. Karhumaa K, Wiedemann B, Hahn-Hagerdal B, Boles E, Gorwa-Grauslund MF. Co-utilization of L-arabinose and Dxylose by laboratory and industrial Saccharomyces cerevisiae strains. Microb Cell Fact 2006;5(18). Chen YF, Dong BY, Qin WJ, Xiao DG. Xylose and cellulose fractionation from corncob with three different strategies and separate fermentation of them to bioethanol. Bioresour Technol 2010;101(18):6994e9. Tucker MP, Nagle NJ, Jennings EW, Ibsen KN, Aden A, Nguyen QA, et al. Conversion of distiller's grain into fuel alcohol and a higher-value animal feed by dilute-acid pretreatment. Appl Biochem Biotechnol 2004 Spring;115:1139e59.
Please cite this article in press as: Fonseca DA, et al., Towards integrated biorefinery from dried distillers grains: Selective extraction of pentoses using dilute acid hydrolysis, Biomass and Bioenergy (2014), http://dx.doi.org/10.1016/ j.biombioe.2014.10.008