densification of a fermented lignocellulosic biomass rich ...densification of a fermented...

Volume 58 2016 CANADIAN BIOSYSTEMS ENGINEERING 3.1

Densification of a fermented lignocellulosic biomass rich in cellulolytic enzymes

Majid Soleimani1, Lope G. Tabil1, Leigh Campbell2, Russell K. Hynes3 and Tim Dumonceaux3

1Department of Chemical and Biological Engineering, University of Saskatchewan, Saskatoon, SK S7N 5A9, Canada 2GNC Bioferm Inc., Bradwell, SK S0K 0P0, Canada 3Agriculture and Agri-Food Canada, Saskatoon, SK S7N 0X2, Canada Email: [email protected] http://dx.doi.org/10.7451/CBE.2016.58.3.1

Received: 2015 September 14, Accepted: 2016 March 29, Published: 2016 June 6.

INTRODUCTION Conversion of the carbohydrate polymers cellulose and hemicellulose to their constituent monosaccharides using cellulases and hemicellulases is an important process in feed, bioproducts, and biofuel industries (Romero et al. 1999; Kang et al. 2004). Beta-glucanase, a marker for the activity of cellulolytic and hemicellulolytic enzymes, consists of a category of digestive enzymes for industrial and animal nutrition applications that releases glucose from the beta-linked glucose polymers in the lignocellulosic biomass. Animal feed is a rich source of non-starch polysaccharides (NSP), which are dissolved in the intestine and lead to an increased viscosity of intestinal contents resulting in a reduced growth rate and a low feed/gain ratio. Application of enzymes in the feed industry has dramatically increased due to their positive role in feed digestion, viscosity reduction in the digestive system and growth rate improvement (Silverside and Bedford 1999). Enzyme-rich lignocellulosic biomass in powder or ground form has a low bulk density (40-200 kg/m3) making for costly handling and storage. Therefore, a densification process such as pelleting needs to be applied to increase the bulk density of the biomass and to convert it to a denser product (pellets with a bulk density in the range of 600-800 kg/m3) (George et al. 2014). Heat treatment of animal diets in processes such as pelletization may cause enzyme destruction. When animal diets are formulated without added enzyme or when enzymes are inactivated during manufacturing process, an increased solubility of NSP at the same time would happen resulting in a high intestinal viscosity (Silversides and Bedford 1999). An ideal feed enzyme should have certain characteristics, namely a high specific activity, high thermostability, long-term stability at ambient condition and a high enough resistance to the animal digestive system (Selle and Ravindran 2007). The influence of different unit operations during feed processing such as grinding, mixing, thermal treatments, conditioning and pelletization on the stability of exogenous enzymes has been reported in detail in a wide array of research studies (Amerahi et al. 2011). However, there is less information on the behavior of enzymes in processes applying steam conditioning that needs to be

3.2 LE GÉNIE DES BIOSYSTÈMES AU CANADA Soleimani et al.

considered as an important subject in this area. To overcome contamination issues by microorganisms such Salmonella and Campylobacter in pelletization, the feed industry is trying to move towards high-temperature processing practices (Doyle and Erickson 2006). Moderate-temperature pelletization can improve the equipment throughput rate (TPR), degree of gelatinization and pellet quality to some extent. High-temperature practices may adversely affect nutrient availability through undesired reactions such as protein denaturation and Maillard reaction and complexing (Ravindran and Amerah 2008; Thomas et al. 1998). On the other hand, pelletization using high-temperature steam can improve mechanical properties of the densified biomass in spite of the negative impact on the nutrients intake and feed conversion ratio by increasing the viscosity resulting from the increased fiber solubility and starch gelatinization (Spring et al. 1996; Mateos et al. 2002; McCracken 2002; Cutlip et al. 2008; Ravindran and Amerah 2008). For an enzyme-rich product, the parameters of densification need to be optimized to minimize the loss of enzyme activity and to reach the optimal pellet quality properties in terms of physical and mechanical properties. High moisture content in an enzyme-rich product especially at a high temperature facilitates heat transfer into the biomass and accelerates enzyme inactivation (Slominski et al. 2007). In comparison to pelletization, granulation can be done using several techniques such as dry granulation, wet granulation, thermal adhesion granulation, pneumatic dry granulation, melt-thermoplastic granulation, layering granulation, fluidized bed granulation, extrusion-spheronization granulation, freeze granulation, steam granulation and spray drying granulation (Saikh 2013). Among these, extrusion-spheronization, powder layering, and solution-suspension layering are the most common techniques of granulation. Despite a number of studies on densification of cellulosic biomass by pelletization, very little information is available in the literature regarding the granulation of lignocellulosic enzyme-rich biomass and comparison of pelletization and granulation. The present work provides detailed technical information on the pelletization and granulation processes of an enzyme-rich biomass. The important variables in each densification process have been investigated and the enzyme activity, energy consumption during densification, bulk and unit densities and particle size distribution of the product are presented and discussed. The two aforementioned densification processes are compared in terms of product quality.

MATERIALS AND METHODS Materials Enzyme-rich biomass produced by solid-state fermentation on a grain-based substrate (barley, Hordeum vulgare) using Aspergillus niger and Trichoderma reesei was provided by GNC Bioferm Inc. (Bradwell, SK). It was packaged in multi-layer polyethylene to prevent exchange of moisture with the environment. Bentonite was obtained

from Canadian Clay Products Inc. (Wilcox, SK) to be used as the binder in the formulations. Silver prills, a hydrogenated palm oil with a melting point of 55°C, was supplied by Trident Feeds (Peterborough, Cambridgeshire, U.K.) to be used as lubricant and energy enhancer in the formulations. Methods Densification of the enzyme-rich biomass consisting of 10% fat in all formulations, with or without 1.5% bentonite, was carried out in two methods, namely pelletization and granulation. A pilot scale pellet mill (California Pellet Mill Co., Crawfordsville, IN) as described by Tabil and Sokhansanj (1996) was used for pelletization, and a combination of extrusion using a bench-top extruder (Fuji-Paudal MG55, LCI Co., Charlotte, NC) and spheronization using a spheronizer (Fuji-Paudal QJ-230T, LCI Co., Charlotte, NC) was used for granulation. For granulation process, the mixture was prepared by combining ingredients (biomass, water, fat, and binder) in a Viking Professional food processor (Viking, Greenwood, MS). The mixture was extruded through a 1.2-mm die using the extruder and the extrudate was transferred into the spheronizer producing granules. Finally, granules were dried in a laminar air flow hood at 23°C in 4 days. The pelletizing experiments were carried out using a ring die rotating at 250 rpm with a pellet die-hole diameter of 6.35 mm (0.25 in) and length:diameter (die) ratio of 6.9. A factorial design was applied for pelletization with two factors, namely, storage temperature of the enzyme-rich biomass (-18°C, 23°C (ambient), and 45°C) and pellet mill heating condition (no heating, dry heating (using heating pads), low-flow rate (0.75 kg/h), medium-flow rate (1.1 kg/h), and high-flow rate (1.45 kg/h) steam conditioning). For the no-heat and dry-heat pelletization, the moisture content of the biomass containing 1.5% bentonite and 10% fat (silver prills) was adjusted to 14% (wb). The moisture-adjusted biomass was stored overnight in the freezer at -18°C or at an ambient temperature at 23°C or was incubated at 45°C prior to pelletization. For the steam conditioning experiments, the biomass containing bentonite and silver prills with the original MC (≈7.5%) was conditioned at the aforementioned levels of steam. For the granulation process, two separate factorial experimental designs were employed on the biomass with or without 1.5% bentonite and all containing 10% silver prills. The first part (3-factor factorial experiment) was on the application of an axial-flow extrusion involving three factors, including bentonite inclusion (with 1.5% bentonite or without bentonite), MC (40, 45 and 50%wb), and residence time (30, 60 and 90 s) in the spheronizer. The second part (2-factor factorial experiment) of granulation was on the application of a radial-flow extrusion involving two factors, including MC (40, 45 and 50%wb) and residence time (30 and 60 s) in the spheronizer. The hemicellulose and cellulose contents in the biomass were determined based on NDF and ADF


following AOAC standards 2002.04 (AOAC, 2005) and 973.18 (AOAC, 1997), respectively. Lignin in the biomass was measured by NREL standard (Sluiter et al. 2008). Enzyme activity was reported based on β-glucanase activity measurement using spectrophotometry (Genesys 10s-vis spectrophotometer, Thermo Scientific, Mississauga, ON) and the release of sugars during a set period of time from a standard substrate lichenan

(Icelandic moss, Sigma-Aldrich Canada Co., Oakville, ON). To determine enzyme activity, 30 mg of the sample was suspended in an 8 ml buffer solution (acetic acid and sodium acetate) providing pH 4, and after 3 h of extraction followed by centrifugation; 1 ml of 0.5% substrate was added. The mixture was incubated at 30°C for 10 min with 1 ml alkaline solution of dinitrosalicylic acid (DNS) consisting of 16 g NaOH, 300 g Na-K-tartrate, and 10 g 3,5-dinitrosalicyclic acid in 1 litre deionized water. After cooling the sample and adding 8 ml deionized water to it, the absorbance was determined at 540 nm using the spectrophotometer. Enzyme activity of the densified biomass was presented as a percentage of the original enzyme activity of the enzyme-rich biomass received from GNC Bioferm, Inc. Bulk density of the pelletized product was determined using a standard funnel and a half-litre container (Adapa et al. 2010). Durability of the samples was tested by tumbling 100 g of pellets using a tumbler at 50 rpm for 10 min (ASABE standard S269, ASABE 2011). Unit density of the pellets was determined by the ratio of mass to the volume calculated by measurement of pellet dimensions for volume. Particle size distribution was carried out using ASABE S319 (ASABE 2011). 100 g of granules, produced by granulation, were shaken through a set of sieves (#4, 6, 8, 12, 16, 20, 30, 40, 50, 70, 100, and 140) using a Ro-Tap sieve shaker. Statistical analyses of the experiments in this study were performed using SAS (Statistical Analysis System, Cary, NC) based on factorial design for analysis of variance (ANOVA), and Duncan’s multiple range test for comparison of the means at P=0.05.

Fig. 1. One-way effect of storage temperature on quality attributes. Quality attributes are enzyme activity, durability, and bulk and unit densities of fungal biomass; letters are the comparison results from Duncan’s multiple-range test (P=0.05).

Fig. 2. One-way effect of pellet mill heating factor on quality attributes. Quality attributes are enzyme activity, durability, and bulk and unit densities of the fermented biomass in pelletization; letters are the comparison results from Duncan’s multiple-range test (P=0.05).


RESULTS AND DISCUSSION The lignocellulosic biomass used in this study for densification was composed mainly of: 27.4% protein, 19.2% cellulose, 22.6% hemicellulose, 5.5% lignin, 2.2% fat, 7.9% ash, and 8.1% water, all in dry basis (db). Pelletization The results of pelletization indicated that the one-way effect of the pellet mill heating variable and the interaction of the storage temperature and pellet mill heating variable were significant (P=0.01) on the enzyme (β-glucanase) activity of the final product. However, the storage temperature effect was not significant (Table 1 and Fig. 1). In the no-heat or dry-heat (temperatures of up to 55ºC) pelletization experiments, enzyme activities of over 90% of the original biomass were obtained (Fig. 2). Whilst, in pelletization with steam conditioning, the flow rate of steam was an important factor (Fig. 2) on the enzyme activity. At the low-flow rate steam conditioning, loss of up to 36% of the enzyme activity was observed; while at the high-flow rate steam conditioning, almost all enzyme activity was lost. Original storage temperature (-18 ºC, 23 ºC, and 45 ºC) of the enzyme-rich biomass was of less

importance to the final enzyme activity of the pelletized biomass. This substantial loss of enzyme activity by steam conditioning agrees with the results reported in the literature (Slominski et al. 2007; George et al. 2014). Durability of the pelletized biomass was significantly (P=0.01) affected by both storage temperature and pellet mill heating condition (Table 1). As shown in Fig. 1, steam conditioned pelletized biomass showed slightly higher durability compared to the non-steam conditioned biomass. This higher durability of the steam conditioned biomass could be due to the higher denaturation of the proteins resulting from heat and moisture added during the process, especially at a higher steam flow rate. A high durability of pellets was also reported by George et al. (2014) using steam conditioning. Bulk density was affected by both storage temperature and heating condition in the pellet mill, and also by the interaction of these factors. However, the pellet mill heating condition and interaction of storage temperature and pellet mill heating condition affected unit density of the pellets. Storage temperature was not solely an important factor to be effective on unit density of the

Table 1. Analysis of variance of enzyme activity, durability, and bulk and unit densities of the fermented biomass in pelletization.

Source Enzyme activity Durability Bulk density Unit density df MS df MS df MS df MS

ST 2 6.39NS 2 41.82** 2 3044.00** 2 2744.65NS PM 4 17042.20** 4 105.39** 4 4141.22** 4 33807.86** ST×PM 8 659.74** 8 6.76NS 8 1919.77** 8 8847.64** Error 30 9.11 30 4.85 60 16.49 135 1576.72 Total 44 1714.71 44 16.02 74 527.03 149 2856.63

ST= storage temperature; PM= pellet mill heating condition; df= degree of freedom; MS= mean of the sum of the squares; NS= not significant; **significant at P=0.01; *Significant at P=0.05.

Table 2. Dependence of enzyme activity, bulk and pellet densities, durability, TPR, and TSE on the combination of the pelletization factors.

Combination Enzyme activity (%)

Bulk density (kg/m3)

Pellet density (kg/m3)

Durability (%)

TPR (kg/h)

TSE (kWh/t.wb)

TSE (kWh/t.db)

A-NH 94.0bc 715b 1239b 85.9de 10.0 50.0 58.1 F-NH 93.4bc 704c 1207bc 87.4cde 9.4 55.1 64.1 H-NH 98.1ab 659hi 1200cd 85.1de 12.9 37.6 43.7 A-DH 99.9a 688d 1195cd 84.5e 10.5 43.0 50.0 F-DH 92.2c 721a 1277a 91.7ab 9.6 52.2 60.7 H-DH 98.3ab 655ij 1199cd 86.1de 12.5 32.0 37.2 A-LS 95.5abc 653j 1182cde 89.0bcd 11.3 31.1 36.2 F-LS 63.6e 675e 1149ef 93.3a 9.0 35.2 40.9 H-LS 77.9d 670ef 1197cd 90.6abc 10.0 36.7 42.6 A-MS 13.9g 663gh 1152ef 90.7abc 9.0 44.4 52.9 F-MS 48.8f 653j 1169cde 93.6a 9.0 50.0 59.5 H-MS 0h 653j 1122f 92.6ab 8.2 40.7 48.5 A-HS 0h 653j 1152ef 94.0a 6.4 54.4 66.4 F-HS 0h 661h 1160def 94.4a 6.4 64.8 79.0 H-HS 4.5h 667fg 1167cde 94.8a 7.5 46.7 56.9

TPR= throughput rate; TSE= total specific energy consumption; A= ambient temperature; F= freezing temperature (-18°C); H= heating temperature (45°C); NH= no heat; DH= dry heat; LS= low steam flow rate (0.75 kg/h); MS= medium steam flow rate (1.1 kg/h); HS= high steam flow rate (1.45 kg/h).


pellets (Table 1). Overall, lower densities were attained with steam conditioning compared to the no-heat or dry-heat densifications as shown in Fig. 2. Throughput of the biomass in the pellet mill for all combinations of the factors was determined and lower throughput values were obtained with steam conditioning, especially at the higher steam flow rate compared with no-heat or dry-heat pelletization (Table 2). This lower throughput rate was due to steam condensation in the conditioning chamber and bridging phenomenon leading to the slow movement of the material. This caused a fairly higher level of power consumption when using a high-flow rate of steam as shown in Table 2. George et al. (2014) and Tumuluru et al. (2011) also reported an inverse relationship between material throughput and specific power consumption during densification. Crumbling test

of the pelletized biomass using a crumbler (Cuthbert Co. Ltd., Winnipeg, MB) resulted in a product with non-uniform particles and wide particle size distribution (0.1 ≤ d ≤ 4 mm), which is not desirable. Granulation Compared to pelletization that binder was required in the formulation for a successful process due to the high fat content (10%), granulation was possible with or without inclusion of the binder. In the granulation (extrusion-spheronization) process, factors of initial MC, bentonite inclusion (B), spheronization time, and extrusion direction were assessed as independent factors. Bentonite inclusion in the formulations lowered the enzyme activity to some extent, especially at higher levels of MC (Table 4). Lower levels of MC resulted in higher enzyme activities in the granulated product as indicated in Fig. 3. The impact of the residence time in the spheronizer on the enzyme activity was not statistically significant and no special trend with the collected data was observed (Table 3). The average bulk density obtained by granulation was much less than the average bulk density obtained by pelletization (Table 2 vs. Table 4). Bulk densities of over 650 kg/m3 and up to 550 kg/m3 were respectively obtained by pelletization and granulation. On the other hand, higher power consumption was determined with the granulation process compared to pelletization. The most important factor affecting the bulk density of the granulated biomass was initial MC such that the higher level of MC (50% wb) resulted in lower bulk densities in the granulated product due to the clustering of the granules during spheronization. Clustering of granules resulted in a higher porosity between the clusters compared to the individual granules. Comparing the combinations of NB-40MC-60s (no

Table 3. Analysis of variance of enzyme activity and bulk density of the fermented biomass in granulation.

Source Enzyme activity Bulk density df MS df MS

B 1 401.83** 1 324.93** MC 2 875.15** 2 33489.23** R 2 43.57NS 2 1351.63** B×MC 2 719.57** 2 864.00** B×R 2 96.49** 2 24.73* MC×RPM 4 51.59* 4 1065.55** B×MC×R 4 41.41NS 4 29.93** Error 36 18.29 36 6.76 Total 53 93.91 53 1469.28

B= bentonite; MC= moisture content; R= residence time (s); MS= mean of the sum of the squares; df= degree of freedom; NS= not significant; **significant at P=0.01; *Significant at P=0.05.

Table 4. Dependence of bulk density, enzyme activity, ESE, SSE, and TSE on the combination of the granulation factors with axial extrusion.

Combination Bulk density (kg/m3)

Enzyme activity (%)

ETPR (kg/h)

ESE (kWh/t.wb)

SSE (kWh/t.wb)

TSE (kWh/t.wb)

TSE (kWh/t.db)

B-40MC-30s 515f 98.8ab 6.7 42.9 6.6 49.5 82.5 B-40MC-60s 545b 93.0abc 6.7 42.9 13.2 56.1 93.5 B-40MC-90s 556a 93.4abc 6.7 42.9 19.8 62.7 104.5 B-45MC-30s 505h 93.3abc 6.7 39.6 6.6 46.2 102.7 B-45MC-60s 517ef 95.6ab 6.7 39.6 13.2 52.8 117.3 B-45MC-90s 515f 94.8ab 6.7 39.6 19.8 59.4 132.0 B-50MC-30s 467i 69.0g 6.7 38.0 6.6 44.6 89.1 B-50MC-60s 461j 71.0fg 6.7 38.0 13.2 51.2 102.3 B-50MC-90s 456k 77.1ef 6.7 38.0 19.8 57.8 115.5 NB-40MC-30s 506gh 100.0a 7.5 36.7 6.6 43.3 72.1 NB-40MC-60s 540c 93.7abc 7.5 36.7 13.2 49.9 83.1 NB-40MC-90s 549b 94.8ab 7.5 36.7 19.8 56.5 94.1 NB-45MC-30s 510g 94.1ab 7.5 35.2 6.6 41.8 92.9 NB-45MC-60s 521e 91.7abc 7.5 35.2 13.2 48.4 107.6 NB-45MC-90s 529d 83.5de 7.5 35.2 19.8 55.0 122.2 NB-50MC-30s 443l 91.0bc 7.5 33.7 6.6 40.3 80.7 NB-50MC-60s 444l 98.3ab 7.5 33.7 13.2 46.9 93.9 NB-50MC-90s 431m 85.9cd 7.5 33.7 19.8 53.5 107.1

ETPR= extrusion throughput rate; ESE= extrusion specific energy consumption; SSE= spheronization specific energy; TSE= total specific energy; B= bentonite; NB= no bentonite; MC= moisture content; s= seconds of residence time in the spheronizer.


granulation of high-MC biomass by radial flow extrusion. Radial flow extrusion resulted in lower enzyme activities in the granulated biomass compared with the axial flow extrusion. For example, enzyme activities of 100% and 86% were obtained for NB-40MC-30s granules obtained by axial- and radial-flow extrusion, respectively. Particle size analysis indicated that the size distribution of the granules could be affected by MC of the biomass and residence time in the spheronizer (Fig. 4). Higher MC in both formulations with and without binder resulted in an increased distribution of particle size. Transition points in the sigmoid graphs show that at lower MC (40%), the size

bentonite-40% MC-60 s residence in the spheronizer), NB-45MC-60s (no bentonite-45% MC-60 s residence in the spheronizer), and NB-50MC-60s (no bentonite-50% MC-60 s residence in the spheronizer), bulk density values of 540 kg/m3, 521 kg/m3, and 444 kg/m3 were obtained, respectively. This shows the negative impact of MC increment on the bulk density of granules (Table 4). Comparing the axial flow and radial flow extrusions indicated that radial flow is much higher in power consumption and significantly lower in throughput rate (ETPR) of the biomass (Tables 4 and 5). For example, for the combination of NB-40MC-30s (no bentonite, 40% MC, 30 s residence in the spheronizer), TSE values of 229 and 72 kWh/t.db were obtained for granulations using axial flow and radial flow extrusion, respectively. Particle size analysis indicated that the MC of the biomass affected the size distribution of the granules. Similar to the axial flow extrusion, clustering of the granules occurred in the

Table 5. Dependence of bulk density, enzyme activity, ESE, SSE, and TSE on the combination of the granulation factors with radial extrusion.

Combination Bulk density (kg/m3)

Enzyme activity (%)

ETPR (kg/h)

ESE (kWh/t.wb)

SSE (kWh/t.wb)

TSE (kWh/t.wb)

TSE (kWh/t.db)

NB-40MC-30s 534b 86.4a 3.8 85.1 6.6 91.7 229.2 NB-40MC-60s 559a 84.1a 3.8 85.1 13.2 98.3 245.7 NB-45MC-30s 513c 85.2a 3.8 76.3 6.6 82.9 184.0 NB-45MC-60s 511c 66.4c 3.8 76.3 13.2 89.5 198.8 NB-50MC-30s 472d 81.7b 3.8 70.4 6.6 77.0 154.0 NB-50MC-60s 473d 68.6c 3.8 70.4 13.2 83.6 167.2

ETPR= extrusion throughput rate; ESE= extrusion specific energy consumption; SSE= spheronization specific energy; TSE= total specific energy; NB= no bentonite; MC= moisture content; s= seconds of residence time in the spheronizer.

Fig. 3. One-way effect of moisture content (MC) on quality attributes. Quality attributes here are enzyme activity and bulk density of the granulated biomass; letters are the comparison results from Duncan’s multiple-range test (P=0.05).

Fig. 4. Particle size distribution of the granules from extrusion-spheronization process with binder (a) and without binder (b).


of the particles was mainly limited to 0.2 to 0.8 mm in diameter, while at higher MC (50%) the size of the particle varied from 0.4 to over 2 mm in diameter. These results indicated that controlling MC and residence time in the spheronizer are important factors to reach a uniform particle size distribution. Comparing the particle size distribution in granulation with that obtained by pellet crumbling (0.1 - 4 mm), granulation would result in more uniform particles with less size variation which is preferred for marketing. However, granulation results in higher energy consumption and less efficiency in terms of product densification (Tables 2 and 4). Therefore, if small particles were not the target of densification, pelletization would be the preferred method due to the higher product density and less energy consumption.

CONCLUSION In pelletization using no-heat or dry-heat (temperatures up to 55ºC), and low-flow (0.75 kg/h) steam conditioning, enzyme activities of over 90% of the original were obtained. However, at high-flow (1.45 kg/h) steam conditioning, a complete enzyme deactivation, lower material throughput, and higher power consumption were attained. In granulation, lower MC (40-45%), no binder inclusion, axial flow extrusion, and shorter residence in spheronizer is recommended to achieve less enzyme deactivation, less power consumption, higher material throughput, and a lower size distribution and more uniform granules. Overall, pelletization is more energy efficient than granulation but is not suitable for uniform small-sized particles.

ACKNOWLEDGMENT The authors would like to acknowledge the technical staff of the Department of Chemical and Biological Engineering, University of Saskatchewan, GNC Bioferm Inc., and also Agri-Food Canada for their support. The authors are also grateful to MITACS Canada, Natural Sciences and Engineering Research Council (NSERC) of Canada, and GNC Bioferm for the financial support and providing equipment and materials for this work.

REFERENCES Adapa, P.K., L.G. Tabil, G.J. Schoenau and A. Opoku.

2010. Pelleting characteristics of selected biomass with and without steam explosion pretreatment. International Journal of Agricultural and Biological Engineering 3(3): 62-79.

Amerah, A.M., C. Gilbert, P.H. Simmins and V. Ravindran. 2011. Influence of feed processing on the efficacy of exogenous enzymes in broiler diets. World Poultry Science Journal 67: 29-46.

http://dx.doi.org/10.1017/S0043933911000031 AOAC. 1997. AOAC Method 973.18-Fiber (acid

detergent) and lignin in animal feeds. In Official Method of Analysis of the Association of Official Analytical Chemists, ed. P. Cunniff, Gaithersburg, MD: AOAC International.

AOAC. 2005. AOAC Method 2002.04-Amylase-treated neutral detergent fiber in feeds. In Horwitz, W. (Ed.), Official Method of Analysis of the Association of Official Analytical Chemists. Gaithersburg, MD: AOAC International.

ASABE. 2012. ASABE S269 - Densified products for bulk handling - definitions and method. In ASABE Standards Agricultural and Biological Engineers Yearbook of Standards. St. Joseph, MI: American Society of Agricultural and Biological Engineers.

ASABE. 2011. ASABE S319 - Method of determining and expressing fineness of feed materials by sieving. In ASABE Standards Agricultural and Biological Engineers Yearbook of Standards. St. Joseph, MI: American Society of Agricultural and Biological Engineers.

Cutlip, S.E., J.M. Hott, N.P. Buchanan, A.L. Rack, J.D. Latshaw and J.S. Moritz. 2008. The effect of steam conditioning practices on pellet quality and growing broiler nutritional value. Journal of Applied Poultry Research 17: 249–261.

http://dx.doi.org/10.3382/japr.2007-00081 Doyle, M.P. and M.C. Erickson. 2006. Reducing the

carriage of foodborne pathogens in livestock and poultry. Poultry Science 85: 960–973.

http://dx.doi.org/10.1093/ps/85.6.960 George, E., S. Emami, L.G. Tabil, and L. Campbell. 2014.

Effect of initial moisture and temperature on the enzyme activity of pelleted high protein-fiber biomass. Canadian Biosystems Engineering 56: 17-24.

Kang, S.W., Y.S. Park, J.S. Lee, S.I. Hong and S.W. Kim. 2004. Production of cellulases and hemicellulases by Aspergillus niger KK2 from lignocellulosic biomass. Bioresource Technology 91 (2): 153–156.

http://dx.doi.org/10.1016/S0960-8524(03)00172-X Mateos, G.G., R. Lazaro and M.I. Gracia. 2002. The

feasibility of using nutritional modifications to replace drugs in poultry feeds. Journal of Applied Poultry Research 11: 437–452.

http://dx.doi.org/10.1093/japr/11.4.437 Mccracken, K.J. 2002. Effects of physical processing on

the nutritive value of poultry diets. In Poultry Feeds, Supply, Composition and Nutritive Value. ed. J.M. Mcnab and K.N. Boorman, New York, NY: CAB International, pp. 301-316.

http://dx.doi.org/10.1079/9780851994642.0301 Ravindran, V. and A.M. Amerah. 2008. Improving the

nutritive value of feedstuffs using new technologies. Proceedings of the 23rd Worlds Poultry Science Congress, Brisbane, Australia, pp. 108.

Romero, M.D., J. Aguado, L. Gonzales and M. Ladero. 1999. Cellulase production by Neurospora crassa on wheat straw. Enzyme and Microbial Technology 25 (3-5): 244–250.

http://dx.doi.org/10.1016/S0141-0229(99)00035-6


Saikh, M.A.A. 2013. A technical note on granulation technology: a way to optimise granules. International Journal of Pharmaceutical Sciences and Research 4 (1): 55-67.

Selle, P.H. and V. Ravindran. 2007. Microbial phytase in poultry nutrition. Animal Feed Science and Technology 135: 1-41.

http://dx.doi.org/10.1016/j.anifeedsci.2006.06.010 Silversides, F.G. and M.R. Bedford. 1999. Effect of

pelleting temperature on the recovery and efficacy of a xylanase enzyme in wheat-based diets. Poultry Science 78: 1184-1190. http://dx.doi.org/10.1093/ps/78.8.1184

Slominski, B.A., T. Davie, M.C. Nyachoti and O. Jones. 2007. Heat stability of endogenous and microbial phytase during feed pelleting. Livestock Science 109: 244–246. http://dx.doi.org/10.1016/j.livsci.2007.01.124

Sluiter, A., B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, D. Templeton and D. Crocker. 2008. Determination of structural carbohydrates and lignin in biomass laboratory analytical procedure. National Renewable Energy Laboratory, Golden, Colorado.

Spring, P., K.E. Newman, C. Wenk, R. Messikommer, and M. Vranje. 1996. Effect of pelleting temperature on the activity of different enzymes. Poultry Science 75: 357-361. http://dx.doi.org/10.3382/ps.0750357

Tabil, L.G and S. Sokhansanj. 1996. Process conditions affecting the physical quality of alfalfa pellets. Applied Engineering in Agriculture 12(3): 345-350.

http://dx.doi.org/10.13031/2013.25658 Thomas, M., T.V. Vliet and A.F.B. Van Der Poel. 1998.

Physical quality of pelleted animal feed 3. Contribution of feedstuff components. Animal Feed Science and Technology 70 (1-2): 59-78.

http://dx.doi.org/10.1016/S0377-8401(97)00072-2 Tumuluru, J.K., C.T. Wright, J.R. Hess and K.L. Kenney.

2011. A review of biomass densification systems to develop uniform feedstock commodities for bioenergy application. Biofuels, Bioproducts, and Biorefining 5 (6): 682- 707. http://dx.doi.org/10.1002/bbb.324

densification of a fermented lignocellulosic biomass rich ...densification of a fermented...

Documents