1. Enzyme Engineering 9. Applications(3): Energy & Environment 9.1 Biodiesel Production 9.2 Bioethanol Production 9.3 Enzymatic Denitrification

2. 9.1 Biodiesel Production

3. Biodiesel Mixture of mono-alkyl esters (fatty acid alkyl esters) Typically made by transesterification of vegetable oils or animal fats. Advantages –Renewable resource –Biodegradable and less toxic –Favourable combustion emission profile Low carbon monoxide, sulphur oxide, and nitrogen oxide –High flash point Less volatile and safer to transport

4. Conversion of triglyceride to fatty acid alkyl esters Bottlenecks for lipase-catalyzed biodiesel production  Enzyme inhibitions by methanol and glycerol  Decrease of enzyme activity  High cost of lipase Transesterification process

5. Transesterification by lipase Enzymatic production is possible using both extracellular and intracellular lipases. In both the cases the enzyme is immobilized and used which eliminates downstream operations like separation and recycling. Both the processes are reported to be highly efficient compared to using free enzymes. Comparison of steps involved in the immobilization of extracellular and intracellular enzymes: (a) extarcellular lipase and (b) intracellular lipase. S.V. Ranganathan et al., Bioresource Technology 99 (2008) 3975–3981

6. Biodiesel production S.V. Ranganathan et al., Bioresource Technology 99 (2008) 3975–3981

7. Continuous reactor AuthorOilEnzymeReactor typeRemarks Conversi on Year S. Halim et al. Waste cooking palm oil Novozyme 435 Packed bed reactor tert-butanol 80% for 120hr 2009 Royon et al.Cottonseed oil Novoayme 435 Fixed bed reactor tert-butanol95%2007 Watanabe et al.Waste edible oil Novozyme 435 Fixed bed reactor Three step reaction90.9%2006 Shimada et al. Vegetable, tuna and waste edible oil Novozyme 435 Packed bed reactor 3 packed bed reactor in series 90% for 100 days 2002 Watanabe et al.Vegetable oil Novozyme 435 Packed bed reactor 2 and 3 packed bed reactors in series 93% for 100 days 2000

8. Extracellular lipase Such enzyme-catalyzed processes in entirely nonaqueous solvents were described by Klibanov et al. (1984). Candida antarctica lipase –Novozyme 435 (CALB immobilized on acrylic resin) –The most effective lipase among any of the lipases tested for methanolysis. Y. Shimada et al., JAOCS, Vol. 77, no. 7 (1999)

9. Extracellular lipase  Methanol has a strong inhibitory effect on the lipase. FIG.1. Effect of methanol content on methanolysis of vegetable oil. A mixture of 10 g oil/methanol and 0.4 g immobilized lipase was shaken at 30°C for 24 h. FIG. 2. Time courses of two- and three-step methanolyses of vegetable oil. Two-step reaction ( ○ ): after 7-h reaction, 2.10 g methanol was added in the reaction mixture, Three-step reaction ( ● ): 1.05 g methanol was added after 7- and 17-h reactions Y. Watanabe et al., JAOCS, Vol. 77, no. 4 (2000) Y. Shimada et al., JAOCS, Vol. 77, no. 7 (1999)

10. Extracellular lipase  Candida antarctica B-lipase (CALB); Novozym 435  Thermomyces lanuginosa (TLL); Lipozyme TL IM P. M. Nielsen et al., Eur. J. Lipid Sci. Technol. 2008, 110, 692–700

11. Intracellular lipase Bacteria, yeast and filamentous fungi that would serve as whole-cell biocatalysts based on their ability of immobilization and the display of functional proteins of interest on their cell surface. FIG. Micrographs of surface (a) and cross-section (b) of a BSP Developed by Matsumoto et al. (2001) –R. oryzae cell –Stepwise methanolysis of plant oil –In solvent free and water containing system –71% after a 165 hr at 37 ℃ K. Ban et al. / Biochemical Engineering Journal 8 (2001) 39–43

12. Intracellular lipase I. Addition of substrate-related compounds to the culture medium improved the lipase activity. II. Gluteraldehyde cross-linking of membrane bound lipase High lipase activity could be maintained over several cycles. III. Fatty acid membrane composition The membrane permeability and stability is significantly affected by the fatty acid composition. IV. Yatalase treatment; An enzyme that partially degrades fungal cell wall and allows the exposure of membrane-bound lipase to the substrate, thereby enhancing its activity. H. Fukuda et al., Trends in Biotechnology Vol.26 No.12, 2008

13. Intracellular lipase Secretory mechanism and localization In the immobilized culture, the fungal mycelium is immobilized into BSPs and large amounts of lipase are retained within the cells. Cleavage of pro-region from the precursor lipase Pre-pro ROL gives rise to two lipases. a.Cell membrane; ROL 31 b.Cell wall; ROL 34 Cell immobilization strongly inhibited the secretion of ROL 31 into the culture medium. H. Fukuda et al., Trends in Biotechnology Vol.26 No.12, 2008

14. Intracellular lipase A yeast cell surface display system for lipase from R. oryzae was developed. Schematic diagram of a yeast whole-cell biocatalyst displaying ROL via an FOL 1 anchor. Lipase displaying system –the N-terminus of ROL including a pro-sequence Pro-ROL has been fused to the flocculation functional domain of FLO p, a lectin-like cell-wall protein of yeast. H. Fukuda et al., Trends in Biotechnology Vol.26 No.12, 2008

15. Extracellular lipase vs. Intracellular lipase An outstanding reaction rate was obtained with Novozym 435 in continuous operation (7 h with 92–94% ME content) and fed-batch operations (3.5 h with 87% ME content) respectively. Considerable reduction in cost can be achieve with intracellular lipase. The reaction rate of intracellular lipase is very slow in a batch reaction and more than 70 h are required to obtain 80–90% ME content.

16. CalB Expression in E.coli  Materials  Cell: E.coli  1) DH5α  2) Rosettagami  3) Novablue  Vector: pCold Ⅰ  Cold shock promoter  N-terminal 6-His tag  Selection : Ampicillin  Chaperone plasmid map

17. CalB Expression in E.coli  Results  (1) Confirmation of lipase activity  (2) Halo test  (3) SDS-PAGE  Co-expression of Cal B and Chaperone DH5α RosettagamiNovablue DH5αRosettagamiNovablue

18. CalB Expression in E.coli  E. coli 를 이용하여 Candida antarctica 에서 유래된 lipase B (Cal B) 의 발현여부를 빠르게 확인하고, 특성을 분석하여 효율적인 발현시스템을 구축하며, 대량발현 시스템에 적용이 가능한지를 확인하였다.  발현 효율이 높고 활성이 좋은 재조합 lipase 개량에 있어서 빠른 확인을 할 수 있는 시스템을 확보하였다.  E. coli 에서 lipase 를 효율적으로 발현하기 위해서는 숙주세포 DH5α 에 Cal B 유전자가 재조합된 pCold Ι vector 와 pGro7(groES/groEL) 을 공동 형질전환 시킨 재조합 균주가 Cal B 발현시스템에 가장 효율이 좋았다.

19. CalB Expression in Pichia pastoris  Materials  Cell: Pichia pastoris X-33  Vector: pPICZα B  α-peptide: protein secretion  AOX1 promoter: induction by methanol  Selection: Zeocin Clone gene of interest into the pPICαA Linearize construct with Sac I Transform X33 using EasyComp TM method Plate transformant s on medium containing Zeocin All transformants integrate at 5’ AOX1 locus by single crossover Select 6-10 colonies for small- scale expression Choose highest expressers for large- scale expression Object : 생촉매 대량생산

20. CalB Expression in Pichia pastoris  Results  (1) Confirmation of lipase activity  (2) Halo test  Halo formation around colony by oil degradation due to lipase secretion Normal colony Halo formation by methanol induction  (3) SDS-PAGE  Confirmation of CalB expression and secretion Non-concentrated mediaConcentrated media

21. CalB Expression in Pichia pastoris  (4) Cal B Zymogram  Confirmation of CalB expression and secretion  (5) Activity test  Measurement of Cal B activity using p-NPP Time (days)OD400Unit/ml 30.05762.71 40.06672.44 50.07784.34 60.092100.55 70.104113.53 80.116126.50 90.129140.56 100.117127.58

22. Summary  The pellet sample's band was much clearer than that of the supernatant sample on the SDS-PAGE gel when the CalB expression was checked in the Pichia pastoris X-33 cell culture that was electroporated with the pPICZαB vector recombining a CalB gene.  From the thickness of the bands the expression level can be confirmed. As the culture time increased, CalB secreted to the media accumulated thereby thickening the bands.  When the activity of the enzyme was measured using p-NPP, the maximum value of 140,560U/L was reached on the day 9 of the culture. During the cell culture the cell growth reached 57.54~71.64g/L when BMGY→BMMY media was used.

23. Biocatalyst immobilization methods Immobilization methods Covalent binding on beads Covalent binding on mesoporous silica Sol-gel entrapment Advantages Easiness of substrate binding on biocatalyst Protection of biocatalyst from solvents Large surface area to maximize enzyme loading amount Protection of biocatalyst from solvents Maintenance of enzyme native structure Simple process Immobilization under mild condition Disadvantages Deactivation by solvent attack Diffusion limitation SNU : Sol-gel 방법을 사용

24. Sol-gel entrapment  Use of hydrophobic precursors  Known as CalB maintains higher activity in hydrophobic circumstance  Easiness of oil feeding to enzyme  Prevention of enzyme deactivation by hydrophilic methanol ETMS (Ethyltrimethoxysilane) MTMS (Methyltrimethoxysilane) iso-BTMS (iso-Buthyltrimethoxysilane) PTMS (Propyltrimethoxysilane) TMOS(Tetramethoxysilane) Enzyme 지금까지의 방법 : Hydrophobic precursor 사용

25. Selection of sol-gel matrix  Comparison of the activity in various matrix  TMOS (hydrophilic)  MTMS, ETMS, PTMS, iso-BTMS (hydrophobic) matrixTMOSMTMS*ETMS*PTMS*iso-BTMS* Activity (U/g support ) 0.1440.4820.5790.2670.460 Specific activity (U/g enzyme ) 9.7113.421.037.545.13 * Mixed with TMOS (TMOS: hydrophobic precursor= 1:4) ETMS/TMOS Lipase immobilized in ETMS mixed with TMOS showed the highest activity and specific activity. IDEA : Hydrophobic-hydrophilic 한 precursor 의 비율을 조정 및 최적화

26. Pretreatment of biocatalyst  Activity maintenance and acquisition of stability of CalB  Methods  Olive oil pretreatment for 1hr at 30 ℃  steric hindrance around active site during immobilization  GA pretreatment for 1hr at 25 ℃  Result : Development of new biocatalyst having the same activity compared with Novozym435 IDEA : Oil 전처리 방법 도입 MatrixUnpretreated GA pretreated Oil pretreated Oil and GA pretreated Novozym 435 Activity (U/g support) 0.5790.4980.8960.5950.916 Specific activity (U/g enzyme) 21.0317.6547.2333.81-

27. Activity of Immobilized lipase  Comparison of the activity  Lipase type  Candida antarctica lipase B  Rhizomucor miehei lipase  Thermomyces lanuginosus lipase  Particle size  100-300 μm and 300-500 μm Sol-gel 제조방법 개선 및 particle 사이즈 변경 & 다른 lipase 도 고정화

28. Stability of immobilized enzymes  Thermostability (60 ℃ )  Methanol tolerance (Oil : MeOH = 1:3) Acquisition of thermostability and methanol tolerance via immobilization of pretreated CALB

29. FAME production by immobilized lipases  FAME conversion using free lipase and immobilized lipase  Meterials  Free lipase : Lipozyme (CalB)  Sol-gel entrapped CalB using ETMS precursor  Results  Over 90% of conversion rate in oil pretreated biocatalyst  Methods  Pre-incubation of oil with methanol for 18 hrs, 300 rpm, at 50 ℃  immobilized lipase: 25%(w/w) of substrate  Methanol stepwise addition

30. CalB immobilization CalB immobilization onto nanocarbon materials High loading of enzyme Enhancement of CalB stability in organic solvent Support materials: functionalized nanocarbon material EDC Cal B

31. Result_ Immobilization of CalB 1ml of Lipozyme CALB L were immobilized with 10mg of nanoparticle. Single layer immobilization was formed around 6mg enzyme/mg support. Maximum ammount : 17.4 mg enzyme/ mg support Unusually higher amount  repeat experiment again.

32. Methanol inhibition Methanol has a strong inhibitory effect on the lipase. –conversion decreased when methanol and oil molar ratio is more then 1.5. To prevent the methanol inhibition, step addition was suggested. FIG 1. Effect of methanol content on methanolysis of vegetable oil. FIG. 2. Time courses of two- and three-step methanolyses of vegetable oil. Two-step reaction ( ○ ), Three-step reaction ( ● ) Y. Shimada et al., JAOCS, Vol. 77, no. 7 (1999), Y. Watanabe et al., JAOCS, Vol. 77, no. 4 (2000)

33.  A deposit of glycerol coating the immobilized catalyst is formed during the process, which reduces the enzymes activity. Glycerol removal by membrane was proposed. The two phase is separated by cellulose acetate membrane. Primary phase: oil-biodiesel-glycerol mixture Secondary phase: aqueous phase  Hydrophobic molecules(FAME and oil) can not pass this membrane.  Only hydrophilic ones, glycerol can pass this membrane.  The continuous inhibition of glycerol can be decreased. Glycerol inhibition K.Belafi-Bako et al., Biocatalysis and Biotransformation, 2002 Vol. 20 (6), pp. 437–439

34. The packed bed reactor has been investigated by several researchers for biodiesel production. –But step addition reactions were needed to prevent methanol inhibition. –And it needs down streams for glycerol separation. To overcome these disadvantages, flat sheet membrane reactor which was used for glyceride synthesis can be applicable for biodiesel production. Using hydrophilic membrane, the inhibition of continuously produced glycerol can be decreased and methanol can be fed continuously to improve the productivity during short operation time. Research objective

35. Continuous system Magnetic stirrer Product (FAME) MeOH & Glycerol Water Bath Immobilized enzyme Membrane OilMeOH

36. Experiment without enzyme Materials Hydrophobic phase : Soybean oil 0.6 ml/min (40 ml) Hydrophilic phase: Methanol 0.6 ml/min (20 ml) Reaction Condition Temperature: 50 ℃ Membrane Snake Skin Dialysis Tubing :Regenerated cellulose (10K MWCO) Analysis Varian 450-GC : HP-5 capillary column 25mx0.32mmx0.17 μ m MeOH Oil Oil&MeOH

37. Result_ Methanol permeability  Methanol permeation was tested without enzyme. Without stirring, 130 mM of methanol was maintained after 2hrs. With stirring; Methanol concentration was increased steadily. Maximum concentration of methanol was 400mM.

38. Experiment with Novozyme 435 Materials Hydrophobic phase : Soybean oil 0.6 ml/min (40 ml) Novozyme 4g Hydrophilic phase: Methanol 0.6 ml/min (20 ml) Reaction Condition Temperature: 50 ℃ Membrane Snake Skin Dialysis Tubing :Regenerated cellulose (10 K MWCO) Analysis Varian 450-GC : HP-5 capillary column 25mx0.32mmx0.17 μ m

39. Result _ Biodiesel production Biodiesel production was tested with continuous system. 25% conversion was achieved at 30 min. After 1 hr conversion rate was dramatically decreased and this low conversion rate (below 6%) was maintained during 10hr. This low conversion rate was based on enzyme deactivation by the methanol inhibition which is caused by high methanol concentration and mixing problem.

40. 9.2 Bioethanol Production

41. 9.3 Enzymatic Denitrification

42. Biological Denitrification  Economically and environmental friendly  Microorganisms use nitrate as a final electron acceptor  Occur under O 2 limitation condition  General process Oxic Anoxic Sediment Influent Return sludgeWaste sludge Effluent

43. Enzyme Reactions NO 3 - NO 2 - NO N 2 O N 2 NADH NAD + + H + NADH dehydrogenase NaRNiR NoRNo 2 R e-e- NaR: nitrate reductase NiR: nitrite reductase NoR: nitric oxide reductase No 2 R: nitrous oxide reductase

44. Screening of Denitrifying Bacteria Microorganisms Denitrification activity (nmol/min mg-cell) References Pseudomonas stutzeri90.2 Carlson and Igrshsm (1983) Pseudomonas aeruginosa45.2 Carlson and Igrshsm (1983) Paracoccus denitrificans34.2 Carlson and Igrshsm (1983) Flexibacter canadensis23.4 Wu and Knowles (1994) Pseudomonas stutzeri67.2 Nussinovitch et al. (1996) Shewanella putrefaciens41.0 Krause et al. (1997) Ochrobactrum anthropi SY509104This study

45. Packed-bed reactor ④ ⑥ ⑤ ③ ② ① ③ ④ ⑦ ⑧ ① Packed-bed reactor (100 ml) ② ORP/pH meter ③ ORP probe ④ pH probe ⑤ monitor ⑥ feeding water ⑦ pump ⑧ nitrogen gas purging - Flow rate : 5ml/min effluent influent

46. Bioelectrochemical Denitrification electrode e- M ox M red NO 3 - NO 2 - cell N2N2 …

47. Cell Permeabilization Control Permeabilized cell

48. Immobilization of Biocatalyst and Mediator on Electrode Surface Carbon nanopowder-Neutral red Polymer matrix (Ca-alginate) Permeabilized cell

49. Reactor System for Bioelectrochemical Denitrification Working electrode Reference electrode v Counter electrode Power supply S P2H 2 O O 2 +4H + +4e -

50. Denitrification Efficiency Figure. The residual nitrate level (a)for various cell concentrations and (b)for various concentrations of mediator (a) (b)  The cell concentration is rate-limiting compared to mediator concentration.

51. Conductive powder was mixed with the cells and alginate for electrode. Novel Biocatalytic Electrode permeabilized cell alginate conductive powder

52. Figure. SEM image (x3500) of novel electrode

53. Denitrification Efficiency

54. Continuous Reactor influent effluent anode: carbon felt cathode combined with biocatalyst

56.  1 st generation: mixed population oxic, anoix tanks  2 nd generation: single microorganism (enzymes) immobilized packed-bed reactor  3 rd generation: whole cell biocatalyst enzymatic-electrochemical reactor Biological Nitrate Removal Technology

57. Biodiesel Fuel Production by Transesterification of Oils Fukuda H, Kondo A, and Noda H, Kobe Univ., Japan Enzyme Engineering Minsuk Kim 2012.06.04. Review

58. Triglycerides!! Hold promise as alternative diesel engine fuel Vegetable oils / Oil blends / Oil waste ▫ Too viscous

59. Transform triglycerides to lower viscosity Pyrolysis – unacceptable in terms of ash, pour point etc. Micro-emulsification - incomplete combustion etc. Transesterification (alcoholysis) Supercritical fluid Chemical catalysis Enzyme Acid-catalysis Alkali-catalysis Extracellular Intracellular

60. Biodiesel fuel Product from alcoholysis of triglycerides. Attractive features ▫ Plant-derived, carbon neutral ▫ Lower dependence on petroleum ▫ Biodegradable ▫ Reduced levels of pollutants (CO, NO, SOx)

63. Transesterification (alcoholysis) Supercritical fluid Chemical catalysis Enzyme Acid-catalysis Alkali-catalysis Extracellular Intracellular Alkali-catalysis High conversionShort reaction timecheapness Water & free fatty acid

64. Transesterification (alcoholysis) Supercritical fluid Chemical catalysis Enzyme Acid-catalysis Alkali-catalysis Extracellular Intracellular Alkali-catalysis vs. Acid-catalysis Lower conversion and rate Low grade material such as sulphur oil

65. Transesterification (alcoholysis) Supercritical fluid Chemical catalysis Enzyme Acid-catalysis Alkali-catalysis Extracellular Intracellular Alkali-catalysis vs. Enzymatic process + waste water treatment

67. Methanol & ethanol are utilized frequently. ▫ Low cost and physical and chemical advantages. Lipases are known to have a propensity to act on long- chain fatty alcohols. ▫ Conversion : Methanol < Ethanol

68. Methanol can inactivate enzyme. ▫ Stepwise addition of methanol. ▫ Methanol should dissolve completely.

69. Transesterification (alcoholysis) Supercritical fluid Chemical catalysis Enzyme Acid-catalysis Alkali-catalysis Extracellular Intracellular Extracellular vs. Intracellular Purification of enzyme is too expensive Not easy to recover enzyme

70. Porous biomass support particles (BSPs) No chemical additives No need to preproduction of cells No aseptic handling of particle Large mass transfer rate Easy scale-up Low cost

71. Cell : Ryzopus oryzae Lipase induction by olive oil or oleic acid Stabilization

72. Transesterification (alcoholysis) Supercritical fluid Chemical catalysis Enzyme Acid-catalysis Alkali-catalysis Extracellular Intracellular Alkali-catalysis vs. Supercritical fluid Without any catalyst ▫ Supercritical methanol - 350℃ & 240s & 45MPa Supercritical methanol is hydrophobic ▫ Fast reaction rate Much simpler purification & environmental friendly

73. Conclusions and Future Prospects Alkali-catalysis is generally employed in actual biodiesel production. However, enzymatic transesterification is also attractive. Further improvement by genetic engineering. ▫ High level of expression of lipase. ▫ Stabilize toward methanol.

74. Comments Why fatty acid methyl ester? i.e. Why not fatty acid ethyl ester? ▫ Lipases are unstable in methanol ▫ Methanol is toxic to cells ▫ Many lipases has higher activity toward ethanol than methanol Vegetable oils  biodiesel fuel vs. Cellulosic biomass  biodiesel fuel ▫ By metabolic engineering. ▫ Which source has bigger potential?

75. The Challenge of Enzyme Cost in the Production of Lignocellulosic Biofuels 2012.06.11 Enzyme engineering (2012, Spring semester) Jeong Eun Shin

76. Introduction Understanding the contribution of enzymes to the cost of lignocellulosic biofuels. Ethanol Fungal cellulases Some authors…  The cost of enzyme is a major barrier for biofuel production Others…  The cost is relatively low or can be decreased with technological innovation or other advances 1)Techno-economic model 2) Sensitivity analysis

77. Techno-economic model The cost of enzymes [dollars per gallon of biofuel] : a “top-down” measure that depends on many factors besides the cost of enzymes themselves, including the choice of feedstock, the enzyme loading, and the overall biofuel yield The inconsistency in the cost of enzymes !!! The development of a techno-economic model for the production of cellulases  The determination of “bottom-up” enzyme costs based on process characteristics  The refinement of costs as better information Hard to perform robust techno-economic analysis of biofuel production processes Adding uncertainty in decision – making at many levels (From researchers & policymakers to investors & industry players)

78. Techno-economic model JBEI’s techno-economic analysis wiki site(http://econ.jbei.org) The process model includes all unit operations required for the production of cellulases from stream-exploded poplar by Trichoderma reesei  Feed stock transport, steam explosion, fungal growth, cellulase production and down stream processing (biomass filtration and incineration) Using model for Corn stover ethanol production with dilute acid pretreatment (Klein-Marcuschamber et al., 2010) to contextulaize the costs of enzymes based on typical ethanol yields.

79. Techno-economic model The costs of enzymes on a $/gal basis depend on the yield of ethanol. (1) The theoretical maximum yield based on conversion of all C5 and C6 sugars present in corn stover  111.4 gal/DT(dry tonne) (2) The yield based on conversion of C6 sugars at a 95% efficiency, but not C5 sugars  67.13 gal/DT (3) The yield based on conversion of all C5 and C6 sugars after a saccharification cellulose conversion of 70%  88.9gal/DT (4) The yield based on conversion of C5 and C6 sugars expected from a typical saccharification and a typical fermentation using engineered yeast  51.6 gal/DT Comparing (1)&(3) to (2)&(4), the main technological bottleneck for improving the overall yield is associated with fermentation Efficient fermentation does not impact the cost of enzyme production Saccharification efficiency is very high, but fermentation efficiency is remained. Saccharification efficiency does not improved, but fermentation efficiency is high.  Merely shift the burden for improvement to fermentation technologies Unattainable scenario

80. Techno-economic model The cost of one of the cheapest proteins available today : Soy protein (~$1.25/kg of protein, ~$500/MT)  If cellulases could be produced at this cost,  One can analyze the cost of production of biotechnologically derived enzymes using the process model presented here. scenarioCost contribution 1$0.08/gal 2$0.14/gal 3$0.11/gal 4$0.18/gal The cost contribution in the literature  Cellulases is cheap as soy protein  Ethanol yield is close to the theoretical maximum Unrealistic!!! scenarioCost contribution 1$0.68/gal 2$1.13/gal 3$0.85/gal 4$1.47/gal The baseline production cost of enzyme was found to be $10.14/kg.

81. Techno-economic model It is not possible to produce enzymes cheaper than plant-derived proteins. It is quite expensive simply to purchase the equipment for the production facility. Enzyme

82. Sensitivity analysis Sensitivity analysis for exploring the effect of parameter choices on the cost of enzymes 1) The fermentation residence time - Using carbon source simpler than poplar, such as amorphous cellulose - Using Trichoderma engineered for faster growth rates  Shorter fermentation residence time 2) The costs of raw materials - The chosen baseline cost for poplar ($60/MT) is low compared to projections of the cost of biomass(~$140/MT) - The choices of carbon sources and inducers used commercially are not publicly discussed.  The costs of raw materials are uncertain.

83. Sensitivity analysis The sensitivity of the contribution of enzyme costs to poplar price and fermentation residence time.  The change of poplar feedstock prices does not help.  Decreasing the growth and fermentation time would help through lowering the capital cost associated with fermentation.

84. Sensitivity analysis Lowering enzyme loadings required for saccharification would be an obvious target for improvement.  Including more stable cellulases  Better pretreatment technologies that enable high saccharification yields at lower enzyme loadings  Reducing the presence of phenolic compounds, which inhibit and deactivate cellulases 10FPU/g cellulases  5FPU/g cellulases (1) $0.68/gal (2) $1.13/gal (3) $0.85/gal (4) $1.47/gal (1) $0.34/gal (2) $0.55/gal (3) $0.42/gal (4) $0.73/gal Improving the economics of enzyme production is a significant hurdle!!!

85. Conclusion This research constructed techno-economic model for the production of fungal cellulases.  The cost of producing enzymes was much higher than that commonly assumed in the literature.  This model would be helpful to decrease the enzyme production cost The sensitivity analysis was performed to study the effect of feedstock prices and fermentation times on the cost contribution of enzymes to ethanol price.  The contribution can be lowered by shifting to lower cost feedstocks, reducing the fermentation times, and reducing the complexity of the process to drive down capital costs.  Achieving high overall biofuel yields at low enzyme loadings is much more promising by improved pretreatment and enzyme technologies, which is associated with biorefinery capital costs. A significant effort is still required to lower the contribution of enzymes to biofuel production costs.