<%BANNER%>

Biomass and Energy Yields of Bioenergy Germplasm Grown on Sandy Soils in Florida

Permanent Link: http://ufdc.ufl.edu/UFE0043046/00001

Material Information

Title: Biomass and Energy Yields of Bioenergy Germplasm Grown on Sandy Soils in Florida
Physical Description: 1 online resource (80 p.)
Language: english
Creator: KORNDORFER,PEDRO H
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ARUNDO -- BIOMASS -- CELLULOSE -- ELEPHANTGRASS -- ENERGYCANE -- FIBER -- HEMICELLULOSE -- LIGNIN -- MARGINAL
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The development of carbon-neutral energy sources has become one of the primary challenges of the twenty-first century. Perennial grasses such as energycane, giant reed, and elephantgrass, have been proposed as promising feedstocks for lignocellulosic ethanol or direct combustion, and are expected to be grown on marginal lands in the State of Florida. The objectives of this study were to evaluate biomass yields and estimate cellulose, hemicellulose, lignin, calorific value and ash concentration of energycane and elephantgrass genotypes and giant reed in low-input rainfed sandy soil farming systems of south Florida. Eight energycane genotypes, ?Merkeron? and ?Chinese Cross? elehphantgrass, giant reed and a check cultivar of energycane (L 79-1002) were compared at two different mineral soil sites with differing organic matter concentration. Our results indicate that energycanes and elephantgrasses are more appropriate bioenergy feedstocks than giant reed for marginal sandy soils of south Florida.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by PEDRO H KORNDORFER.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Gilbert, Robert A.
Local: Co-adviser: Sollenberger, Lynn E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043046:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043046/00001

Material Information

Title: Biomass and Energy Yields of Bioenergy Germplasm Grown on Sandy Soils in Florida
Physical Description: 1 online resource (80 p.)
Language: english
Creator: KORNDORFER,PEDRO H
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ARUNDO -- BIOMASS -- CELLULOSE -- ELEPHANTGRASS -- ENERGYCANE -- FIBER -- HEMICELLULOSE -- LIGNIN -- MARGINAL
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The development of carbon-neutral energy sources has become one of the primary challenges of the twenty-first century. Perennial grasses such as energycane, giant reed, and elephantgrass, have been proposed as promising feedstocks for lignocellulosic ethanol or direct combustion, and are expected to be grown on marginal lands in the State of Florida. The objectives of this study were to evaluate biomass yields and estimate cellulose, hemicellulose, lignin, calorific value and ash concentration of energycane and elephantgrass genotypes and giant reed in low-input rainfed sandy soil farming systems of south Florida. Eight energycane genotypes, ?Merkeron? and ?Chinese Cross? elehphantgrass, giant reed and a check cultivar of energycane (L 79-1002) were compared at two different mineral soil sites with differing organic matter concentration. Our results indicate that energycanes and elephantgrasses are more appropriate bioenergy feedstocks than giant reed for marginal sandy soils of south Florida.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by PEDRO H KORNDORFER.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Gilbert, Robert A.
Local: Co-adviser: Sollenberger, Lynn E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043046:00001


This item has the following downloads:


Full Text

PAGE 1

1 BIOMASS AND ENERGY YIELDS OF BIOENERGY GERMPLASM GROWN ON SANDY SOILS IN FLORIDA By PEDRO HENRIQUE KORNDRFER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 P edro Henrique Korn d rfer

PAGE 3

3 To my be loved family

PAGE 4

4 ACKNOWLEDGMENTS I wish to extend my deep thanks and gratitude to Dr. Robert A. Gilbert, Chairman of the Supervisory Comm ittee, for his guidance, assistance, and useful criticism throughout the duration of my graduate program. I feel indebted for his help, support, and understanding. I also would like to thank Dr. Zane R. Helsel, for his guidance, assistance and advice throu ghout my work. I would like to express my appreciation to Drs. John E. Erickson and Lynn E. Sollenberger, who where always there to answer my questions, contribute with their expertise to this research and for serving as members of the Supervisory Committe e. I am thankful for the endless love, encouragement and support of my family. To my parents, Gaspar and Clotilde, to my sisters, Let cia, Ana Paula, and Andr a Luisa, and to my fianc, Camilla, who pushed me forw a rd throughout my academic progress, my tha nks cannot be adequately acknowledged with words. The encouragement, help, and friendship of Miguel Castillo, Daniel Pereira, and Barbara Martin, is also sincerely acknowledged. Thanks also are due to the staff of the Everglades Research and Education Cent er at Belle Glade, Florida. I wish to thank Jairo Sanchez and Miguel Baltazar for their assistance in planting, sampling, and harvesting of the experiments, and laboratory analysis. Finally, I would like to acknowledge the support from Florida Crystals/Oke elanta Corporation and U.S. Sugar Corporation for providing the experiment al areas and the mechanical equipment for field preparation, crop management and harvests.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 ABSTRACT ................................ ................................ ................................ ................................ ... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 12 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 15 2.1 Background ................................ ................................ ................................ ................... 15 2.1.1 Use of Perennial Grasses as Energy Crops ................................ ....................... 15 2.1.2 Cellulosic Ethanol ................................ ................................ ............................. 15 2.1.3 State o f Florida: Potential for Biomass Energy Crop Production ..................... 16 2.2 Grasses as Lignocellulosic Feedstock ................................ ................................ ........... 16 2.2.1 Energycane ( Sacchar um spontaneum hybrids) ................................ ................. 18 2.2.2 Elephantgrass ( Pennisetum purpureum Schum.) ................................ .............. 19 2.2.3 Giant reed ( Arundo donax L.) ................................ ................................ ........... 21 3 CROP GROWTH AND BIOMASS YIELDS FROM BIOENERGY GRASS SPECIES GROWN ON SANDY SOILS OF FLORIDA ................................ ................................ ....... 23 3.1 Materials and Methods ................................ ................................ ................................ .. 25 3.1 .1 Soil and Climatic Conditions ................................ ................................ ............ 25 3.1 .2 Field Experiment Design and Management ................................ ...................... 26 3.1 .3 Genotype Selection ................................ ................................ ........................... 27 3.1 .4 Measurements and Analysis ................................ ................................ .............. 28 3.1 .5 Statistical Analysis ................................ ................................ ............................ 30 3.2 Results ................................ ................................ ................................ ........................... 31 3.2 .1 Tecan Analysis ................................ ................................ ................................ .. 31 3.2 .2 Townsite Analysis ................................ ................................ ............................. 32 3.2 .3 Combined Analysis ................................ ................................ ........................... 33 3.3 Discussion ................................ ................................ ................................ ..................... 34 4 FIBER CONCENTRATION, CALORIFIC VALUE, AND ASH CONCENTRATION OF BIOENERGY GRASS SPECIES GROWN ON SANDY SOILS OF FLORIDA .......... 49 4.1 Materials and Methods ................................ ................................ ................................ .. 50 4.1 .1 Measurements and Analysis ................................ ................................ .............. 51 4.1 .2 Statistical Analysis ................................ ................................ ............................ 53

PAGE 6

6 4.2 Results ................................ ................................ ................................ ........................... 54 4.2 .1 Tecan Analysis ................................ ................................ ................................ .. 54 4.2 .2 Townsite Analysis ................................ ................................ ............................. 55 4.2 .3 Combined Analysis ................................ ................................ ........................... 56 4.3 Discussion ................................ ................................ ................................ ..................... 57 5 SUMMARY AND CONCLUSIONS ................................ ................................ ..................... 72 LIST OF REFERENCES ................................ ................................ ................................ ............... 73 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 80

PAGE 7

7 LIST OF TABLES Table page 3 1 Analysis of variance F ratios and level of significance for juice concentration, Brix, and moisture concentration for crop and genotype effects and their interaction at the Tecan location. ................................ ................................ ................................ ................... 38 3 2 Juice concentration, Brix, and moisture concentration for giant reed, energy cane, and elephantgrass genotypes at Tecan. ................................ ................................ ..................... 38 3 3 Crop x genotype interaction means for Brix of giant reed, energycane, and elephantgrass genotypes at the Tecan location. ................................ ................................ 39 3 4 Analysis of variance F ratios and level of significance for fresh and dry yields for crop and genotype effects and their interaction at the Tecan location. .............................. 39 3 5 Fresh and dry yields for giant reed, energycane, and elephantgrass genotypes at Tecan. ................................ ................................ ................................ ................................ 40 3 6 Analysis of variance F ratios and level of significance for juice concentrat ion, Brix, and moisture concentration for crop and genotype effects and their interaction at the Townsite location. ................................ ................................ ................................ .............. 40 3 7 Juice concentration, Brix, and moisture concentration for giant reed and energycane genotypes at Townsite. ................................ ................................ ................................ ....... 41 3 8 Analysis of variance F ratios and level of significance for fresh and dry yields for crop and genotype effects and their interaction at Townsite. ................................ ............ 41 3 9 Fresh and dry yields for giant reed and energycane genotypes at Townsite. .................... 42 3 10 Analysis of variance F ratios and level of significance for juice concentration, Brix, and moisture concentration for crop, genotype, and site effects and their interactions. .... 42 3 11 Crop x genotype interaction means for Brix yield. ................................ ............................ 43 4 1 Analysis of variance F ratios and level of significance for cellulose, hemicellulose, and lignin concentration (conc.) and yield for crop and genotype effects and their interaction at Tecan. ................................ ................................ ................................ ........... 61 4 2 Cellulose, hemicellulose, and lignin concentrations and yield for giant reed, energycane, and elephantgrass genotypes at Tecan. ................................ .......................... 61 4 3 Analysis of variance F ratios and level of significance for ash concentration and HHV value for crop and genotype effects and their interaction at the Tecan location. ..... 62

PAGE 8

8 4 4 Crop x genotype interaction means for hemicellulose and HHV of giant reed, energycane, and elephantgrass genotypes at the Tecan location. ................................ ...... 62 4 5 Analysis of variance F ratios and level of s ignificance for cellulose, hemicellulose, and lignin concentration and yield for crop and genotype effects and their interaction at Townsite. ................................ ................................ ................................ ........................ 63 4 6 Genotype effect for cellulose, hemicellulos e, and lignin concentrations and yield for giant reed and energycane genotypes at Townsite. ................................ ............................ 63 4 7 Analysis of variance F ratios and level of significance for ash concentration and HHV for cro p and genotype effects and their interaction at the Townsite location. ......... 63 4 8 Analysis of variance F ratios and level of significance for cellulose, hemicellulose, and lignin concentration and total yield for crop, site, and genotype effects and their interaction. ................................ ................................ ................................ ......................... 64 4 9 Crop x genotype interaction means for hemicellulose yield. ................................ ............. 64 4 10 Genotype x site interaction means for hemicellulose yields of giant reed and energycane genotypes. ................................ ................................ ................................ ....... 65 4 11 Genotype x site x crop interaction means for cellulose concentrati on of giant reed and energycane genotypes ................................ ................................ ................................ 66 4 12 F ratios and level of significance for ash concentration and HHV for crop, site, and genotype effects and their interaction. ................................ ................................ ............... 67

PAGE 9

9 LIST OF FIGURES Figure page 3 1 Monthly rainfall, maximum and minimum temperatures at Tecan and Townsite locations. ................................ ................................ ................................ ............................ 44 3 2 Genotype leaf area index (LAI) in first ratoon crop at Tecan (A) and Townsite (B). ....... 45 3 3 Crop treatment effect means for juice concentration (A), Brix (B), and moist ure concentration (C) in the plant cane and first ratoon crops at Tecan ................................ 45 3 4 Crop means for fresh yield (A) and dry yield (B) in the plant cane and first ratoon crops at Tecan. ................................ ................................ ................................ ................... 46 3 5 Crop means for juice concentration (A), Brix (B), and moisture concentration (C) in the plant cane and first ra toon crop at Townsite ................................ ................................ 46 3 6 Crop by genotype interaction means for Brix in plant cane and first ratoon crops at Townsite ................................ ................................ ................................ ............................ 47 3 7 Crop effect means for fresh yield (A) and dry yield (B) in the plant cane and first ratoon crops at Townsite ................................ ................................ ................................ ... 47 3 8 Crop by site interaction means for juice concentration (A), Brix (B), and moisture concentration (C) in the plant cane and first ratoon crops at Tecan and Town site. ........... 48 3 9 Crop x site interaction means for fresh yield (A) and dry yield (B) in the plant cane and first ratoon crops at Tecan and Townsite. ................................ ................................ ... 48 4 1 Crop means for cellulose, hemicellulose, and lignin concentration (A) and dry yield (B) at Tecan ................................ ................................ ................................ ........................ 67 4 2 Crop treatment means for ash concentration an d higher heating va lue at Tecan. ............. 68 4 3 Crop effect means for cellulose, hemicellulose, and lignin concentration (A) and dry yield (B) at Tecan. ................................ ................................ ................................ .............. 68 4 4 Crop x genotype interaction means for cellulose concentration in the plant cane and first ratoon crops at Townsite. ................................ ................................ ........................... 69 4 5 Crop treatment effect means for ash concentration (A) and HHV (B ) in plant cane and first ratoon crop at the Townsite location. ................................ ................................ .. 70 4 6 Crop x site interaction effect means for cellulose, hemicellulose, and lignin concentration (A) and total yield (B) in th e plant cane and first ratoon crops at Tecan (TC) and Townsite (TS). ................................ ................................ ................................ .... 70 4 7 Crop x site interaction effect means for ash concentration (A) and HHV (B). ................. 71

PAGE 10

10 Abst ract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science BIOMASS AND ENERGY Y IELDS OF BIOENERGY G ERMPLASM GROWN ON SA NDY SOILS IN FLORIDA B y Pedro Henrique Korndrfer May 2011 Chair: Robert A. Gilbert Cochair: Lynn E. Sollenberger Major: Agronomy The development of carbon neutral energy sources has become one of the primary challenges of the twenty first century. Perennial grasses such as energycane [cross es of commercial sugarcane hybrids ( Saccharum spp.) with S. spontaneum ], giant reed ( Arundo donax L.), and elephantgrass ( Pennisetum purpureum Schum. ), have been proposed as promising feedstocks for lignocellulosic ethanol or direct combus tion, and are expected to be grown o n marginal land s The objectives of this study were to evaluate biomass yields and estimate fiber concentration (cellulose, hemicellulose, and lignin), calorific value and ash concentration of energycane and elephantgras s genotypes and g iant reed in low input rainfed sandy soil farming systems of south Florida. Eight energycane genotypes (875 3, US 74 1010, US 78 1011, US 78 1013, US 78 1014, US 82 1655, US 84 1047, and US 84 1066), Merkeron and Chinese Cross elehphan tgrass, g iant reed and a check cultivar of energycane (L 79 1002) were compared in a randomized complete block design at two different mineral soil sites with differing organic matter concentration Tecan and Townsite, with 50 and 15 g kg 1 organic matter, respectively. The experiment was evaluated during a two crop cycle (2008 to 2010). At Tecan, fresh yields ranged from 53.9 to 69.3 Mg ha 1 in plant cane and 44.1 to 56.7 Mg ha 1 in first ratoon. Fresh and

PAGE 11

11 dry yields for g iant reed at Tecan were extremely low, 10.1 and 5.3 Mg kg 1 respectively. Overall plant cane yields were lower and first ratoon yields were higher at Townsite than Tecan, but the same yield and fiber concentration differences between energycane and giant reed were observed. Our results sh owed that mean cellulose, hemicellulose, and lignin (mean of plant cane and first ratoon crops) were 400, 273, and 81 g kg 1 respectively across species. The overall higher heating value (calorific value) of plant cane and first ratoon crops for the speci es in the present study were 19.2 and 19.9 MJ kg 1 A sh concentration ranged from 24.6 to 31.6 g kg 1 and was positively correlated to the Si concentration Our results indicate that energycanes and elephantgrasses are more appropriate bioenergy feedstock s than g iant ree d for marginal sandy soils of south Florida.

PAGE 12

12 CHAPTER 1 INTRODUCTION The growing reliance on energy, the exhaustion of fossil fuel energy sources and significant changes in climate are combined factors that drive the increasing demand for alternative energy sources. The carbon dioxide concentration of the atmosphere is projected to increase by approximately 50% over the first 50 years of this century (IPCC, 2001). The major cause of this increase is continued combustion of fossil fuels. As a result, significant changes in climate will be accelerated, in particular global temperature increases. Therefore, the development of carbon neutral energy sources has become one of the primary challenges of the twenty first century. There has been incre asing concern in the U.S. regarding both sustainability and national security issues surrounding imported fuels. Biomass has always been a major source of energy for mankind and is presently estimated to contribute approximately 10 to en ergy supply (Kaygusuz and T rker, 2002; Gross et al., 2003; Parikka, 2004). In 2008, bioenergy provided 53% of all renewable energy, and 5% of the total energy produced in the U.S. (EIA, 20 09 ). Renewable energy sources can improve energy security, decrease urban air pollution, and reduce accumulation of carbon dioxide in the atmosphere (Lynd et al., 1991). Many energy production and utilization cycles based on cellulosic biomass have near zero greenhouse gas emissions on a life cycle basis (Lynd et al., 199 1), meaning that all the carbon dioxide released from plant energy will be sequestered in the subsequent crop, becoming a carbon neutral cycle. Biomass energy sources are attractive due to their carbon neutral balance and as renewable alternatives to exist ing transport fuels (Schell et al., 2004). The technology of ethanol production from non food plant sources is being developed rapidly for large scale production in the near

PAGE 13

13 future (Lin and Tanaka, 2006). However, production costs and species selection ha v e been major concern s (Cardinale et al., 2007). C ellulosic ethanol is defined as the production of ethanol from feedstock containing cellulose, such as agricultural wastes (sugarcane bagasse and corn stover), forestry wastes (woodchips), grasses ( Panicum v irgatum and Miscanthus x giganteus ) and others. S tudies on cellulosic ethanol production (Sticklen, 2008; Rubin, 2008; Lynd et al., 2008) have found that t he three main pillars are viable biomass production, cell wall degradation, and optimization of ferm entation of formed sugars ( US DOE, 2006). Species and genotype selection for biomass production is a fundamental step in the production of cellulosic ethanol (McKendry, 2002) Total biomass yield, as well as fiber concentration, cellulose, hemicellulose, an d lignin, must be taken into account. Identifying plants with high biomass production for marginal lands is critical for the successful implementation of cellulosic biofuels Perennial grasses are commonly considered as potential lignocellulosic feedstocks for both energy and fiber production. These grasses possess higher cellulose and lignin concentration than annual species (Dien et al., 2006), and are characterized by having relatively low moisture concentration at the end of their growing cycle, high wa ter and nutrient use efficiencies and low susceptibility to pest and diseases. Perennial plant species may have high adaptation ability to produce in marginal lands In addition, C4 perennial grasses are more efficient converter s of light, water, and nutri ent s into biomass than are C3 temperate species In summary, giant reed energycanes, and elephantgrasses are perennial grasses species that have many of the desired traits for addressing bioenergy crops, and they represent important species for cellulosic ethanol production and industrial biomass combustion in the southeastern

PAGE 14

14 USA. However, little research has been performed on biomass and energy yields of these species when grown on marginal lands in Florida. The objective of this thesis research was to d ocument realistic biomass yields, fiber characteristics, and energy cont ent of these species for emerging bioenergy industries in Florida

PAGE 15

15 CHAPTER 2 LITERATURE REVIEW Background Use of Perennial Grasses as Energy Crops Since the mid 1980s there has been significant interest in the use of perennial grasses as a renewable source of energy in the U.S. and Europe (Lewandowski et al., 2003). The production of transportation fuels, such as ethanol, as well as cogeneration for electricity and steam heat producti on are the two primary markets for bioenergy (McLaughlin and Kszos, 2005 ). Perennial grasses have high biomass production and lignocellulose concentration (McKendry, 2002) as well as high nutrient and water use efficiency (Lewandowski et al., 2003) attr ibutes which have contributed to interests in perennial grasses for biofuels. The literature review that follows discus s es the potential of grasses as lignocellulosic feedstocks, explores the potential of selected perennial grass genotypes as bioenergy cro ps, and evaluates their biomass production and composition when grown in marginal lands. Cellulosic E thanol Lignocellulosic biomass from grasses provide s an abundant and renewable source of sugars for the fermentation of cellulosic ethanol. Lignocellulose represents the majority of plant dry biomass. Lignocellulose is composed primarily of organic carbohydrates (cellulose, hemicellulose, and pectin), aromatic polymers (lignin) ( Fengel and Weneger, 1984; Ingram et al., 1999), and inorganic component s which remain as ash following combustion (Wiselogel et al., 1996). The plant cell wall is composed of fibrils of cellulose (polysaccharide with glucose as its monomer) embe dd ed in a polymeric matrix of hemicellulose (xylan, glucuronoxylan, arabinoxylan, or gluco mannan) and lignin (guaiacyl) (Aspinall, 1980) Lignocellulosic composition may vary among grass species, however it is generally composed of 20 0 to 50 0 g

PAGE 16

16 kg 1 cellulose, 20 0 to 40 0 g kg 1 hemicellulose, 2 0 to 20 0 g kg 1 pectin, and 10 0 to 20 0 g kg 1 ligni n (Ingram et al., 1999). Each of these components contributes to fiber properties, which ultimately impact the lignocellulosic feedstock properties. In the cellulosic ethanol conversion process cellulose and hemicellulose components are broken down into simple sugars and fermented. Another fraction of the biomass consists of lignin and insoluble extractives which are not fermented to ethanol (Chang and Holtzapple, 2000; Laureano Perez et al., 2005) L ignin suppress es the rate of enzymatic hydrolysis in li gnocellulosic ethanol processing by preventing the digestible parts of the substrate from hydroly zing (Chang and Holtzapple, 2000). Lignin and other extractives become a byproduct and may be used in combustion systems to generate heat energy or electricit y. State of Florida: P otential for B iomass E nergy C rop P roduction The southeastern regions of the U.S. where sunlight and rainfall are abundant and growing seasons are longer, is the optimum zone of the country for biomass production ( USDA, 2010 ). The sta te of Florida has favorable climatic conditions and low opportunity cost land for production of biomass crops, and can be considered as one of the leading areas in the United States to produce biomass as a source of renewable energy. Florida is a subtropic al region, with one of the wettest climates in the U.S. receiving 1150 to 2050 mm of rain (Stricker et al ., 1993). The average number of frost free days per year ranges from 240 in the P anhandle to 365 in the Keys (Sanford, 2003). The precipitation and so lar radiation throughout the year create favorable climatic conditions for biomass accumulation in Florida. Grasses as L ignocellulosic F eedstock Since the mid 1980s, various perennial grasses have been proposed for bioenergy in temperate and sub tropical r egions of the world. Giant reed ( Arundo donax L.) (Angelini et al.,

PAGE 17

17 2009), miscanthus ( Miscanthus x giganteus ) (Lewandowski et al., 2003; Atkinson, 2009; and Burner et al 2009), energycane ( Saccharum spontaneum hybrids) (Wang et al., 2008), and elephantg rass ( Pennisetum purpureum Schum.) (Woodard and Prine, 1993; Overman and Woodard, 2006; and Morais et al., 2009) have been evaluated for bioenergy potential in both Europe and the U.S. Perennial grasses are ideal energy crops for cellulosic ethanol product ion, because they contain high amounts o f renewable lignocellulosic biomass, may have low input production requirements, and are relatively inexpensive to produce (Kim and Dale, 2004; Bransby, 2008). According to Long (2008), an ideal energy crop would b e a photosynthetically efficient C4 plant, recycle nutrients through its root system, require low inputs (nutrients, water and pesticides), not have invasive characteristics, be resistant to pests and outcompete weeds, and would not be difficult to harvest. In cellulosic ethanol production, p lant biomass is converted to fermentable sugars for the production of biofuels using pretreatment processes that break down the lignocellulose and remove the lignin, thus allowing microbial enzyme access for cellulosic fe rmentation to ethanol (Sticklen, 2006). C ultivars containing high levels of cellulose would be preferred over those with high levels of lignin. Lignin plays an important role in structurally supporting the plant to prevent lodging. Thus, the benefits of li gnin will need to be balanced against its negative effect on the efficiency of cellulosic ethanol production in an energy crop breeding and selection program (Tew and Cobill, 2008). Successful production of cellulosic biofuel feedstocks will depend on impr ovement of plant characteristics through traditional and molecular breeding methods. When breeding for cellulosic ethanol production, selection criteria should include high fiber/sugar ratio, tolerance to

PAGE 18

18 biotic and abiotic stresses, high biomass yields, l imited or late flowering, and no seed (Jakob et al., 2009). Energycane ( Saccharum spontaneum hybrids) Saccharum spontaneum is one of the six species of the genus Saccharum (Daniels and Roach, 1987). Soltwedel in Java and Harrison and Bovell in Barbados, (l ater reviewed by Mangelsdorf, 1946, and Stevenson, 1965), performed the first hybridization efforts involving S officinarum (sugarcane) and S. spontaneum (wild sugarcane) species. This cross resulted in a hybrid (F 1 ) which was more robust than the parenta l material. Clones of S. spontaneum have long been used as sources of disease and pest resistance, ratooning performance, cold tolerance and vigor in sugarcane breeding programs for commercial sucrose production. However it was not until 2007 that L 79 100 2, a high fiber S. spontaneum hybrid, was released for energy proposes in the U.S. (Bischoff et al., 2008). Energycane and sugarcane are from the same genus, Saccharum however energycane s have higher fiber/sugar ratios, thinner stalks and higher plant pop ulations then sugarcane. L 79 1002 was evaluated from 2002 through 2005 by Legendre and Gravois (2007) at the USDA ARS Sugarcane Research Unit, in Houma, LA. They recorded average fresh weight cane yields of 70.7 Mg ha 1 (Bischoff et al., 2008). Giamalva e t al. (1984) reported average dry weight cane yields of L 79 1002 over a five harvest period of 21 Mg ha 1 in Louisiana. Brix and fiber concentration in the stalks of this cultivar were approximately 12 0 and 28 0 g kg 1 respectively. The reported fiber con centration was more than twice that of commercial sugarcane. L 79 1002 has shown promising energy characteristics, however, L 79 1002 has also shown increasing susceptibility to smut disease (caused by Ustilago scitaminea Sydow & P. Sydow) (Bischoff et al. 2008) in the field in LA and FL, thus new high yielding, disease free germplasm is needed.

PAGE 19

19 In the Everglades Agricultural Area in south Florida, where soils are primary organic Histosols, Deren et al. (1991) tested two inter specific hybrids between comm ercial sugarcane and S. spontaneum (US 72 1288 and US 79 1010) under seasonally flooded conditions. They reported dry biomass yields of 20 and 60 Mg ha 1 for the c ultivar US 72 1288 and 7 and 42 Mg ha 1 for the cultivar US 79 1010, in the plant cane (first year crop) and first ratoon crops (second year crop) respectively. In phosphatic clay soils, where fertility and water holding capacity are high energycane may be grown with minimal inputs thus reducing production costs Stricker et al. ( 1993 ) evaluated two energycane accessions in phosphatic clays in central Florida, and reported mean annual dry biomass yields of 43 and 49 Mg ha 1 yr 1 for L 79 1002 and US 72 1153 energycane, respectively. En ergycane yield s increased through the 4 y r study, suggesting strong ratooning ability According to Burner et al. (2009) it is important to select site specific crop species that fit an overall bioenergy management plan. To date, Florida and Louisiana are the largest sugarcane producers in the U.S. having sugarcane production areas of 150 660 and 157 950 hectares respectively (USDA, 20 09 ). E stablished sugarcane production infrastructure (planting, fertilizing, weeding and harvesting equipment ), grower experience in tall grass production and a favorable climate are important factors that make energycane production attractive in Florida. Elephantgrass ( Pennisetum purpureum Schum.) Elephantgrass is a tall rhizomatous perennial bunchgrass that may reach a height of 6 m and similar appearance to sugarcane introduced f rom tropical Africa in 1913 (Thompson, 1919). P opularly named as Napiergrass, usually propagated vegetatively and can also reproduce sexually but seed size are small and seedlings are weak (Burton, 1993). Initi ally exhibited

PAGE 20

20 potential as forage for cattle when introduced in the USA (Sollenberger et al., 1987), because of its high biomass production elep hantgrass exhibits potential for use as a bioenergy crop Burton (1989) registered Merkeron elephantgrass ( Pennisetum purpureum Schum. ) as a hybrid between dwarf elephantgrass, no. 208, and a tall selection, no. 1. Woodard et al (1991) compared elephantgrass (PI 300086 and Merkeron) to energycane (L 79 1002) for silage on sandy soils of north Florida, harvested one, two and three times per year. They reported average dry biomass yields of 27.3, 27.0 and 17.5 Mg ha 1 yr 1 for PI 300086, Merkeron and L 79 1002, respectively. However, biomass yields were reduced with increased frequency of harvest for all three genotypes. Prine et al. (1991) recorded elephantgra ss dry biomass yields ranging from 20 to 30 Mg ha 1 yr 1 in north Florida Prine et al. (1997) also grew elephantgrass on a range of soils in southern and central Florida using different cultural practices and reported dry biomass yields between 30 and 60 Mg ha 1 yr 1 Elephantgrass is adapted to soil s ranging from low fertility acid soils to slightly alkaline soils (Hanna et al., 2004). Elephantgrass is a C4 crop with high water use efficiency and high biomass yields that could produce large amounts of bi omass per acre, thus reducing land requirements needed for biomass feedstocks (Lewandowski et al., 2003). However, desirable traits for bioenergy crops may overlap with those of invasive non native species (Barney and DiTomaso, 2008) Ele phantgrass raises concerns to sugarcane growers in south Florida, since is commonly seen growing along canals and roadsides in sugarcane production areas (Rainbolt, 2005) and is found to have a high probability of becoming invasive (Gordon et al., 2011). Since there are no labeled herbicide for the selective control of elephantgrass and cultural and mechanical control less effective at the later stages of growth (Rainbolt, 2005), proper

PAGE 21

21 management practices should be taken into to consideration when growing elephantgrass nea r already established sugarcane areas. Giant reed ( Arundo donax L.) G iant reed i s a tall perennial grass species puta tively native to East Asia (Rossa et al., 1998), with a rapid growth rate of up to 8 m within a growing season (Perdue, 1958). Giant ree d i s one of the tallest of the herbaceous grasses (Lewandowski et al. 2003), and is found in all subtropical regions of both hemispheres (Herrera and Dudley, 2003). The Arundo genus belongs to the Poaceae family and is commonly named giant reed, bamboo re ed, Spanish reed, Italian reed, or wild cane. Despite its rapid growth rate, g iant reed is a C3 plant, although its high rates of photosynthesis and biomass productivity are similar to those of a C4 plant (Rossa et al., 1998; Christou, 2001). Giant reed fl owers in the late summer, but apparently does not produce viable seeds. Consequently, it is planted by rhizome propagation or vegetative planting of the stalk s Giant reed has exhibited potential for use as a bioenergy crop, with rapid growth, low nutrient requirements, and adaptation to different soil conditions, tolerating both heavy clays and loose sands (Perdue, 1958). According to Hartmann (2001), the lignin and cellulose concentration of g iant reed is higher than annual crops, which explains its abili ty to stand upright at lower water concentration s. Thus, giant reed may be better suited to late harvest than other biomass crops (Lewandowski et. al., 2003) In an experiment in Italy by Angelini et al. (2009), the biomass yield of giant reed was reported up to 42 Mg ha 1 in the plant cane crop under optimal conditions in a warm climate with sufficient irrigation. In the first ratoon crop, biomass dry yield increased by 43% (from 29 to 51 Mg ha 1 yr 1 ). Lewandowski et al. (2003) has shown that g iant reed c an maintain a relatively stable biomass yield from the second to fourth ratoon crop.

PAGE 22

22 Giant reed is considered by some scientists to be a noxious or invasive plant (Gordon et al., 2006; Barney and DiTomasso, 2008). Despite these concerns, g iant reed has als o been cited as a potential biomass energy crop (Lewandoski et al., 2003; Angelini et al., 2009). The date and location of the introduction of this species in the U.S. is unknown, but g iant reed has become an invasive weed problem in riparian areas of Cali fornia and watersheds of Texas (Herrera and Dudley, 2003). In wetlands of Florida, especially south Florida, giant reed invasive potential in areas near Lake Okeechobee has been evaluated (Gordon et al., 2006) The risk of g iant reed becoming an invasive w eed is considered low to moderate, although it has been found in 21 of 67 Florida counties. Other than its growth characteristics and its invasive ability, little is known about appropriate management practices for large scale production of this crop, and little has been published on its use for bioenergy production in the U.S

PAGE 23

23 CHAPTER 3 CROP GROWTH AND BIOM ASS YIELDS FROM BIOE NERGY GRASS SPECIES GROWN ON SANDY SOILS OF FL ORIDA The growing reliance on energy, the exhaustion of fossil fuel energy sources and significant changes in climate are factors that drive the increasing demand for alternative energy sources. Biomass energy sources are attractive due to their favorable carbon balance and as renewable alternatives to existing transport fuels (Schell et al., 2004). The technology of ethanol production from non food plant sources is being developed rapidly for large scale production in the near future (Lin and Tanaka, 2006) and the production costs and species selection ha ve been a major concern (Cardina le et al., 2007). According to Long (2008), ideal characteristics of a biomass crop are low input requirements C4 photosynthesis long canopy duration, high water use efficiency, ability to manage using existing farm equipment, disease and pest resistance eas e of harvest and storage in the field Successful production of cellulosic biofuel feedstocks will depend on improvement of plant characteristics through traditional and molecular breeding methods. Other selection criteria when breeding for cellulosi c ethanol include high fiber/sugar ratio, tolerance to biotic and abiotic stresses, high biomass yields, low or late flowering, and no grain set (Jakob et al., 2009). Some p erennial grasses have desirable attributes, including high amounts of renewable lig nocellulosic biomass, high nutrient and water use efficiency (Lewandowski et al., 2003) and relative ly inexpensive production (Kim and Dale, 2004; Bransby, 2008) These qualities have contributed to increased interest in perennial grasses as feedstocks fo r biofuel production. Since the mid 1980s, various perennial grasses have been proposed for bioenergy in temperate and sub tropical regions of the world. Giant reed ( Arundo donax L.) (Angelini et al., 2009), miscanthus ( Miscanthus x giganteus ) (Lewandowski et al., 2003; Atkinson, 2009; and Burner et al 2009), energycane ( Saccharum spontaneum hybrids) (Wang et al., 2008), and elephantgrass

PAGE 24

24 ( Pennisetum purpureum Schum.) (Woodard and Prine, 1993; Overman and Woodard, 2006; and Morais et al., 2009) have been evaluated for bioenergy potential in both Europe and the U.S. The southeastern regions of the U.S. where sunlight is abundant and growing seasons are long, is the optimum zone of the country for biomass production ( USDA, 2010 ). The state of Florida has fa vorable climatic conditions and low opportunity cost land for production of biomass crops, and can be considered as one of the leading areas in the United States to produce biomass as a source of renewable energy. Identifying plants with high biomass produ ction for marginal lands such as sandy soils in south Florida is critical for the successful implementation of cellulosic biofuels. In the Everglades Agricultural Area in south Florida, where soils are prima rily organic Histosols, Deren et al. (1991) tes ted two inter specific hybrids between commercial sugarcane and S. spontaneum (US 72 1288 and US 79 1010) under seasonally flooded conditions. They reported annual dry biomass yields of 20 and 60 Mg ha 1 for the cultivar US 72 1288 and 7 and 42 Mg ha 1 for the cultivar US 79 1010, in the plant cane and first ratoon crops, respectively. In phosphatic clays soils, where fertility and water holding capacity are high energycane may be grown with minimal inputs thus reducing production costs Stricker et al. ( 1 993 ) evaluated two energycane accessions in phosphatic clays in central Florida, and reported mean annual dry biomass yields of 43 and 49 Mg ha 1 yr 1 for L 79 1002 and US 72 1153 energycane, respectively. En ergycane yield s increased through the 4 yr stud y, suggesting strong ratooning ability Giamalva et al. (1984) reported average dry weight cane yields of L 79 1002 over a five year harvest period of 21 Mg ha 1 yr 1 in Louisiana. Brix and fiber concentration in the stalks of this cultivar were approximat ely 12 0 and 28 0 g kg 1 respectively. The reported fiber concentration is more than twice that of commercial sugarcane. L 79 1002 has shown promising

PAGE 25

25 energy characteristics, however, L 79 1002 has also shown increasing susceptibility to smut disease (cause d by Ustilago scitaminea Sydow & P. Sydow) in the field in LA and FL Elephantgrass is adapted to soil s ranging from low fertility acid soils to slightly alkaline soils (Hanna et al., 2004). Woodard et al (1991) compared elephantgrass (PI 300086 and Merker on) to energycane (L 79 1002) for silage on sandy soils of north Florida, harvested one, two and three times per year. They reported average dry biomass yields of 27.3, 27.0 and 17.5 Mg ha 1 yr 1 for PI 300086, Merkeron and L 79 1002, respectively. Prine et al. (1991) recorded elephantgrass yields ranging from 20 to 30 Mg ha 1 yr 1 in north Florida Prine et al. (1997) also grew elephantgrass on a range of soils in southern and central Florida using different cultural practices and reported yields between 30 and 60 Mg ha 1 yr 1 In an experiment in Italy by Angelini et al. (2009), the dry biomass yield of g iant reed was reported up to 42 Mg ha 1 yr 1 in the plant cane crop under optimal conditions in a warm climate with sufficient irrigation. In the first r atoon crop, biomass dry yield increased by 43% (from 29 to 51 Mg ha 1 yr 1 ). While energycane, elephantgrass and giant reed have been evaluated for biomass production in Europe and U.S. there is a lack of information on biomass yields when grown on margi nal sandy soils in Florida. This information is crucial as this is the area where biomass production is most likely to occur. The objectives of this study were to : 1) compare plant growth; 2) quantify fresh and dry biomass yields; and 3) quantify Brix, jui ce and moisture concentration of selected perennial grass genotypes in a two crop cycle grown on marginal lands in Florida. Materials and Methods Soil and Climatic Conditions The experiment was conducted at two mineral soil sites with differing soil organ ic matter concentration The first location, managed by Florida Crystals Corporation (Tecan), contained soil of the Pomona series (sandy, siliceous, hyperthermic Ultic Alaquods). This soil is

PAGE 26

26 characterized by an Ap horiz on (to a depth 15 20 cm) with 50 g k g 1 organic matter concentration soil pH of 4.7, and Mehlich 1 extractable P, K, Ca Mg and Si of 6.2, 22, 847, 24 and 1.1 mg kg 1 respectively. The second location, managed by United States Sugar Corporation (Townsite), also contained soil of the Pomo na series. However the Townsite location had a lower organic matter con centration of 15 g kg 1 average soil pH of 7.3, and higher P, K, Ca Mg and Si of 10.4, 31, 2515, 148 and 113.1 mg kg 1 respectively. The coordinates of the field boundaries for eac h experiment were as follows: Tecan (26 o 37.84 N, 80 o 56.19 W), (26 o 37.64 N, 80 o 56.19 W), (26 o 37.64 N, 80 o 56.10 W), (26 o 37.84 N, 80 o 56.09W); and Townsite (26 o 43.95 N, 80 o 59.18 W), (26 o 44.37 N, 80 o 59.18 W), (26 o 44.37 N, 80 o 59.23 W), (26 o 43.95 N, 80 o 59.23 W). Air temperature and rainfall were recorded by a meteorological station of the Florida Automated Weather Network (FAWN), Clewiston, FL at a distance of 1 km from Townsite and 10 km from Tecan field trials. Average monthly maximum and minimum temperatures, as well as precipitation during the 27 mo. period of the study are shown in Figure 3 1. Field Experiment Design and Management Large scale plots were used at both locations to facilitate mechanical harvesting with commercial equipment Plots measured four rows wide by 12. 2 m long, with 1.5 m and 1.7 m between row spacing, at Townsite and Tecan, respectively. The experiment was planted in a randomized complete block design with 3 replications. Fertilization, weed and pest control at both sites followed custo mary commercial sugarcane practices. Our goal was to determine how the germplasm used in this experiment would perform under standard field management practices and under rainfed conditions All plots at both locations received 150, 170, and 170 kg ha 1 of N, P 2 O 5 and K 2 O respectively, one week prior to

PAGE 27

27 planting and 160, 160, and 170 kg ha 1 of N, P 2 O 5 and K 2 O respectively, eight weeks after plant cane harvest. Genotype Selection Energycane genotypes were initially selected from wide crosses of S. spo ntaneum x commercial sugarcane made at the U.S. Department of Agriculture Agricultural Research Service (USDA ARS) Sugarcane Field Station at Canal Point, FL. Fifty three of these genotypes were planted in December 2006 in single replicate plots on a Laude rhill muck soil (euic, hypertherm ic and Lithic Haplosaprist) at the University of Florida Everglades Research and Education Center (EREC) in Belle Glade, FL. From these 53 genotypes, 21 were selected for this study based on superior plant cane yields at ER EC. For this present research project, only 12 genotypes from the experimental plots at Tecan and Townsite were selected based on plant cane yields at those sites to proceed with a complete fiber concentration, ash concentration and high er heating value a nalysis for this study. From the selected energycane genotypes, only nine were common to both sites (Tecan and Townsite), 875 3; US 74 1010; US 78 1011; US 78 1014; US 78 1013; US 82 1655; US 84 1047; US 84 1066 (CP energycane genotypes) and g iant reed Th ree other genotypes were added to the selection from the Tecan site, L 79 1002 (Bischoff et a l., 2008), used as an energycane check, and Merkeron and Chinese Cross elephantgrass ( Pennisetum purpureum Schum.). The check cultivar, L 79 1002, was not plan ted at Townsite because insufficient seedcane was produced to plant a second experiment. Elephantgrass genotypes were also not used at Townsite because of grower concerns regarding the weed potential of this specie s in adjacent sugarcane fields (Rainbolt, 2005) The Tecan location was planted on 19 Dec. 2007 and the Townsite location on 21 Feb. 2008. All plots were planted vegetatively using stalk cuttings from the germplasm collection at

PAGE 28

28 EREC. The EREC seedcane was harvested at a 10 cm stubble height on t he day of planting. Two stalks were placed horizontally side by side in furrows 15 to 20 cm deep, and chopped into sections approximately 60 cm long to remove apical dominance and improve seedcane germination. Measurements and Analysis Leaf area index (LAI ) for all selected genotypes was measured once per month for a period of 9 months in the first ratoon crop (2009 2010) at Tecan on 27 Jan 17 Feb 17 Ma r. 14 Apr 18 May, 16 Jun e 16 Jul y. 18 Aug and 18 Sep t. 2009 (41, 62, 90, 118, 152, 181, 211, 24 4 and 275 days after plant cane harvest) and Townsite for a period of six months, with measurements performed on 15 Apr 12 May, 12 June, 20 July, 19 Aug and 23 Sept 2009 (34, 61, 92, 130, 160 and 195 days after plant cane harvest). A SunScan Canopy Analysis System (Dynamax Inc., Houston, TX) was used to estimate leaf area index This system uses a 1 .0 m wand placed beneath the crop canopy to measur e transmitted photosynthetic active radiation (PAR) and compares it to simultaneous measurements of unsh aded radiation from a globe sensor outside the plots to calculate LAI. Since the inter row spacing of the plots was 1.5 m and the SunScan wand is 1.0 m, two measurements were taken diagonally across the measurement row, spanning from mid row to mid row, an d these readings averaged to get one LAI measurement. This procedure was repeated twice per plot, and all readings were performed between 10 and 14 h (Gilbert et al., 2008). Plant cane crop results were not conducted because LAI equipment was not available at the time. LAI equipment failure precluded measurements after the month of September on the first ratoon crop at Townsite. Two days prior to final biomass yield measurements at Tecan (plant cane, 15 Dec. 2008; and first ratoon, 13 Dec. 2009), and at Town site (plant cane, 10 Mar. 2009; and first ratoon, 8 Mar. 2010), ten stalks were collected at random from the two center rows of each plot. These ten

PAGE 29

29 stalk samples were chopped in a modular sugarcane disintegrator (Codistil S/A Denini, Mod: 132S, Piracicaba SP, Brazil). All chopped plant material (including leaf, stem and tops) from each sample was mixed in an individual 200 L barrel. From ea ch of the plot samples, two sub samples were collected. Subsample 1, composed of 800 to 1000 g fresh weight was used to measure juice concentration and Brix (total soluble solids). Subsample 2, composed of 500 to 700 g of fresh weight biomass was used to measure moisture concentration. For juice concentration, Subsample 1 fresh weight was recorded and then samples were pla ced in a hydraulic press (Codistil S/A Dedini, Mod: D 2500 II, Piracicaba SP, Brazil), for 30 sec. until pressure attained 3000 psi. The resultant filter cake press sub sample was weighed and the extracted juice, juice volume and weight were measured to det ermine the juice concentration from biomass. Approximately 5 ml of juice was used to determine total soluble solids (Brix) in a refractometer (Bellingham and Stanley Inc., RFM 91, England). Subsample 2 was placed in a paper bag (Duro Bag Mfg., #25, Florenc e, KY), fresh weight was recorded using a balance with weighing capacity up to 8100 g x 0.1 g (Metter Toledo, Mod: PB 8001 S), and the sub sample was dried at 60C until constant weight to determine dry biomas s concentration and moisture concentration Plot s were harvested at Tecan on 19 Dec. 2008 (plant cane) and 15 Dec. 2009 (first ratoon). Townsite harvests occurred on 12 Mar. 2009 (plant cane) and 10 Mar. 2010 (first ratoon). The border rows (rows 1 and 4) were removed and the center two rows of each plo t were cut using a commercial sugarcane harvester. Harvesters used at Tecan on plant cane harvest and firs t ratoon were Case IH 7700 (Piraci caba, Brazil) and Case IH 8800 (P iraci caba, Brazil), respectively, and at Townsite a John Deere 3510 model (Catalo, Brazil) was used for both crop harvests. These harvesters were followed by a tractor (John Deere, 7210, Moline, IL) pulling a

PAGE 30

30 weigh wagon (Cameco Ind., 3 Ton Sample Wagon, Thibodaux, LA). All primary and secondary fans from the sugarcane harvester were tu rned off to ensure all harvested aboveground biomass, including leaves and tops, w as deposited into the weigh wagon. The weigh wagon was equipped with load cells (Avery Weigh Tronix, 715), and the two center rows of each plot were weighed individually befo re taring the scale and proceeding to the following plot. Total d ry biomass yields were calculated using fresh weights recorded by the weigh wagon at harvest multiplied by the dry biomass concentration of subsample 2. Statistical Analys i s Data were analyze d using SAS software (SAS Inc., 1996 ; Littell et al., 2002). Analys es of variance for all yield data were performed for a randomized complete block design using the general linear model (PROC GLM) of the SAS software. LAI measurements at each sampling date were an alyzed using ANOVA with genotype as the independent variable. Mean separations P = 0.05 unless otherwise noted. Three separate analys i s were performed and are prese nted separately for genotypes grown at the 1) Tecan location ( 875 3; US 74 1010; US 78 1011; US 78 1014; US 78 1013; US 82 1655; US 84 1047; US 84 1066 ; giant reed; Merkeron; and Chinese Cross) 2) Townsite location ( 875 3; US 74 1010; US 78 1011; US 78 10 14; US 78 1013; US 82 1655; US 84 1047; US 84 1066 ; and giant reed) and 3) C ombined analysis ( 875 3; US 74 1010; US 78 1011; US 78 1014; US 78 1013; US 82 1655; US 84 1047; US 84 1066 ; and giant reed) for the common set of genotypes grown at both Tecan an d Townsite. Main effects of crop and genotype and their interactions were analyzed for the Tecan and Townsite data sites whereas main effects of crop, genotype site and their interactions were analyzed for the combined data set.

PAGE 31

31 Results Significant diffe rences in first ratoon crop LAI were noted beginning 75 days after plant cane harvest at Tecan ( Figure 3 2 A) and Townsite ( Figure 3 2 B). By the final measurement date at 275 days after harvest ( DAH ) LAI of the Merkeron elephantgrass and CP energycane geno types was significantly greater than the Chinese Cross elephantgrass and energycane genotype L 79 1002, which was significantly greater than giant reed at the Tecan location ( Figure 3 2 A). LAI of the CP energycanes was significantly greater than giant reed from 120 to 190 days after plant cane harvest ( Figure 3 2 B). Tecan Analysis At Tecan, the effects of crop and genotype were highly significant for juice concentration Brix and moisture concentration and the crop x genotype interaction was significant f or Brix (Table 3 1). Figure 3 3 presents plant cane and first ratoon crop treatment effect means for juice concentration (A), Brix (B), and moisture con centration (C) at Tecan. The plant cane crop had greater means than the first ratoon crop for all three traits with plant cane recording 424 118 and 648 g kg 1 greater juice concentration Brix and moisture concentration respectively The juice concentration and Brix means of 14 6 and 93 g kg 1 for giant reed and 26 5 and 77 g kg 1 for Merkeron, were lowe r than the energycane genotypes Energycane means ranged from 38 6 to 427 g kg 1 for juice concentration and 114 to 127 g kg 1 for B rix (Table 3 2) Data for these traits were not obtained for the elephantgrass genotype Chinese Cross. All energycane and el ephantgrass moisture concentration s were higher than giant reed at Tecan (Table 3 2). The crop x genotype interaction for Brix at Tecan (Table 3 3 ). The genotypes giant reed, Merkeron, and US 78 1013 increased from the plant cane to first ratoon crops for Brix, however, all other genotypes decreased in Brix from plant cane to first ratoon.

PAGE 32

32 Crop and genotype had significant effects on fresh yields, however only the genotype effect was significant for dry yields. The interaction of crop x genotype was not s ignificant for either trait (Table 3 4 ). Figure 3 4 presents the crop treatment effect means for fresh (A) and dry yield (B) in the plant cane and first ratoon crops at Tecan. The plant cane crop had 18.5% greater fresh yields than the first ratoon crop, h owever the dry yields were similar across crops Fresh and dry yields for giant reed at Tecan were lower (only 10.1 and 5.3 Mg ha 1 ) than the energycane and elephantgrass genotypes (Table 3 5 ). Merkeron elephantgrass was notable for high fresh and dry yiel ds. US 74 1010 recorded lower fresh and dry yields than US 78 1013, but in general there were few differences in yields among energycane genotypes. Townsite A nalysis At the Townsite location, the crop and genotype effects were significant for juice concent ration, Brix and moisture concentration (Table 3 6 ). However, the crop x genotype interaction was significant only for Brix Figure 3 5 presents plant cane and first ratoon crop means for juice concentration (A), Brix (B), and moisture concentration (C). J uice and moisture concentration s were significantly lower in plant cane than first ratoon at Townsite. However, Brix was significantly greater in plant cane than the first ratoon crop. At Townsite, the juice concentration Brix and moisture concentration s of giant reed were lower than energycane genotypes (Table 3 7 ). Giant reed had only 77 g kg 1 juice concentration compared to a range of 331 to 363 g kg 1 for the energycane genotypes. Thus while the Brix values of giant reed juice concentration were 17% percent lower than the energycanes, the total sugar yields would be much lower due to the lower total juice concentration in the plant.

PAGE 33

33 Figure 3 6 presents the significant crop x genotype interaction means for Brix at Townsite. Brix was greater for all ene rgycane genotypes in plant cane than the first ratoon crop with an average decrease of 30% between crops. In contrast, giant reed had a 13.5% increase from the plant cane to t he first ratoon crop Crop and genotype had significant effects on fresh and dry yields at Townsite, but the crop x genotype interaction was not significant for these traits (Table 3 8 ). Figure 3 7 presents the crop treatment effect means for fresh (A) and dry yield (B) in the plant cane and first ratoon crops at Townsite. The plant ca ne crop had lower fresh and dry yields than the first ratoon crop. Fresh yield increased 29.5 percent and dry yield increased 30.6 percent from plant cane to first ratoon. Fresh yield s of the energycane genotypes ( 4 4 .0 to 52 .1 Mg ha 1 Table 3 9 ) were gre ater than fresh yields of giant reed (10.6 Mg ha 1 ). Dry yield means of the energycane genotypes, 17. 9 to 20.4 Mg ha 1 per year, were also greater than giant reed ( 6.0 Mg ha 1 ). Fresh and dry yields were not different among energycane genotypes. Combined A nalysis Crop, genotype, site, and crop x site interactions had highly significant effects on juice concentration, Brix and moisture concentration The crop x genotype interaction was significant only for Brix Thus, this analysis will focus primarily on the crop x site interaction means (Table 3 10 ). The c rop by genotype interaction was significant for Brix yield (Table 3 1 1 ). Giant reed recorded an increase in Brix from plant cane to the first ratoon crop, whereas all the energycane genotypes recorded a decrease in Brix between crops. Figure 3 8 presents the significant crop x site interaction for juice concentration (A), B rix (B), and moisture concentration (C). The three traits exhibited different trends across crops at the

PAGE 34

34 two locations. Juice concent ration was greater in plant cane than first ratoon at Tecan, but greater in first ratoon than plant cane at Townsite. The opposite trend was recorded for moisture concentration with first ratoon means greater than plant cane at Tecan but plant cane greater than first ratoon at Townsite. Brix means were higher in plant cane than first ratoon at both locations, however the difference was greater at Townsite than Tecan. The effect of genotype, site and the crop x site interaction was highly significant for fre sh yiel d crop, genotype and crop x site interactions were significant for dry yield. All other interactions were not significant for either trait (Table 3 1 2 ). Figure 3 9 presents the significant crop x site interaction effect means for fresh yield (A) an d dry yield (B). At Tecan, plant cane fresh and dry yields were greater than first ratoon, but at Townsite, first ratoon yields were greater than plant cane. Discussion A t Tecan, the leaf area index of the energycane clones declined after reaching peak est imated LAI of 6.0 at 241 days after plant cane harvest. The LAI of L 79 1002 energycane, Merkeron and Chinese Cross elephantgrass increased steadily up to 210 days after plant cane harvest. Giant reed had a slow increase in LAI, reaching its peak LAI of on ly 2.8 at 241 days after plant cane harvest. At Townsite, giant reed LAI was much lower than CP energycanes. The low LAI of giant reed impacted photosynthesis and consequently fresh and dry biomass yields. The average air temperature at Tecan during the in itial 5 months of growth w as 19.7C during plant cane and 18.6C during the first ratoon crop. Total rainfall during the same period was 25 2 mm during plant cane crop and only 19 mm during first ratoon crop. At Tecan, plant cane fresh yields were greater t han first ratoon. The difference between crops may be explained by the fact that rainfall was greater during the initial period of crop establishment in plant cane than first ratoon crop.

PAGE 35

35 For thermal conversion, feedstock requires low moisture concentratio n typically less than 500 g kg 1 while lignocellulosic ethanol processing can utilize high moisture concentration feedstocks (McKendry, 2002). Energycane and elephantgrass genotypes are thus more appropriate for lignocellulosic ethanol than thermal conve rsion because their moisture concentration s ranged from 600 to 667 g kg 1 Giant reed recorded a moisture concentration of 470 g kg 1 and thus may be better suited for thermal conversion. L ow Brix is desired for lignocellulosic ethanol production. High Brix may suppress enzymatic hydrolysis by microorganisms. When plants are subjected to drought stress, Brix increases. At Townsite, energycane genotypes had greater Brix in plant cane compared to the first ratoon. This occurred because rainfall was only 49 mm from 12 Nov 2008 to 12 Mar 2009, prior to plant cane harvest. In contrast, higher rainfall (17 7 mm) occurred from 10 Nov 2009 to 12 Mar 2010, prior to first ratoon harvest, which likely increased moisture concentration and decreased Brix Results from the combined data set, which include d giant reed and CP energycane genotypes common to both locations, showed that Brix was lower at Tecan than at Townsite. This may be explained by the difference in soil organic matter content between locations. The Tecan locations had 50 g kg 1 soil organic matter and Townsite only 15 g kg 1 Higher levels of soil organic matter mean more available water to plants, lower stress and lower Brix Greater available water in the soil stimulate s plant s to increase biomass (leaf area), and this may cause reduction of sucrose due to the dilution effect and energy consumption. The average dry biomass yields recorded for Merkeron and Chinese Cross elephantgrass in this study (26.3 and 20. 4 Mg ha 1 respectively) are lower than dry b iomass yields reported by Bouton (2002) for Merkeron ranging from 27.7 to 30.7 Mg kg 1 and elephantgrass lines tested

PAGE 36

36 by Prine et al. (1997) ranging from 30 to 60 Mg kg 1 in south and central Florida. These results are approximately one third of those re ported under optimal conditions by Andrade et al. (2005) The low dry biomass yields observed for elephantgrass in this study can be explained by the low fertility soil (sandy soil) and climatic conditions. However at Tecan, Merkeron had the greatest yield s among the genotypes evaluated in this study. The cultivar L 79 1002 was released because of its high fiber concentration excellent rat o oning ability and vigorous growth habit. According to Bischoff et al. (2008), L 79 1002 is moderately susceptible to smut disease. Stricker et al. (1993) reported dry biomass yield s reaching up to 42.6 Mg ha 1 (4 yr means) when grown in a phosphatic clay soil in central Florida. This productivity is much higher than that observed in sandy soil s of our experiment ( 18.6 Mg ha 1 ) for the same cultivar. The results of the experiment demonstrated that CP energycanes recorded similar yields to L 79 1002. The overall dry biomass yields from giant reed 3.8 and 5.1 Mg ha 1 in plant cane and first ratoon, respectively, were lower t han dry biomass yields observed in trials from southern Italy (10.6 and 22.1 Mg ha 1 in plant cane and first ratoon, respectively) (Cosentino et al., 2006), northern Italy (13.0 and 37.0 Mg ha 1 in plant cane and first ratoon, respectively) (Angelini et al., 2005), and Spain (17.0 and 22.5 Mg ha 1 in plant cane and first ratoon, respectively) (Christou, 2001). One reason for lower dry biomass yields of giant reed in our study could be related to the method of propagation. In the present study, seedcane c uttings was used rather than rhizomes, which may explain the large gaps in rows of giant reed plots and lower LAI and biomass yields. However, rhizome propagation for giant reed under rain fed conditions on marginal lands may not be economically viable for commercial biofuel production in Florida. Our study used commercial seedcane production practices used in sugarcane cultivation and

PAGE 37

37 giant reed performance was extremely poor in this farming system. According to Lewandowski et al. (2003), the production of giant reed is more economical and environmentally favorable under moderate irrigation. Our results showed that giant reed was approximately one fourth as productive as energycane and elephant grass genotypes. The poor performance of giant reed in sandy soi ls could also be partially associated with the wide row spacing and 12 mo. harvest frequency that followed commercial sugarcane standards, whereas planting s performed with narrower rows (Gilbert et al., 2008) and more frequent cuttings could increase the l ow yields reported in this study. Based on yields, o ur results indicate that energycanes and elephantgrasses are more appropriate bioenergy feedstocks than giant reed for marginal sandy soils of south Florida managed under sugarcane cultivation practices R ainfall during the establishment period is essential for high biomass yields on sandy soils. Fresh and dry yields from new energycane genotypes were similar to L 79 1002 energycane.

PAGE 38

38 Table 3 1 Analysis of variance F ratios and level of significance f or juice concentration Brix, and moisture concentration for crop and genotype effects and their interaction at the Tecan location. Treatment Juice Concentration Brix Moisture Concentration Crop (C) 157.8 *** 7.3 ** 144.0 *** Genotype (G) 25.6 *** 17.6 ** 56.4 *** C x G 0.8 2.6 1.8 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively. Table 3 2 Juice concentration Brix, and moisture concentration for giant reed ener gycane and elephantgrass genotypes at Tecan. Genotype Juice Concentration Brix Moisture Concentration ________________________ g kg 1 ________________________ Giant reed 4 63c 93 c 471 f Merkeron 26 5b 77 d 600e L 79 1 002 42 7a 11 7ab 667 a 875 3 394 a 119 ab 627 cd US 74 1010 39 7a 121 ab 65 4ab US 78 1011 38 6a 12 7a 63 3cd US 78 1013 3 90a 126 ab 645 bc US 78 1014 392 a 12 5a 63 7bc US 82 1655 38 6a 127 a 643 bc US 84 1047 39 2a 11 9ab 6 40bc US 84 1066 39 5a 114 b 653 ab Chinese Cross 61 4de LSD 0.05 45.96 10 5 19.33 Treatment means followed by the sa me letter are not significantly different ( P Elephantgrass ( Pennisetum purpureum ) genotypes.

PAGE 39

39 T able 3 3 Crop x genotype interaction means for Brix of giant reed, energycane, and elephantgrass genotypes at the Tecan location. Genotype Brix _____ g kg 1 _____ Crop PC 1R Giant reed 101 115 Merkeron 68 87 L 79 1002 122 110 875 3 124 115 US 74 1010 123 119 US 78 1011 123 121 US 78 1013 116 130 US 78 1014 131 118 US 82 1655 133 121 US 84 1047 124 113 US 84 1066 122 107 Chinese Cross PC = plant cane; 1R = first ratoon. Table 3 4 Analysis of variance F ratios and level of significance for fresh and dry yields for crop and genotype effects and their interaction at the Tecan location. Treatment Fresh Yield Dry Yield Crop (C) 24.8*** 2.6 Genotype (G) 14.2** 12.0*** C x G 0.8 0.7 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively. Yields include the entire aboveground portion of the plant (stalks, tops, and dead leaves).

PAGE 40

40 Table 3 5 Fresh and dry yields for giant reed energycane and elephantgrass genotypes at Tecan. Genotype Fresh Yield Dry Yield __________________ Mg ha 1 __________________ Giant reed 10.1 d 5.3 d Merkeron 65. 8a 26. 4a L 79 1002 56. 6abc 18.6 bc 875 3 59.3 abc 21.8 bc US 74 1010 51. 7c 17.9 c US 78 1011 5 6.0abc 20. 5bc US 78 1013 63. 1ab 22. 2b US 78 1014 52. 6bc 18. 9bc US 82 1655 58. 4abc 20.6 bc US 84 1047 59. 4abc 21.0 bc US 84 1066 60.0abc 20.6 bc Chinese Cross 53.0 bc 20. 5bc LSD 0.05 10.74 4.04 Treatment means followed by the same letter ar e not different ( P Yield includes the entire aboveground portion of the plant (stalks, tops, and dead leaves). Table 3 6 Analysis of variance F ratios and level of significance for juice concentration Brix, and moisture concentration for crop and genotype eff ects and their interaction at the Townsite location. Treatment Juice Conc. Brix Moisture Conc. Crop (C) 24.0 *** 150.8 *** 37.2 *** Genotype (G) 44.1 *** 2.7 49.4 *** C x G 0.6 5.1 *** 1.6 *, **, *** Significa nt at the < 0.05, 0.01, and 0.001 levels of probability respectively.

PAGE 41

41 Table 3 7 Juice concentration Brix, and moisture concentration for g iant reed and energycane genotypes at Townsite. Genotype Juice Concentration Brix Moisture Concentration __ _____________________ g kg 1 _______________________ Giant reed 7 7 b 122 b 431 b 875 3 33 7a 146 a 586 a US 74 1010 340 a 14 6a 58 7a US 78 1011 363 a 144 a 587 a US 78 1013 333 a 149 a 57 9a US 78 1014 341 a 14 3a 586 a US 82 1655 349 a 153 a 589 a US 84 1047 331 a 150 a 57 7a US 84 1066 35 5a 14 3a 587 a LSD 0.05 38.79 1.55 20.95 Treatment means followed by the same letter are not significantly different ( P Table 3 8 Analysis of variance F ratios and level of significance for fresh and dry yields for crop and genotype effects and their interaction at Townsite. Treatment Fresh Yield Dry Yield Crop (C) 55.5 *** 62.4 *** Genotype (G) 17.6 *** 13.7 ** C x G 1.2 1.4 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively.

PAGE 42

42 Table 3 9 Fresh and dry yields for giant reed and energycane genotypes at Townsite. Genotype Fres h Yield Dry Yield _____________________ Mg ha 1 ______________________ Giant reed 10.6 b 6.0b 875 3 46.4 a 18.2 a US 74 1010 46. 2a 18. 2a US 78 1011 50. 2a 19.7 a US 78 1013 48.5 a 19.6 a US 78 1014 47. 6a 18. 8a US 82 1655 52.4 a 20.4 a US 84 1047 4 4 .0a 17. 9a US 84 1066 51. 8a 20. 4a LSD 0.05 8.82 3.48 Treatment means followed by the same letter are not different ( P Yields includes the entire aboveground portion of the plant (stalks, tops, and dead leaves). Table 3 10 Analysis of vari ance F ratios and level of significance for juice concentration Brix and moisture concentration for crop, genotype and site effects and their interaction s Treatment Juice Conc. Brix Moisture Conc. Crop (C) 27.8 *** 131.4 *** 11.4 ** Genotype (G) 52.3 ** 7.6 *** 99.7 *** Site (S) 40.0 *** 124.6 *** 229.6 *** C x G 0.2 5.9 *** 0.6 C x S 117.8 *** 62.6 *** 134.2 *** G x S 0.4 0.5 1.1 C x G x S 0.9 1.4 1.4 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively.

PAGE 43

43 Table 3 1 1 Crop x genotype interaction means for Brix yield. Genotype Brix __________ g kg 1 __________ Crop PC 1R Giant reed 99 116 875 3 14 9 11 7 US 74 1010 146 12 1 US 78 1011 15 2 119 US 78 1013 153 119 US 78 1014 14 8 119 US 82 1655 1 60 12 0 US 84 1047 14 8 12 1 US 84 1066 14 5 11 2 PC = plant cane; 1R = first ratoon

PAGE 44

44 Figure 3 1 Monthly rainfall, maximum and minimum temperatures at Tecan and Townsite locations.

PAGE 45

45 Figur e 3 2 Genotype leaf area index (LAI) in first ratoon crop at Tecan (A) and Townsite (B). Figure 3 3 Crop treatment effect means for juice concentration (A), Brix (B), and moisture concentration (C) in the plant cane and first ratoon crops at Tecan. Different letters represent significant differences among crop means ( P < 0.05).

PAGE 46

46 Figure 3 4 Crop means for fresh yield (A) and dry yield (B) in the plant cane and first ratoon crop s at Tecan. Different letters represent significant differences among crop means ( P < 0.05). Figure 3 5 Crop means for juice concentration (A), Brix (B), and m oisture concentration (C) in the plant cane and first rat oon crop at T ownsite. Different letters represent significant differences between crop means ( P < 0.05).

PAGE 47

47 Figure 3 6 Crop by genotype interaction means for Brix in plant cane and first ratoon crops at Townsite. Figure 3 7 Crop effect means for fresh yield (A) and dry yield (B) in the plant cane and first ratoon crops at Townsite. Different letters represent significant differences among crop means ( P < 0.05).

PAGE 48

48 Figure 3 8 Crop by site interaction means for juice concentration (A), Brix (B), and moisture concentration (C) in the plant cane and first ratoon crops at Tecan and Townsite. Figure 3 9 Crop x site interaction means for fresh yield (A) and dry yield (B) in the plant cane and first ratoon crops at Tecan and Townsite.

PAGE 49

49 CHAPTER 4 FIBER CONCENTRATION, CALORIFIC VALUE, AND ASH CONCENTRATION OF BIOENERGY GRASS SPECIES GROWN ON SANDY SOILS OF FLORIDA The production of ethanol from lignocellulosic biomass is a promising alternative source of energy (USDOE, 2006) with a comparative advantage over sources like corn and sugarcane due to the abundance of lignocellulose compared to grain and sucrose combin ed and the divers ity of raw material s that can be used as feedstocks. However, lignocellulosic ethanol requires greater processing to extract fermentable sugars from feedstock. Perennial grasses, such as elephantgrass, energycane and g iant reed have been identified as potential bioenergy feedstocks. Some author s suggest that these species can be grown o n marginal lands thus reducing land compet ition with food crops ( Field et al., 2008 ) while causing less harm to the environment by reducing greenhouse gas ( GHG ) emissions up to 88% (Farrell et al., 2006). Biomass feedstocks are primarily composed of cellulose, hemicellulose and lignin. The six and five carbon carbohydrate components from cellulose and hemicellulose are fermented into ethanol, whereas the li gnin fraction is no t ferment ed Lignin is thus a byproduct in the cellulosic ethanol process and may be used for direct combustion (Huang et al 2009) to produce steam, heat energy and electricity. L ignocellulosic biomass composition may change depending on climatic conditions and within season plant growth patterns (Naik et al., 2010). Thus careful characterization of biomass feedstocks is necessary, since the chemical composition of plant cell wall will affect biomass conversion efficiency For example high nitrogen, moisture and ash concentration from feedstock reduces the efficiency of thermo chemical conversion (Sanderson et al., 1996). High m oisture concentration of feedstock s can negatively affect the combustion quality of biomass, because it caus es ignition problems and reduc es combustion temperatures (Demirbas, 2004). Biomass heating value, also

PAGE 50

50 called calorific value, can be defined as the higher heating value (HHV), which is the energy co ncentration of biomass on a dry basis when combusted (Kha n et al., 2009). The quantification of cell polymers is also important to both lignocellulosic ethanol and direct combustion bioenergy production systems In the lignocellulosic conversion process, m ost of the fermentable sugars are from cellulose and hemi cellulose in plant cell wall s and lignin limits the availability of these sugars for enzymatic breakdown ( Chang and Holtzapple, 2000; Laureano Perez et al., 2005) However, in thermo chemical conversion, feedstocks with greater lignin concentration are de sirable, because lignin has greater heating value than cellulose and hemicellulose (Demirbas, 2005) H eating value is reduced proportionally with increasing ash con centration in biomass (McKendry, 2002). Many of the ash forming elements in biomass, such a s Si, Al, Fe, Ca, Mg, Na, K, S, and P (Khan et al., 2009) can form slag in boiler tubes and result in the increase of maintenance and operational costs (McKendry, 2002; Khan et al., 2009). Quantification of feedstock compositio n, heating value and ash con centration enables calculations of theoretical yields for conversion processes, assists with economic analys i s, and provides selection criteria for plant breeders. The objectives of this study were: 1) to quantify and estimate concentration of cellulose, h emicellulose, and lignin from plant biomass of giant reed elephantgrass and energycane ; and 2) to quantify HHV and ash con centration of these genotypes grown on marginal lands in Florida. Materials and Methods The experimental design, planting and harvest were performed as r eported in Chapter 3. Two days prior to final biomass yield measurements at Tecan (plant cane 15 Dec 200 8; and first ratoon 13 Dec 2009), and at Townsite (plant cane 10 Mar 2009 ; and first ratoon 8 Mar 2010), ten stalks were col lected at random from the two center rows of each plot. These ten stalk

PAGE 51

51 samples were chopped in a modular sugarcane disintegrator (Codistil S/A Denini, Mod: 132S, Piracicaba SP, Brazil). All chopped plant material (including leaf, stem and tops) from each sample was mixed in an individual 200 L barrel. From ea ch of the plot samples, two sub samples were collected. Subsample 1, composed of 800 to 1000 g fresh weight biomass was used estimate fiber concentration Subsample 2, compose d of 500 to 700 g of fresh weight biomass was used to estimate fiber component concentration s and measure higher heating value and ash con centration. Measurements and Analysis For fiber concentration S ubsample 1 fresh weight was rec orded and then sample s were placed in a hydraulic press (Codistil S/A Dedini, Mod: D 2500 II, Piracicaba SP, Brazil), for 30 sec until pressure attained 3000 psi. The resultant filter cake press sub sample was weighed and the estimated fiber concentration of the subsample was calculated using Eq. 1. Est imated f iber concentration = (Initial subsample 1 weight Filter cake press weight) / (Initial subsample 1 weight) x 100 Eq. 1 Subsample 2 was placed in a paper bag (Duro Bag Mfg., #25, Florence, KY), fresh weight was recorded using a balance wi th weighing capacit y up to 8100 g x 0.1 g (Metter Toledo, Mod: PB 8001 S), and the sub sample was dried at 60C until constant weight to determine dry biomass concentration. After drying, subsamples were ground t o pass a 1 mm screen in a Wiley Mill, standar d model #3 (Thomas Scientific, Philadelphia, PA). These ground samples were stored in whirlpak bags ( 300 g ) at room temperature. Approximately 40 g of ground sample s w ere sent to a commercial laboratory (Dairy One Lab., Ithaca, NY) for wet chemistry anal ysis to calculate acid detergent fiber (ADF), neutral detergent fiber (NDF), and acid detergent lignin (ADL), following the method of Van Soest et al. (1991).

PAGE 52

52 Simple subtraction rules are used to estimate cellulose and hemicellulose concentration from ADF, NDF and ADL (Mandre, 2005 ; Hodgson et al., 2010 ). ADF ADL = cellulose Eq. 2 N DF ADF = hemicellulose Eq. 3 Cellulose and hemicellulose were calculated using these equations and reported both as a concentr ation and in Mg ha 1 on a total dry weight basis. Lignin was determined using the ADL values directly (Van Soest et al 19 91 ) and reported as concentration on a total dry weight basis and in total Mg ha 1 Total d ry biomass yields from each fiber component was calculated using the total dry biomass yields (chapter 3) and multiplied by the concentration of each fiber component The HHV from the ground dry biomass material was measured at a commercial laboratory (Dairy One, Ithaca, NY) using an IKA C2000 basi c Calorimeter System. The instrument was set of 25 C. Analysis time was between 7 and 12 min. Dried samples were weighed into polyethylene bags. Samples were then placed in a crucible, and ignited in an oxygen rich atmosphere in a sealed decomposition vessel where the increase in temperature of the system was measured. The specific HHV of the sample was calculated from the weight of the sample, the heat capacity of the calorimeter system determined fro m benzoic acid calibration standards, and the increase in temperature of the water within the inner vessel of the measuring cell and expressed in MJ kg 1 on a dry biomass weight basis For determination of ash con centration from plant biomass, the laborat ory analytical procedure NREL/TP 510 42622 (Sluiter et al., 200 8 ) was used. Results were expressed as the concentration of residue remaining after dry oxidatio n at 575 25 C ( l oss on i gnition) All results were reported relative to the 105 C oven dry we ight of the sample. The procedure

PAGE 53

53 consists of weighing a sample of 0.5 to 2.0 g to the nearest 0.1 mg. This sample was placed in a porcelain ashing crucible to heat the biomass sample in a muffle furnace (Thermolyne, Mod: F A1740, Dubuque, IA, USA) with a ramping program (furnace temperature ramp from room temperature to 105 C hold at 105 C for 12 min., ramping to 250 C at 10 C /min., holding at 250 C for 30 min., ramping to 575 C at 20 C /min., holding at 575 for 180 min., allowing temperature to dr op to 105 C and holding furnace temperature at 105 C until samples are removed) for 24 6 h. Samples were then placed into a desiccator and cooled until constant weight A sh con centration was expressed as g kg 1 Statistical Analys i s Yield traits assoc iated with lignocellulosic ethanol production ( cellulose, hemicellulose and lignin concentration and dry weight yields ) and direct combustion (HHV and ash concentration ) were analyzed using analysis of variance with Proc GLM in SAS with crop genotype si difference (LSD) test was used for comparisons of means in all statistical analys i s. Three separate analys i s were performed and are presented separately by genotypes g rown at the 1) Tecan location ( 875 3; US 74 1010; US 78 1011; US 78 1014; US 78 1013; US 82 1655; US 84 1047; US 84 1066 ; giant reed; Merkeron; and Chinese Cross) 2) Townsite location ( 875 3; US 74 1010; US 78 1011; US 78 1014; US 78 1013; US 82 1655; US 84 1047; US 84 1066 ; and giant reed) and 3) Combined analysis ( 875 3; US 74 1010; US 78 1011; US 78 1014; US 78 1013; US 82 1655; US 84 1047; US 84 1066 ; and giant reed) for common set of genotypes grown at both Tecan and Townsite as in Chapter 3

PAGE 54

54 Results Tecan A nalysis At Tecan, crop and genotype effects were significant for cellulose, hemicellulose and lignin concentration and yield w ith the exception of the crop effect on lignin yield (Table 4 1). The crop x genotype interaction was significant only fo r hemicellulose concentration Figure 4 1 presents the crop effect means for cellulose, hemicellulose, and lignin concentration (A) and dry yield (B) for plant and first ratoon crops at Tecan. There was a greater cellulose and lignin concentration in first ratoon than plant cane, but higher concentration hemicellulose in plant cane than first ratoon. Dry yields of cellulose were lower in the plant cane than the first ratoon crop, but hemicellulose yields were higher in the plant cane than in first ratoon. L ignin yields were not different between crops. Giant reed produced significant ly lower cellulose, hemicellulose and lignin yields than energycane and elephantgrass (Table 4 2). Merkeron elephantgrass had the highest yields for all three traits. Elephantgr ass genotypes Merkeron and Chinese Cross had higher cellulose and lignin concentrations than most energycane genotypes, with giant reed a lso having higher lignin concentration than the energycanes. The f iber components concentration s and yields were simila r among energycane genotypes. Crop had a significant effect on ash concentration and HHV value s whereas genotype did not have a significant effect on these traits. The genotype x crop interaction effect was only significant for HHV (Table 4 3 ). The crop x genotype interaction for hemicellulose concentration and HHV at Tecan (Table 4 4 ). An increase in hemicellulose occurred for L 79 1002, US 78 1011, US 78 1014, US 84 1047, and US 84 1066 from plant cane to the first ratoon crop, whereas all other genotypes decrease d in hemicellulose concentration between plant cane and first ratoon crops. Increases in

PAGE 55

55 HHV were noted from plant cane to first ratoon for all genotypes excep t Merkeron and Chinese Cross elephantgrass. Figure 4 2 presents the crop treatment effe ct means for ash concentration and HHV H igher heating value was greater in first ratoon (20.1 MJ kg 1 ) than in plant cane (19.6 MJ kg 1 ) However, ash concentration was lower at first ratoon ( 219 g kg 1 ) than in plant cane ( 313 g kg 1 ). Townsite A nalysis At the Townsite location crop had a significant effect on cellulose and lignin concentration and cellulose, hemicellulose and lignin yields (Table 4 5 ). The genotype effect was significant for cellulose, hemicellulose and lignin yields and hemicellulose c oncentration. Crop x genotype interaction effects were significant only for cellulose concentration. Figure 4 3 presents the crop effect means for cellulose, hemicellulose, and lignin concentration (A) and dry yield (B) for plant and first ratoon crops at Townsite. Cellulose and lignin concentration was greater in the first ratoon than the plant cane crop, whereas hemicellulose concentration was not different between crops. Dry biomass yields were lower in plant cane than the first ratoon crop for all fiber components. Giant reed yielded less cellulose, hemicellulose and lignin yields than the energycane genotypes (Table 4 6 ) due to lower total biomass rather than lower fiber component concentrations. Giant reed had higher hemicellulose concentration than t he energycane genotypes, with the exception of US 82 1655. In general the energycane genotypes recorded similar fiber concentrations and yields Figure 4 4 presents the crop x genotype interaction effect means for cellulose concentration at Townsite. Firs t ratoon crop cellulose concentrations were greater than the plant cane crop for all genotypes, however the differences were smaller for giant reed than the energycane genotypes.

PAGE 56

56 The crop effect was significant for a sh concentration and higher heating valu es at Townsite, whereas genotype and crop x genotype interaction did not have a significant effect on these traits (Table 4 7 ). Figure 4 5 presents the crop means for ash concentration (A) and HHV (B) at Townsite. Ash concentration was greater in plant ca ne than the first ratoon crop, but HHV were greater in the first ratoon than the plant cane crop. Combined A nalysis Crop, site, and genotype had significant effects on all fiber concentrations and yields, with the exception of the crop effect on hemicellul ose concentration and the effect of genotype on cellulose concentration (Table 4 8 ). The crop x site interaction effect was significant for all traits with the exception of hemicellulose concentration. The crop x genotype interaction effect was significant for hemicellulose concentration the site x genotype interaction had a significant effect on hemicellulose yield, and the crop x site x genotype interaction was significant for cellulose concentration. Figure 4 6 presents the site by crop interaction effe ct means for cellulose, hemicellulose, and lignin concentration (A) and total yield (B) for plant cane and first ratoon crops. Cellulose concentration and yield differences between crops were greater at Townsite than Tecan. Similarly, differences in lignin concentration and yield between crops were greater at Townsite than Tecan. Differing trends in hemicellulose yields between crops were noted at Townsite and Tecan with plant cane hemicellulose yields greater than first ratoon at Tecan but first ratoon yie lds greater than plant cane at Townsite. The c rop by genotype interaction was significant for hemicellulose yield (Table 4 9 ). Hemicellulose yields decreased between crops for giant reed, 875 3, US 78 1013, US 78 1014,

PAGE 57

57 and US 82 1655 energycane, but increa sed for US 74 1010 US 78 1011 US 84 1047 and US 84 1066 energycane. Hemicellulose yields were markedly lower at Townsite than Tecan for 875 3, US 78 1011, US 78 1013, and US 84 1047 energycane genotypes (Table 4 1 0 ). Giant reed and US 74 1010 recorded s imilar hemicellulose yields between sites. The 3 way g enotype x site x crop interaction had a significant effect on cellulose concentration ( Table 4 1 1 ) For g iant reed cellulose concentration differences between crops were larger at Tecan (9.9 %) than To wnsite (5.7 %), but for all the energycane genotypes cellulose concentration differences between crops were larger at Townsite than Tecan. Crop and site effects and the crop x site interaction were significant for ash concentration and HHV (Table 4 12 ). Va rietal effects and their interactions were not significant for these traits. There was a crop x site interaction for ash concentration (A) and HHV (B) for plant cane and first ratoon crops at the Tecan and Townsite locations ( Figure 4 7 ) Ash concentration was greater in plant cane than first ratoon at both locations, but the difference between crops was greater at Tecan. Tecan and Townsite recorded similar HHV (B) across crops. Discussion Th is study is the first to our knowledge to characterize the cell wa ll components from energycane genotypes, g iant reed and Merkeron and Chinese Cross elephantgrass. Our results showed that mean cellulose, hemicellulose, and lignin con centration (mean of plant cane and first ratoon crops) were 40 0 27 3 and 8 1 g kg 1 resp ectively across species. McKendry (2002) reported fiber concentration means for switchgrass, ranging from 30 0 to 50 0 10 0 to 40 0 and 5 0 to 25 0 g kg 1 for cellulose, hemicellulose, and lignin, respectively. Our data on energycane, giant reed and elephant grass fall within the same range. Hodgson et al. (2010) working with Miscanthus genotypes in Europe, reported cellulose, hemicellulose, and lignin concentrations

PAGE 58

58 ranging from 412 to 539, 235 to 338, and 76 to 115 g kg 1 respectively. Their data are simil ar to those for g iant reed elephantgrasses and energycanes in this study, for which cellulose, hemicellulose, and lignin concentrations ranged from 395 to 444, 252 to 305, and 65 to 112 g kg 1 respectively, from two locations. According to Ingram et al. (1999), fiber concentration from grass species vary between 20 0 to 50 0 g kg 1 cellulose, 20 0 to 4 0 0 g kg 1 hemicellulose, and 10 0 to 20 0 g kg 1 lignin The average lignin concentration from the selected species in this experiment was lower than means obse rved by Ingram et al. (1999). The lignin concentration of giant reed, Merkeron and Chinese Cross elephantgrass (96, 93, and 113 g kg 1 respectively) was higher and moisture concentration lower (471, 600, and 614 g kg 1 respectively) than the energycanes with overall average lignin of 77 g kg 1 and moisture concentration of 615g kg 1 Genotypes with greater lignin concentration, such as Merkeron elephantgrass and g iant reed could cause an inhibitor y effect on enzymatic hydrolysis, making the fermentation of cellulose and hemicellulose less efficient. For the production of ethanol, a biomass feedstock with high cellulose/hemicellulose concentration is needed to provide a high amount of ethanol per metric ton of biomass. To illustrate the effect of cellulos e concentration o n yield, up to 280 l/t of ethanol can be produced from switchgrass, compared with 205 l/t from wood, largely due to the increase d proportion of lignin in wood (McKendry, 2002). The energycane cellulose, hemicellulose, and lignin concentr at ion reported in our study (397, 272, and 7 7 g kg 1 respectively) is considered adequate for ethanol production via microbial digestion (McKendry, 2002). Jung and Lamb (2004) reported that hemicellulose and cellulose concentrations are overestimated and l ignin underestimated using the detergent fiber system and ADL which was used to calculate the concentrations of cellulose and hemicellulose (Eq. 1 and Eq. 2) However,

PAGE 59

59 this method is the predominant method used by agronomist s and animal scientist s to measu re forage quality (Jung and Lamb, 2004) and is a much less expensive method (approximately $30/sample) to estimate fiber concentration in bioenergy crops than NREL methods (approximately $500/sample). NIR calibrations with NREL wet chemistry methods may r educe the cost and processing time for bioenergy fiber analys i s. Cellulose and lignin con centration from all genotypes at both locations were higher in first ratoon than plant cane. To our knowledge this is the first report of changes in fiber composition across crops. This information indicates that the calorific value across crops may increase since lignin is positively correlated to HHV The overall HHV of plant cane and first ratoon crops for the species in the present study were 19.2 and 19.9 MJ kg 1 which is similar to overall values reported by Naik et al. (2010) for firewood biomass (19.6 MJ kg 1 ) and by Hu et al. (2010) for switchgrass (18.8 MJ kg 1 ). However, Mantineo et al (2009) found averages of 16.4 MJ kg 1 for g iant reed and M. x giganteus and 15.9 MJ kg 1 in C cardunculus that were lower than the values found for the grass species in this study. The major component of ash is silica. Ash con centration has been linked to silica (Si) con centration in plant tissue (Crocker et al., 1998) whic h is also linked to soil clay con centration (Sander, 1997). High silica con centration of feedstock can combine to cause fouling and slagging of combustion systems when temperature exceeds the melting point of ash. The development of low ash agricultural fe edstock (< 30 g kg 1 ash) could significantly expand biofuel markets (Samson and Mehdi, 1998). Results from our soils analysis, show a high difference on the available Si between locations, Tecan (1.1 mg kg 1 ) and Townsite (113.1 mg kg 1 ). Plants grow n in soils with high Si con centration recorded high ash con centration In the first ratoon crop,

PAGE 60

60 Tecan with low available Si in the soil had ash con centration of 24.6 g kg 1 and Townsite, with high available Si in the soil had ash con centration of 31.6 g kg 1 The low overall ash con centration observed in the present experiment is due to the high sand concentration linked to low Si uptake. Thus the low ash con centration observed in this study may be due to the soil characteristics (sandy soils). Based on the li gnocellulosic characteristics all the genotypes tested in this study with exception of g iant reed are promising biomass candidates to be grown under commercial management practices in sandy soils in south Florida. The cellulose, hemicellulose, and ligni n concentration s were 40 0, 273, and 81 g kg 1 respectively. Direct combustion ash concentration appears to be related to available Si in the soil. Merkeron elephantgrass and g iant reed had lower moisture and high er lignin concentration s than the energycan e genotypes. Those species appear to be more appropriate for direct combustion than the energycanes whereas the energycanes would be better suited for cellulosic ethanol production

PAGE 61

61 Table 4 1 Analysis of variance F ratios and level of significan ce for cellulose, hemicellulose, and lignin concentration (conc.) and yield for crop and genotype effects and their interaction at Tecan. Treatment Cellulose Hemicellulose Lignin Conc. Yield Conc. Yield Conc. Yield Crop (C) 202.1 *** 4.1 5. 5 4.4 8.0 ** 0.4 Genotype (G) 3.2 ** 11.4 *** 4.6 *** 12.3 *** 7.1 *** 7.9 *** C x G 1.4 0.9 2.8 ** 1.0 1.9 1.7 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively. Table 4 2 Cellul ose, hemicellulose, and lignin concentration s and yield for g iant reed energycane and elephantgrass genotypes at Tecan. Genotype Cellulose Hemicellulose Lignin g kg 1 Mg ha 1 g kg 1 Mg ha 1 g kg 1 Mg ha 1 Giant reed 41 3 cd 2.2 c 305 a 1.6 f 9 6b 0.5 c M erkeron 44 5a 11. 7a 296 abc 7. 9a 9 3bc 2. 5a L 79 1002 4 20bcd 7. 8b 28 3bcd 5. 3cde 7 2d 1.3 b 875 3 400 d 8.7 b 30 6a 6. 7b 80cd 1.7 b US 74 1010 416 bcd 7.4 b 275 d 4.9 e 78 cd 1.4 b US 78 1011 400 d 8.2 b 289 abcd 5.9 bcde 70 d 1.4 b US 78 1013 408 cd 9. 1b 27 8cd 6. 2bcde 69 d 1.5 b US 78 1014 39 6d 7.4 b 293 abcd 5.5 bcde 7 7d 1.4 b US 82 1655 41 9bcd 8.6 b 301 ab 6 2 bcd 68 d 1.4 b US 84 1047 432 abc 9.0 b 301 ab 6.3 bc 6 5d 1.3 b US 84 1066 417 bcd 8.6 b 29 4abcd 6. 1bcde 7 9cd 1.6 b Chinese Cross 44 1ab 9.1 b 252 e 5. 1de 11 3a 2.3 b LSD 0.05 24 6 1.82 20 4 1.20 14 8 0.50 Treatment means followed by the same letter within a column are not significantly different ( P

PAGE 62

62 Table 4 3 Analysis of variance F ratios and level of significance for ash concentration and HHV value for crop and genotype effects and their interaction at the Tecan location. Treatment Ash Concentration H HV Crop (C) 53.8 *** 46.0 *** Geno type (G) 1.0 0.8 C x G 1.4 2.1 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively. Table 4 4 Crop x genotype interaction means for hemicellulose and HHV of giant reed, energycane and elephantgrass genotypes at the Tecan location. Genotype Hemicellulose HHV _____ g kg 1 _____ _______ MJ kg 1 _______ Crop PC 1R PC 1R Giant reed 340 27 1 19. 9 20.2 Merkeron 310 282 20. 1 20.0 L 79 1002 280 28 5 19. 6 20.2 875 3 319 292 19.7 20. 1 US 74 1010 275 275 19. 5 20. 3 US 78 1011 28 4 294 19. 4 20.0 US 78 1013 285 270 19. 8 20.0 US 78 1014 28 7 300 19. 5 20.3 US 82 1655 3 10 293 19. 5 20.3 US 84 1047 29. 3 30.9 19.3 20.1 US 84 1066 29. 4 29. 4 19. 5 20. 3 Chinese Cross 25.5 2 5.0 20. 2 20.0 PC = plant cane; 1R = first ratoon.

PAGE 63

63 Table 4 5 Analysis of variance F ratios and level of significance for cellulose, hemicellulose, and lignin concentration (conc.) and yield for crop and genotype eff ects and their interaction at Townsite. Treatment Cellulose Hemicellulose Lignin Conc. Yield Conc. Yield Conc. Yield Crop (C) 400.3 *** 182.1 *** 0.0 65.9 *** 273.0 *** 263.0 *** Genotype (G) 0.5 12.5 *** 2.4 12.5 *** 1.9 8.1 *** C x G 2. 6 1.6 1.6 1.2 1.3 2.0 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively. Table 4 6 Genotype effect for cellulose, hemicellulose, and lignin concentration s (conc.) and yield for g iant reed and e nergycane genotypes at Townsite. Genotype Cellulose Hemicellulose Lignin Conc. Yield Conc. Yield Conc. Yield g kg 1 Mg ha 1 g kg 1 Mg ha 1 g kg 1 Mg ha 1 Giant reed 385 5 2.39 b 283 2 a 1.58 c 95 2 0.65 c 875 3 383 8 7.14 a 247 8 bc 4.55 ab 81 8 1.54 ab U S 74 1010 372 3 6.89 a 258 3 bc 4.68 ab 82 0 1.53 ab US 78 1011 380 2 7.60 a 243 0 c 4.75 ab 80 2 1.62 ab US 78 1013 390 2 7.83 a 253 0 bc 4.96 ab 82 8 1.69 ab US 78 1014 383 3 7.29 a 254 8 bc 4.79 ab 82 8 1.63 ab US 82 1655 375 8 7.80 a 266 2 ab 5.42 a 79 0 1.66 ab US 8 4 1047 385 8 7.19 a 248 7 bc 4.36 b 76 0 1.47 b US 84 1066 384 2 8.18 a 250 0 bc 5.14 ab 85 8 1.86 a LSD 0.05 23 7 ns 1.42 22 4 0.92 11 1 ns 0.35 Treatment means followed by the same letter are not significantly different ( P ns = not significant at P= 0.05 Table 4 7 Analysis of variance F ratios and level of significance for ash concentration and HHV for crop and genotype effects and their interaction at the Townsite location. Treatment Ash Concentration HHV Crop (C) 10.1 *** 123.5 *** Genotype (G ) 1.3 0.8 C x G 0.6 1.4 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively.

PAGE 64

64 Table 4 8 Analysis of variance F ratios and lev el of significan ce for cellulose, hemicellulose, and lignin concentration (conc.) and total yield for crop, site, and genotype effects and their interaction. Treatment Cellulose Hemicellulose Lignin Conc. Yield Conc. Yield Conc. Yield Crop (C) 561.6 *** 142.1 *** 1.5 19.6 *** 152.6 *** 148.6 *** Genotype (G) 1.1 31.5 *** 2.9 ** 31.2 *** 4.5 ** 13.8 *** Site (S) 54.4 *** 11.3 ** 104.8 *** 42.4 *** 10.2 ** 5.0 C x G 1.0 1.3 3.8 *** 0.9 0.9 1.3 C x S 15.5 *** 55.7 ** 2.1 44.7 *** 44.3 *** 81.5 *** G x S 1.2 0.9 1.6 2.2 0.9 0.5 C x G x S 2.2 0.9 0.8 0.6 1.1 1.8 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively. Table 4 9 Crop x genotype interaction means for hemicellulose yield. Genotype Hemicellulose __________ Mg ha 1 __________ Crop PC 1R Giant reed 32.25 26.60 875 3 27.88 27.45 US 74 1010 26.63 26.73 US 78 1011 25.77 27.43 US 78 1013 27.13 25.9 3 US 78 1014 36.87 27.93 US 82 1655 28.75 28.00 US 84 1047 27.28 27.68 US 84 1066 26.82 27.57 PC = plant cane; 1R = first ratoon.

PAGE 65

65 Table 4 1 0 Genotype x site interaction means for hemicellulose yields of giant reed and energycane genotypes. Genotype Hemicellulose ________ Mg ha 1 ________ Tecan Townsite Giant reed 1.61 1.58 875 3 6.65 4.55 US 74 1010 4.92 4.68 US 78 1011 5.90 4.75 US 78 1013 6.15 4.96 US 78 1014 5.50 4.79 US 82 1655 6.03 5.42 US 84 1047 6.31 4.36 US 84 1066 6.0 8 5.14

PAGE 66

66 Table 4 1 1 G enotype x site x crop interaction means for cellulose concentration of giant reed and energycane genotypes Genotype Site Crop Cellulose g kg 1 Giant reed TC PC 363 1R 462 TS PC 354 1R 411 875 3 TC PC 354 1R 446 TS PC 317 1R 450 US 74 1010 TC PC 387 1R 445 TS PC 310 1R 435 US 78 1011 TC PC 365 1R 435 TS PC 312 1R 439 US 78 1013 TC PC 378 1R 445 TS PC 335 1R 445 US 78 1014 TC PC 359 1R 433 TS PC 341 1R 425 US 82 16 55 TC PC 377 1R 461 TS PC 321 1R 430 US 84 1047 TC PC 392 1R 472 TS PC 331 1R 441 US 84 1066 TC PC 377 1R 457 TS PC 325 1R 443 TC= Tecan; TS= Townsite PC = plant cane; 1R = first ratoon.

PAGE 67

67 Table 4 12 F ratios and level of significance for ash concentration and HHV for crop, site, and genotype effects and their interaction. Treatment Ash Concentration HHV Crop (C) 49.3 *** 180.1 *** Genotype (G) 0.8 1.1 Site (S) 10.4 ** 173.5 *** C x G 0.6 1.5 C x S 8.0 ** 4.9 G x S 0.5 0.4 C x G x S 0.4 1.0 *, **, *** Significant at the < 0.05, 0.01, and 0.001 levels of probability respectively. Figure 4 1 Crop means for c ellulose, hemicellulose, and lignin concentration (A) and dry yield (B) at Tecan. Different letters represent significant differences among crop means ( P < 0.05).

PAGE 68

68 Figure 4 2 Crop treatment means for ash concentration and higher heating value at Tecan. Different letters represent significant differences between crop means ( P < 0.05). Figure 4 3 Crop effect means fo r c ellulose, hemicellulose, and lignin concentration (A) and dry yield (B) at Tecan. Different letters represent significant differences between crop means ( P < 0.05).

PAGE 69

69 Figure 4 4 Crop x genotype interaction means for cellulose concentration in the plant cane and first ratoon crops at Townsite.

PAGE 70

70 Figure 4 5 Crop treatment effect means for ash concentration (A) and HHV (B) in plant cane and first ratoon crop at the Townsite location Different letters represent significant differences between crop means ( P < 0.05). Figure 4 6 Crop x site interaction effect means for cellulose, hemicellulose, and lignin concentration (A) and total yield (B) in t he plant cane and first ratoon crops at Tecan (TC) and Townsite (TS)

PAGE 71

71 Figure 4 7 Crop x site interaction effect means for ash concentration (A) and HHV (B).

PAGE 72

72 CHAPTER 5 SUMMARY AND CONCLUSI ONS Our results indicate that energycanes and elephantgrasses are more appropriate bioenergy feedstocks than g iant reed for marginal sandy soils of south Florida. Rainfall during the establishment period of growth is essential for high biomass yields on sandy soils. Fres h and dry yields from new energycane genotypes were similar to L 79 1002 energycane Based on the lignocellulosic characteristics all the genotypes tested in this study, with exception of g iant reed are promising biomass candidates to be grown under comm ercial management practices in sandy soils in south Florida. The cellulose, hemicellulose, and lignin concentration s averaged 400, 273, and 8 1 g kg 1 respectively (overall average). Merkeron and g iant reed had high er lignin concentration than energycane g enotypes. Those species are more appropriate for combustion than the energycanes. The feedstock ash concentration appears to be related to available Si in the soil. Future Research Needs Because energy crops are more likely to be grown on marginal lands, s uch as the sandy soils used for this study, and expected to be grown under minimal inputs, the rat o oning ability and nutrient uptake of perennial grasses under these stress conditions should be further addressed in order to select bioenergy feedstocks for upcoming commercial scale biofuel plants in Florida.

PAGE 73

73 LIST OF REFERENCES Andrade, A.C., D.M. da Fonseca, R. dos S. Lopes, D. do Nascimento Jnior, P.R. Cecon, D.S. Queiroz, D.H. Pereira, and S.T. Reis. 2005. fert ilized and irrigated. Revista Cinc Agrotc 29: 415 423. Angelini, L.G., L. Ceccarini, and E. Bonari. 2005. Biomass yield and energy balance of giant reed ( Arundo donax L.) cropped in central Italy as related to different management practices. Eur. J. Ag ron. 22:375 389. Angelini, L.G., L. Ceccarini, N. Nassi o Di Nasso and E. Bonari. 2009. Comparison of A rundo donax L. and M iscanthus x giganteus in a long term field experiment in central Italy: Analysis of productive characteristics and energy balance. Biomass Bioenergy 33:635 643. A s pinall, G.O. 1980. Chemistry of cell wall polysaccharides. p.473 500 In J. Preiss (ed.) The biochemistry of plants. A comprehensive structure and function. Academic Press New York Atkinson, C.J. 2009. Establishing perenn ial grass energy crops in the UK: A review of current propagation options for miscanthus. Biomass Bioenergy 33:752 759. Barney, J.N., and J.M. DiTomasso. 2008. Non native species and bioenergy: Are we cultivating the next invader? Bio s cience. 58:64 70. B ischoff, K.P., K.A. Gravois, T.E. Reagan, J.W. Hoy, C.A. Kimbe n g, C.M. LaBorde, and G.L. Hawkins. 2008. Registration of L 79 1002Sugarcane. J. of Plant Registrations 2:211 217. Bouton, J.H. 200 2 Bioenergy crop breeding and production research in the sou theast. Final report for 1996 to 2001. Technical Report. ORNL/SUB 02 19XSV810C/O1 [Online] Available at http://www.ornl.gov/info/reports/2002/3445605360123.pdf (verified 21 Nov 2010). Bransby, D.I. 2008. Synchronization of biofeedstock and conversion t echnologies: Current status and future prospects. NABC 20:123 124. Burner, D.M., T.L. Tew, J.J. Harvey and D.P. Belesky. 2009. Dry matter partitioning and quality of M iscanthus P anicum and S accharum genotypes in A rkansas, USA. Biomass Bioenergy 33:610 619. Burton, G.W. 1989. Registration of 'Merkeron' napiergrass. Crop. Sci. 29:1327. Burton, G.W. 1993. African grasses. p. 294 298. In : J. Janick and J.E. Simon (ed.) New Crops. Wiley, New York. Cardinale, B.J., J.P. Wright, M.W. Cadotte, I.T. Carroll, A. Hector, D.S. Srivastava, M. Loreau and J.J. Weis. 2007. Impacts of plant diversity on biomass production increase through time because of species complementarity. Proc. Natl. Acad. Sci. U. S. A. 104:18123 18128.

PAGE 74

74 Chang, V.S. and M.T. Holtzapple. 2000. F undamental factors affecting biomass enzymatic reactivity. Applied Biochemistry and Biotechnology 84:5 37. Ch r istou, M. 2001. Giant ree d in Europe. p. 2089 2091. Proceedings of the First World Conference on Biomass for Energy and Industry. Sevil l a, Spain. evaluation of Arundo donax L. clones collected in southern I taly. Industrial Crops and Products 23:212 222. Crocker, L.M., E.J. Depeters, J.G. Fadel, S.E. Esse x, H. Perez Monti and S.J. Taylor. 1998. Ash content of detergent fibers in feeds, digesta, and feces and its relevance in fiber digestibility calculations. J. Dairy Sci. 81:1010 1014. Daniels, J., and B.T. Roach. 1987. Taxonomy and evolution. p. 7 84. I n D.J. Heinz (ed ) Sugarcane improvement through breeding. Elsevier, New York. Demirbas, A. 2004. Combustion characteristics of different biomass fuels. Progress in Energy and Combustion Science 30:219 230. Demirbas, A. 2005. Potential applications of ren ewable energy sources, biomass combustion problems in boiler power systems and combustion related environmental issues. Progress in Energy and Combustion Science 31:171 192. Deren, C.W., G.H. Snyder, P.Y.P. Tai, C.E. Turick and D.P. Chynoweth. 1991. Biom ass production and biochemical methane potential of seasonally flooded inter generic and inter specific S accharum hybrids. Bioresour. Technol. 36:179 184. Dien, S.B., H.J.G. Jung K.P. Vogel M.D. Casler, J.F.S. Lamb L Iten, R.B. Mitchell, and G. Sarath 2006. Chemical composition and response to dilute acid pretreatment and enzymatic saccharificat i on of alfalfa, reed canarygass, and switchgrass. Biomass and Bioenergy 30: 880 891. EIA. 20 09 U.S. Dep ar t ment of Energy information administration Office of Coal, Nuclear, Electric and Alternate Fuels [Online] Available at http://eia.gov/cneaf/alternate/page/renew_energy_consump/pretrends09.pdf (verified 15 Nov 2010). W ashington, DC. Farrell, A.E., R.J. Plevin, B.T. Turner, A.D. Jones, M. O'Hare and D.M. Kammen. 2006. Ethanol can contribute to energy and environmental goals. Science 311:506 508. Fengel, D., and G. Weneger. 1984. Wood: chemistry, ultra structure, and rea ctions. W. de Gruyter, New York. Field, C.B., J.E. Campbell, and D.B. Lobell. 2008. Biomass energy: the scale of potential resource. Trends Ecol. Evol. 23:65 72.

PAGE 75

75 Giamalva, M.J., S.J. Clarke and J.M. Stein. 1984. Sugarcane hybrids of biomass. Biomass 6:61 68. Gilbert, R.A., C.R. Rainbolt, D.R. Morris and J.M. McCray. 2008. Sugarcane growth and yield responses to a 3 month summer flood. Agric. Water Manage. 95:283 291. Gilbert, R.A., J. A. Ferrell, and Z.R. Helsel. 2008. Production of giant reedgrass for biofuel. SS AGR 318 [Online]. Available at http://edis.ifas.ufl.edu/ag327 (verified at 3 Mar. 2011). Inst. of Food and Agric. Sci., Cooperative Extension Service, IFAS, Gainesville, FL. Gordon, D.R., A.M. F ox and R.K. Stocker. 2006. Testing a predictive s creening tool for reducing the introduction of invasive plants in Florida. Final Report to the USDA APHIS/PPQ. The Nature Conserva n cy, VA, Univ ersity of Florida, Gainesville FL Gordon, D.R., K.J. Tancig, D.A. Onderdonk, and C.A. Gantz. 2011. Assessing th e invasive potential of species proposed for Florida and the United States using the Australian Weed Risk Assessment. Biomass Bioenergy 35:74 79. Gross, R., M. Leach and A. Bauen. 2003. Progress in renewable energy. Environ. Int. 29:105 122. Hanna, W.W. C.J. Chaparro, B.W. Mathews, J.C. Burns, L.E. Sollenberger and J.R. Carpenter. 2004. Perennial P ennisetums p. 503 535. In L.E Moser et al. (ed.) Warm season (C 4 ) grasses monograph. ASA, CSSA, and SSSA, Madison, WI. Hartmann, H.B. 2001. Energie aus bio masse. p. 248 271. In M. Kaltschimitt and H. Hartmann (ed.) Springer, New York. Herrera, A. and T.L. Dudley. 2003. Reduction of riparian arthropod abundance and diversity as a consequence of giant reed ( A rundo donax ) invasion. Biological Invasions 5:167 177. Hodgson, E.M., S.J. Lister, A.V. Bridgwater, J. Clifton Brown and I.S. Donnison. 2010. Genotypic and environmentally derived variation in the cell wall composition of miscanthus in relation to its use as a biomass feedstock. Biomass Bioenergy 34:65 2 660. Hu, Z., R. Stukes, M.F. Davis, E.C. Brummer, and A.J. Ragauskas. 2010. Chemical profiles of switchgrass. Biosource Technology. 101:3243 3257. Huang, H., S. Ramaswamy, W. Al Dajani, U. Tschirner and R.A. Cairncross. 2009. Effect of biomass species and plant size on cellulosic ethanol: A comparative process and economic analysis. Biomass Bioenergy 33:234 246. Ingram, L.O., H.C. Aldrich, A.C.C. Borges, T.B Causey, A. Martinez, F. Morales, A. Saleh, S.A. Underwood, L.P. Yomano, S.W. York, J. Zaldivar, and S. Zhou. 1999. Enteric bacterial catalysts for fuel ethanol production. Biotechnol. Prog. 15:855 866.

PAGE 76

76 IPCC. 2001. Climate change 2001: mitigation. Contribution to the third assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge United Kingdom Jakob, K., F. Zhou and A.H. Paterson. 2009. Genetic improvement of C4 grasses as cellulosic biofuel feedstock. In Vitro Cell. Dev. Biol. Plant. 45:291 305. Jung, H.G ., and J.F.S. Lamb. 2004. Prediction of cell wall polysaccharide and lignin concentrations of alfalfa stems from detergent fiber analysis Biomass Bioenergy 27:365 373. Kaygusuz, K. and M.F. Trker. 2002. Biomass energy potential in T urkey. Renewable Energy 26:661 678. Khan, A.A., W. de Jong, P.J. Jansens and H. Spliethoff. 2009. Biomass combustion in fluidized bed boilers: Potential problems and remedies. Fuel Process Technol 90:21 50. Kim, S. and B.E. Dale. 2004. Global potential bioethanol production from wasted crops and crop residues. Biom ass Bioenergy 26:361 375. Laureano Perez, L., F. Teymouri, H. Alizadeh and B.E. Dale. 2005. Understanding factors that limit enzymatic hydrolysis of biomass: Characterization of pretreated corn stover. p. 1081 1100. In Understanding factors that limit en zymatic hydrolysis of biomass: Characterization of pretreated corn stover. 26 th S ymposium on biotechnology for fuels and chemicals, Chanttanooga, TN. Legendre, B.L. and K.A. Gravois. 2007. The 2006 Louisiana sugarcane variety survey. Sugar Bull. 85 : 23 27. Lewandowski, I., J.M.O. Scurlock, E. Lindvall and M. Christou. 2003. The development and current status of perennial rhizomatous grasses as energy crops in the US and Europe. Biomass Bioenergy 25:335 361. Lin, Y. and S. Tanaka. 2006. Ethanol fermentati on from biomass resources: current state and prospects. Appl. Microbiol. Biotechnol. 69:627 642. Littell, R.C., W.W. Stroup, and R.J. Freund. 2002. SAS Publishing. SAS for linear models. 4th ed. SAS Inst., Cary, NC. Long, S.P. 2008. Opportunities for enhan cing the productivity of biofeedstocks and minimizing inputs: theory and practice. In A. Eaglesham, S.A. Slack, and R.W.F. Hardy (ed.) Reshaping American agriculture to meet its biofuel and polymer roles. NABC 20:1 09 1 18 Lynd, L.R., J.H. Cushman, R.J. Nic hols and C.E. Wyman. 1991. Fuel ethanol from cellulosic biomass. Science 251:1318 1323. Lynd L R M.S. Laser, D. Bransby, B.E. Dale, B. Davison, and R. Hamilton 2008 How biotech can transform biofuels. Nat Biotechnol 26:169 172

PAGE 77

77 Ma n dre, M. 2005. Res ponses of Norway spruce ( Picea abies L.) to wood ash application. Forest S tudies 42:34 47. Mangelsdorf, A.J. 1946. Sugarcane breeding in Hawaii, Part I. 1778 50:141 160. Cosentino. 2009. Biomass yield and energy balance of three perennial crops for energy use in the semi arid M editerranean environment. Field Crops Res. 114:204 213. McKendry, P. 2002. Energy production from biomass (part 1): Overview of biomass. Bioresou r. Technol. 83:37 46. McLaughlin, S.B. and L.A. Kszos. 2005. Development of switchgrass ( P anicum virgatum ) as a bioenergy feedstock in the United States. Biomass Bioenergy 28:515 535. Morais, R.F., B.J. Sousa, J.M. Leite, L.H.B. Soares, B.J.R. Alves, R. M Boddy, and S. Urquiaga. 2009. Elephantgrass genotypes for bioenergy production by direct biomass combustion. Pesq. A gropec. B ras. 44:133 140. Naik, S., V.V. Goud, P.K. Rout, K. Jacobson and A.K. Dalai. 2010. Characterization of Canadian biomass for alt ernative renewable biofuel. Renewable Energy 35:1624 1631. Overman, A.R. and K.R. Woodard. 2006. Simulation of biomass partitioning and production in elephantgrass. Comunic. Soil Sci. Plant Analysis. 37:1999 2010. Parikka, M. 2004. Global biomass fuel r esources. Biomass Bioenergy 27:613 620. Perdue, R.E. 1958. Arundo donax source of musical reeds and industrial cellulose. Econ. Bot. 12:368 404. Prine, G.M., P. Mislevy, R.L. Stanley, L.S. Dunavin Jr. and D.I. Bransby. 1991. Field production of energy cane elephantgrass and sorghum in southe astern United States. p. 1 12. In D.L. Klass (ed.) Proc. F inal P rogram of C onference on E nergy from B iomass and W aste XV. Inst. Gas Tech., Washington, DC. Prine, G.M., J.A. Stricker and W.V. McConnell. 1997. Oppor tunities for bioenergy development in L ower S outh USA. p. 227 235. In R.P. Overend and E. Chornet (ed.) Opportunities for bioenergy development in lower south USA. Proc. 3 rd C onference of America: Making a business from biomass in energy, environment, chem icals, fibers and materials Elsevier Science, Oxford, United Kingdom. Rainbolt, C. 2005. Napiergrass: biology and control in sugarcane.SS AGR 242 [Online]. Available at http://edis.ifas.ufl.edu/sc071 (verifie d at 11 Nov. 2010). Inst. of Food and Agric. Sci., Cooperative Extension Service, IFAS, Gainesville, FL. Rossa, B., A.V. Tuffers, G. Naidoo and D.J. von Willert, D.J. 1998. Arundo donax L. (poaceae) a C3 species with unusually high photosynthetic capacity Botanica Acta 111:216 221.

PAGE 78

78 Rubin, E.M. 2008. Genomics of cellulosic biofuels. Nature 454: 841 845. Samson, R. and B. Me hdi 1998. Strategies to reduce the ash content of perennial grasses. Proc. 4 8 Oct. 1 998, Madison, WI Sander, B. 1997. Properties of Danish biofuels and the requirements for power production. Biomass Bioenergy 12:177 183. Sanderson, M.A., F. Agblevor, M. Collins and D.K. Johnson. 1996. Compositional analysis of biomass feedstocks by nea r infrared reflectance spectroscopy. Biomass Bioenergy 11:365 370. Sanford, M.T. 2003. Florida's climate and its beekeeping. Fact Sheet ENY 134 [Online] Available at http://edis.ifas.ufl.edu/aa264 (verified 11 Dec. 2010). Inst. of Food and Agric. Sci., C ooperative Extension Service, IFAS, Gainesville, FL. Schell, D.J., C.J. Riley and N. Dowe. 2004. A bioethanol process unit: Initial operating experiences and results with a corn fiber feedstock. Bioresource Technology 91:179 188. Sluiter, A., B. Hames, R. Ruiz, C. Scarlata, J. Sluiter and D. Templeton. 200 8 Determination of ash in biomass. Technical Report. NREL/TP 510 42622. National Renewable Energy Laboratory. G olden, Colorado. Sollenberger, L.E., G.M. Prine, W.R. Ocumpaugh, S.C. Schank, R.S. Kalmbacher, and C.S. Jones, Jr. 1987. Dwarf elephantgrass: A high quality forage with potential in Florida and the tropics. Soil Crop Sci. Soc. Fla. Proc. 46:42 46. Stevens on, G.C. 1965. Genetics and breeding of sugarcane. Longmans, Green and Co Ltd., London. Sticklen, M. B. 2006. Plant genetic engineering to improve biomass characteristics for biofuels. Current Opinion in Biotechnology 17:315 319. Sticklen, M.B. 2008. Plant genetic engineering for biofuel production: towards affordable cellulosic ethanol. Nature Reviews Genetics 9:433 443. Stricker, J.A., G.M. Prine and K.R. Woodard. 1993. Biomass yield of tall grass energy crops on phosphatic clay in central Florida. Proc. Soil and Crop Science Society of Florida 52:4 6. Tew, T.L. and R.M. Cobill. 2008. Genetic improvement of sugarcane ( Saccharum spp.) as an energy crop. p. 249 272. In W. Vermerris (ed.) Genetic improvement of bioenergy crops. Springer New York. Thompson J.B. 1919. Napier and merker grasses. Univ. Fla. Agric. Exp. Stn. Bull. 153, Gainesville, FL.

PAGE 79

79 USDA. 20 09 National agricultural statistics service. State crop progress and condition [Online] Available at http://www.nass.usda.gov/Statistics_by_State (verified 15 Mar 2010). USDA NASS, Washington, DC. USDA. 2010. Biofuels strategic production report [Online] Available at http://www.usda.gov/documents/USDA_Biofuels_Report_6232010.pdf (verified 11 Nov 2010). USDA, Washington, DC. USDOE. (United States Department of Energy). 2006. Breaking the biological barriers to cellulosic ethanol: A joint research agenda. DOE/SC 0 095. U.S. Department of Energy Office of Energy Efficiency and Renewable Energy [Online] Available at http://genomicscience.energy.gov/biofuels/2005workshop/b2blowres 63006.pdf ( verified 12 Mar 2010) Van Soest, P.J., J.B. Robertson and B.A. Lewis. 1991. Methods for dietary fiber, neutral detergent fiber, and non starch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74:3583 3597. Wang, L.P., P.A. Jac kson, L. Xin, Yuan Hong, F., Foreman, J.W., Xue Kuan C., D. Hai Hua, F. Cheng, M. Li and K.S. Aiken. 2008. Evaluation of sugarcane X Saccharum spontaneum progeny for biomass composition and yield components. Crop Science 48:951 961. Wiselogel, A., J. Tyso n, and D. Johonsson. 1996. Biomass feedstock resources and its composition. p.105 118. In C.E. Wyman (ed.) Handbook of bioethanol: production and utilization. Washington, DC. Woodard, K.R. and G.M. Prine. 1993. Dry matter accumulation of elephantgrass, en ergycane, and elephantmillet in a subtropical climate. Crop Sci. 33:818 824. Woodard, K.R., G.M. Prine, D.B. Bates and D.P. Chynoweth. 1991. Preserving elephantgrass and energycane biomass as silage for energy. Bioresource Technology 36:253 259.

PAGE 80

80 BIO GRAPHICAL SKETCH Pedro H. Kornd rfer was born in 1984 in Porto Alegre, RS, Brazil. Pedro developed his interests in agriculture from a family business producing sugarcane rum and by following in his a professor in plant nutrition He e arn ed his B.S. degree in Agronomy in 2008 from Gois State University Ipameri, G O Brazil. Pedro began his experience at University of Florida through short term scholar programs in the fall of 2005, 2007 and 2008, participating in sugarcane, weed science and bioenergy crops research programs with Drs. Gilbert, Rainbol t and Helsel respectively. He joined the Agronomy Department at the University of Florida in spring 2009, pursuing his graduate studies under the supervision of Dr. Robert A. Gilbert and co