1 Article

2 Effect of Moisture Content on Lignocellulosic Power

3 Generation: Energy, Economic and Environmental
4 Impacts
5 Karthik Rajendran 1,2, *
6 1 MaREI Centre, Environmental Research Institute, University College Cork, Cork, Ireland
7 2 School of Engineering, University College Cork, Cork, Ireland
8 * Correspondence: Karthik.1988@gmail.com; Karthik.Rajendran@ucc.ie; Tel.: +353-83-034-4162

9 Academic Editor: name

10 Received: date; Accepted: date; Published: date

11 Abstract:

12 Moisture content in the biomass affects its thermal processing for further application such as
13 electricity or steam. The effect of variation in moisture content of two different biomass including
14 Banagrass and Energy cane was evaluated using techno-economic and lifecycle assessments. The
15 variation considered was a 25% change in the loss of moisture from the original moisture content of
16 the biomass before field drying. Techno-economic analysis revealed that high moisture of both the
17 biomass wasn’t economically feasible. Comparing Banagrass with Energycane, the latter was more
18 economically feasible that would be mainly attributed to the less moisture and ash content in
19 Energycane. About 32 GWh/year of electricity could be produced by field drying 60,000 dry
20 MT/year Energy cane. The investments for different scenarios ranged between $17-22 Million while
21 the only field dried Energycane could recover the investment after 11 years of operation. This
22 scenario was also the environmental friendly that released 16 g CO2 eq./MJ of electricity produced.

23 Keywords: techno-economic analysis; lifecycle assessments; thermochemical processes;

24 lignocelluloses; energy analysis;
25

26 1. Introduction

27 Climate models from IPCC predict the global surface temperature will increase between 0.3 and
28 4.8 °C in the 21st century [1]. Between 1880 and 2012, mean surface temperature increased by 0.85 °C
29 that is alarming and needs to be controlled. Human interventions including usage of fossil fuels for
30 energy and transportation, emission from agricultural practices, industrial developments etc. were
31 reported to be the major contributors to the climate change [1]. About 40% of the global warming
32 emissions could be reduced when the demands of energy supply and transportation sectors were met
33 [1]. The quest for alternative energy sources includes wind, solar, geothermal, tidal, hydroelectric and
34 bioenergy. Wind and solar have certain problems including the produced electricity are difficult to
35 use for other applications, storing the produced energy and potential losses during the transmission
36 on conversion.

37 Islands like Hawaii is deprived of resources for power generation where more than 70% of
38 electricity is produced from imported oil and 13% from coal as a primary energy precursor [2]. The
39 mainland US generates only 1% of its electricity from oil as a source shows the dependency of fossil
40 resources in these islands. Currently, biomass-based electricity accounts for 3% of the electricity
41 generated in Hawaii [3]. Lignocelluloses are plentiful organics accessible today accounting up to 50
Processes 2017, 5, x; doi: FOR PEER REVIEW www.mdpi.com/journal/processes
 Processes 2017, 5, x FOR PEER REVIEW 2 of 15

42 billion tons in dry weight [4]. Of the available 1.3 Million acres’ land owned by the State of Hawaii
43 for agriculture, only 8% is used for cultivation while the rest is used for gazing [5,6]. This land acts as
44 a potential source for the cultivation of energy crops and other lignocelluloses that could help in the
45 production of green energy. Producing green-energy reduces the dependence on oil, economic
46 burden for the state and reduce the environmental impacts by producing electricity from a clean
47 source.

48 Lignocellulosic ethanol is one of the possible alternatives for clean energy that requires a
49 pretreatment process [7]. The pretreatment step increases the production cost and also has a
50 technological hindrance in the form of effective sugar release and inhibitor formation. Some attempts
51 were also made on integrating first and second generation of ethanol production earlier that reduces
52 the production cost [8]. Unlike biochemical processes, thermochemical processes offer access to clean
53 energy with less technological hindrances. Different thermochemical routes include gasification,
54 combustion, pyrolysis, hydrothermal carbonization, hydrothermal liquefaction etc. [9]. Combustion
55 is a thermochemical process where an exothermic redox reaction happens between the fuel and an
56 oxidant that produces steam and other gases. These hot gases could be used for potential applications
57 for heat and power using a turbine [10].

58 The disadvantage with the thermochemical process is that the biomass needs to be free of excess
59 moisture and failing to do so consumes energy to vaporize the water present in reducing the
60 efficiency of the process. For the same, only dried biomasses such as wood chips are usually preferred
61 in thermochemical processes. Sometimes, the energy crops are also field dried before it is used in the
62 plant reducing the moisture content [11]. The variation in the moisture content affects the overall
63 viability of the process. No previous research considered the variation of moisture content on trifold
64 sustainability perspective including techno-economic and environmental impact analysis. This tri-
65 fold sustainability metrics check the feasibility of a process in multi-disciplinary angle ensuring the
66 process is not just environmentally friendly but also economically viable.
67 The objective of this study include:
68  Analyze the effect of varying moisture content of two different lignocelluloses
69 (Banagrass and Energy cane) for electricity production.
70  Evaluate the techno-economic potentials of power production in Maui (Hawaii).
71  Carry out an energy analysis to understand energy flow due to the moisture in
72 lignocelluloses.
73  Conduct a lifecycle assessment to identify the environmental impacts of the
74 thermochemical process.

75 2. Methods

76 2.1. Description

77 The plant site is foreseen to be built on Maui, one of the islands in Hawaii that mainly uses diesel
78 as a source for electricity production. Using diesel as an electricity source in contrasting with the
79 mainland USA shows that the emissions rating will be different. The change in emissions will affect
80 the global warming potential and environmental impacts. The plant was designed to process 60,000
81 dry MT/year that was calculated based on the land availability in Maui island where a pilot-scale
82 field trial was conducted. This pilot trials estimated different aspects of growing crops including crop
83 yield, crop rotation, nutrient requirement etc. One could argue that 60,000 dry MT/year is a small
84 capacity in comparison with the mainland USA however, for an island the land use has a significant
85 effect and hence the plant has to designed based on the availability.

86 In this study, four different scenarios were considered using two different biomasses. For
87 Banagrass “BG” was used as a code name while for Energycane it was “EC”. The variability of
 Processes 2017, 5, x FOR PEER REVIEW 3 of 15

88 moisture content by drying them on the land was estimated for electricity production where high
89 moisture implies the biomass was transported to the site upon harvest, that has a code name of “HM”.
90 When the biomasses were left on the field for the excess moisture to be dried and then transported to
91 the site that has a low moisture, “LM” was abbreviated. For example, BGLM refers to the scenario of
92 Banagrass with a high-moisture content. In total, four scenarios were estimated for electricity
93 production with two biomass under two different moisture content. When the biomass was left on
94 the field, about 25% of the moisture was expected to be lost for both the biomass from their wet
95 weight (Table 1).

96 2.2. Feedstock
97 Two different feedstocks were considered in this study including Banagrass and Energycane.
98 NREL procedures were followed to find the composition of Banagrass while the composition of
99 Energy cane was based on Kim and Dale [12]. The biomass yield, emissions, nutrient requirements,
100 water, agricultural machinery requirements etc. for both the biomass was obtained from the field
101 trials conducted on Maui. Table 1 shows the composition of both the biomasses in wet and dry basis.
102 Harvest rate of the biomasses was 15 MT/year that had a collection efficiency of 90%.
103
104 Table 1. The composition of Banagrass and Energycane on a wet and dry basis.
Banagrass Energy cane
105 Composition
Wet Basis Dry Basis Wet Basis Dry Basis
106 Cellulose 10.22% 37.48% 10.05% 33.44%
Hemicellulose 6.39% 23.43% 6.36% 21.16%
107
Lignin 4.49% 16.46% 3.78% 12.58%
108 Extractives 3.56% 13.05% 7.92% 26.36%
Ash 2.61% 9.57% 1.94% 6.46%
109 Moisture 72.73% - 69.95% -

110

111 2.3. Model development

112 Intelligen Superpro Designer (V-10.0) was used to simulate the process that calculates the mass
113 and energy balance. Based on the mass and energy balance economic analysis were performed. The
114 biomass harvested were transported using trucks to the plant site. When the biomass reaches the
115 plant site, the biomass is unloaded and stored in a silo using a conveyor before it was further
116 processed. The silo has a storage capacity to hold the biomass for 10 days, as a reserve in case of any
117 delays in transportation. The stored biomass is shredded using a knife mill that uses energy at the
118 rate of 0.09 kW/(kg/h). For field dried biomass with low moisture, 0.15 kW(kg/h) energy was used
119 [13]. This was due to the reason that wet biomass consumes less energy to shred while dry biomass
120 requires high energy. The shredded biomass is sent to the boiler for steam generation. Figure 1 shows
121 the process schematics from Superpro Designer. The developed models were attached as electronic
122 supplementary to facilitate transparency and reproducibility.

123 The boiler was operated at 257°C and 4.5MPa that used 10% excess oxygen from air depending
124 on the incoming biomass [14]. The excess oxygen was adjusted with the flow to provide complete
125 combustion of the material. The flu gas exists the boiler at 200°C where the overall heat loss was
126 assumed to be 5% [15]. Elemental composition of individual components within the lignocelluloses
127 was mentioned. The available heat in the system varied depending on the moisture content of the
128 biomass as less heat was required to combust when the biomass had a lower moisture content. Ash
129 exists the boiler at 400°C. The steam from the boiler is sent to a turbine where the steam is expanded
130 to produce electricity. The cooled steam after the gas expansion is condensed and reused to the
 Processes 2017, 5, x FOR PEER REVIEW 4 of 15

131 process, and about 5% of the water was considered as losses or waste that was not recycled. The
132 amount of feed water was auto adjusted depending on the throughput of the biomass in each
133 scenario. Some of the electricity produced was utilised for the plant and only the excess electricity
134 was sold and the internally used electricity was credited to avoid confusions in the economic
135 calculations.

Feed Water Recycled Water

Energy Cane
Flue Gas Waste Water
P-10 S-101 S-102

Conveyors P-2
P-4 S-105
P-31 S-103 Flow Splitting
S-104
Shredding
Storage

P-5
P-3
Gas Expansion
Combustion
Air
Ash
136
137 Figure 1. Schematics of the flowsheet developed in Superpro Designer.

138 2.4. Economic analysis

139 2.4.1. Assumptions

140 The plant had a lifetime of 15 years with a construction period of 2.5 years and startup period of
141 4-months. The construction period was higher as most of the construction materials should be barged
142 from the mainland. The annual operation time of the plant was 330 days, while the remaining 35 days
143 was used for plant maintenance and other purposes. The sizing and costing of the different
144 equipment’s were based on the Superpro Designer. The interest rate was assumed to be 9% while the
145 annual inflation was presumed at 4%. Table 2 shows the detailed assumptions that were used to carry
146 out this study. Biomass was purchased at the cost of $80/dry MT as the price of biomass is constantly
147 increasing a higher value was considered [16]. Electricity was sold at a cost of $0.27/kWh that was
148 obtained based on the data from Hawaiian Electric. The straight-line depreciation method was used
149 and the depreciation was assumed to be for 10 years.

150 2.4.2. Uncertainty analysis

151 The uncertainty analysis depicts the vulnerability of a process and it is essential to understand
152 the robustness. For the same, different uncertainty analysis were carried out including the fluctuation
153 in capacity and the minimum electricity selling price (MESP) for a zero net present value (NPV). In
154 the base scenarios, the capacity of the plant was 60,000 dry MT/year. For the uncertainty analysis, the
155 capacity varied between 30,000 and 360,000 dry MT/year. Similarly, the selling price of electricity was
156 altered to find MESP at which no profit or loss for the process was obtained.

157 2.5. Lifecycle assessments

158 2.5.1. Goal, scope and boundary definition

159 The main goal of this work is to carry out the LCA was to develop a well-to-pump life cycle
160 inventory and assess the environmental impacts of electricity production from two biomasses in
161 Hawaii region. The scope included comparing environmental impacts of electricity production from
162 lignocelluloses under high moisture and low moisture content that replaces the conventional
163 electricity production in Hawaii. 1MJ was used as a functional unit to carry out the assessment. Figure
164 2 shows the system boundary that included both techno-economic and LCA boundary. Agricultural
 Processes 2017, 5, x FOR PEER REVIEW 5 of 15

165 inputs like machinery used, irrigation water, nutrient requirement, harvesting, transportation of the
166 biomass etc. were included in the system boundary when calculating the lifecycle assessments. Open
167 LCA (V 1.6.3) software was used to estimate the environmental impacts while eco-invent database
168 (V 3.1) was used to run the background connected processes integrating the inventory. TRACI 2.1
169 was used as an impact assessment method as this was the ISO preferred method to carry out the LCA
170 in USA [17,18].
171 Table 2. List of assumptions used in this study.
172
Type Assumption
173 Annual processing 60,000 dry

174 capacity MT/year

Biomass cost $80/ dry MT
175 Electricity cost $0.27/kW-h
Interest rate 9%
176
Annual operational hours 7920 h
177 Start-up time 4 months
Construction period 30 months
178
Income tax 40%
179 Inflation 4%
Project lifetime 15 years
180 Depreciation method Straight line

181 Salvage value 5%

Depreciation years 10 years
182

183
184 Figure 2. System boundary showing the techno-economic boundary and LCA boundary for the
185 lignocellulosic power production.
 Processes 2017, 5, x FOR PEER REVIEW 6 of 15

186 2.5.2. Lifecycle inventory

187 The outputs from the techno-economic analysis of the four scenarios were imported as inventory
188 that was used to carry out lifecycle assessments. The process model developed yielded inputs and
189 outputs such as raw material, water consumption, energy consumption and energy produced from
190 the process. The electricity from the plant was assumed to replace the electricity produced from the
191 Hawaii electricity grid. The data for biomass production including crop yields, irrigation and field
192 emissions were gathered based on the pilot field trials conducted in Maui, Hawaii. Eco-invent
193 database and the imported data from the techno-economic analyses was used to conduct LCA
194 analyses.

195 2.5.3. Lifecycle impact assessments

196 The lifecycle impact assessment method used in this study was Tool for the Reduction and
197 Assessment of Chemical and other environmental Impact (TRACI 2.1) that was developed by USEPA
198 [19]. TRACI contains ten different impact categories including acidification, ecotoxicity,
199 eutrophication, global warming, ozone depletion, photochemical ozone formation (POF), resource
200 depletion – fossil fuels, carcinogenics, non-carcinogenics and respiratory effects [17]. The inputs and
201 outputs from the different scenarios that were used to carry out the LCA were attached as an
202 electronic supplementary to facilitate reproducibility and increase transparency of the results.

203
204 Figure 2. Block flow diagram showing the mass balance for different scenarios. BG and EC refer to
205 Banagrass and Energycane, HM and LM refer to high moisture and low moisture respectively.
 Processes 2017, 5, x FOR PEER REVIEW 7 of 15

206 3. Results

207 3.1. Techno-economic analysis

208 The techno-economic analysis was carried out using Superpro Designer that considered the
209 variation in moisture content of two-different biomass including Banagrass and Energycane. 60,000
210 dry MT/year was considered as an annual processing capacity in the base case scenario. The change
211 in moisture content by field drying the biomass and their subsequent effect on the electricity
212 production was assessed. Figure 2 shows the block flow diagram on the overall mass balance of
213 different scenarios. The moisture content in the high moisture scenarios such as BGHM and ECHM
214 was 72.73% and 69.95% respectively. The moisture content in the BGLM and ECLM scenarios was
215 reduced to 25% in both the scenarios. Banagrass and Energycane as a feedstock released 94,700 and
216 98,000 MT carbon dioxide/year. The electricity production was highest when Energycane with low
217 moisture content (ECLM) at the rate of 32.6 GWh/year while the lowest electricity production was in
218 BGHM at 15.6 GWh/year (Figure 2). The electricity reported here were the net electricity produced
219 after consumption of different types of equipment.
220 Table 3. The purchase cost of different equipment’s used in different scenarios with their sizing
221 data.
222
High Moisture Low Moisture
BGHM ECHM BGLM ECLM
Banagrass Energycane Banagrass Energycane
Unit Amount Cost($) Amount Cost($) Amount Cost($) Amount Cost($)
Conveyor
ft. 400 458,000 400 458,000 400 458,000 400 458,000
belt
Shredder kg/h 27,780 459,000 25,210 433,000 15,872 328,000 13,773 302,000
Steam
kg/h 29,635 372,000 42,048 485,000 41,585 481,000 53,607 583,000
generator
Expansion
kW 2,591 590,000 3,791 900,000 3,747 892,000 4,922 1,092,000
turbine
Unlisted
470,000 569,000 540,000 610,000
equipment
Total cost
2,349,000 2,845,000 2,699,000 3,045,000
($)
223
224 Table 3 shows the cost of different equipment used in the four scenarios with their processing
225 throughputs. The shredder load decreased with the decrease in moisture content of the field dried
226 biomass. The cost of turbine ranged between $ 590,000 – 1,092,000 for a capacity of 2,591 – 5,922 kW.
227 The boiler had a throughput between 30,000 and 54,000 kg/h that costed $ 372,000 (BGHM) – $ 583,000
228 (ECLM). The total cost of the equipment for different scenarios ranged between $ 2.3 and 3.0 Million.
229 The equipment costs were a fraction of the total capital investments while the other costs include
230 installation, piping, instrumentation, insulation, electrical, building, auxiliary facilities, indirect costs,
231 contractor fee, and contingencies. Figure 3 shows the different economic indices including CAPEX,
232 OPEX and revenues generated from different scenarios. The capital investment ranged between $ 17-
233 22M for different scenarios. In general, a decrease in moisture content by field drying the biomass
234 increased the CAPEX that was due to the reason that higher capacity boiler and the turbine was used
235 as more power was produced.
236
 Processes 2017, 5, x FOR PEER REVIEW 8 of 15

CAPEX OPEX Revenues

$25

Millions
$22
$20
$17
$15 $10
$9
$5 $7
$10

$5

$-
BGHM ECHM BGLM ECLM

High Moisture Low Moisture

237
238 Figure 3. Economic indexes including CAPEX, OPEX and revenues generated from different
239 scenarios (Million USD).
240
241 The OPEX varied between $ 8.7 – 9.5 Million for all the scenarios, while the revenues meeting
242 the OPEX was only generated from ECLM scenario (Figure 3). The cost of the feedstock at $ 80/dry
243 MT accounted for ca. 50% of the OPEX for all scenarios (Figure 4). Raw material cost was the most
244 significant contributor for OPEX followed by facility dependent costs that ranged between 32 and
245 39%. This shows that the cost of producing biomass needs to be controlled for any biomass-based
246 energy/chemicals and failing to do so questions the economic viability of the project. Utilities, i.e., the
247 electricity consumed corresponds to 7 - 9%, while the wages for the labour was 3% of OPEX.
248
BGHM

ECHM Raw Materials

Labor
Facility
Laboratory
Utilities
ECLM

BGLM

249
250 Figure 4. Fragmentation of operational costs for different scenarios.
251
252 Every kWh of electricity sold, generated a unit revenue of $ 0.27, while the production cost for
253 the different scenarios including BGHM, ECHM, BGLM and ECLM was $ 0.47, $ 0.34, $ 0.34 and $0.27
254 respectively (Figure 5). The cut off production cost was only met in ECLM that generate the unit
255 revenue as same as the costs occurred in that scenario. Return on investment (ROI) that equates the
256 profitability index of a project that showed that only ECLM had a positive ROI at 9%, while the rest
257 of the scenarios yielded <0% meaning that drying Energycane in the field and producing electricity
258 out of it was the only scenario that was economically viable for Maui. However, drying the biomass
259 results the land use change and that effects need to evaluated in the future.
260
 Processes 2017, 5, x FOR PEER REVIEW 9 of 15

$0.50 10%
9%
$0.40 $0.47 5%

Cost ($/kWh)
$0.27

ROI (%)
$0.30 0%

$0.20 -5%

$0.10 -10%
-14%
$- -15%
BGHM ECHM BGLM ECLM

Production cost Unit Revenue ROI

261
262 Figure 5. Profitability indexes including production cost ($/kWh), unit revenue ($/kWh) and return
263 on investment (%) for the four different scenarios exploited in this study.
264
(a) (b)

(c) (d)

265 Figure 6. Energy analyses of different scenarios including consumption and production of
266 electricity. All the numbers here were presented as MWh.
 Processes 2017, 5, x FOR PEER REVIEW 10 of 15

267 3.2. Energy analysis

268 Figure 6 shows the energy analysis of different scenarios via Sankey diagram on the breakdown
269 of consumption and production pattern. For scenario BGHM, about 15% of the produced electricity
270 was consumed in different forms including a shredder, conveyors, general load and unlisted
271 equipment. Around 70% of the electricity consumed was dominated by the shredder, leaving a net
272 electricity of 15,653 MWh/year for sale. Similarly, for other scenarios, i.e., BGLM, ECHM and ECLM
273 the consumption was 9.5%, 10.1% and 6.8% respectively. The net electricity production for different
274 scenarios was 15,653 (BGHM), 24,433 (BGLM), 24,005 (ECHM) and 32,696 (ECLM) Mwh/year. It is
275 also worth to mention that LM scenarios had higher electricity consumption for shredder (0.15
276 kW/(kg/h)) that HM scenarios that had (0.09 kW/(kg/h). However, the low moisture had produced
277 more electricity in comparison with high moisture biomass.

278 3.3. Lifecycle assessments

279 The data from techno-economic assessments and field data from pilot trials were used for LCI,
280 while LCA was conducted using Open LCA and TRACI 2.1 was used as an impact assessment
281 method [20]. 1 MJ was used as a functional unit for all the scenarios. Seven environments related and
282 three health-related impacts were tabulated in Table 4 for different scenarios. Low moisture of
283 biomass resulted in lower environmental impacts for both the substrates in comparison with high
284 moisture content. This shows that field drying the biomass has a positive effect on the environmental
285 impacts, however, the effect of land use change needs to be assessed as the biomass will be on the
286 field for a longer period of time.

287 Acidification refers to the increase in hydrogen ions i.e., a number of acids into the environment
288 by carrying out this process. The acidification potential was measured in kg SO2 eq. Banagrass with
289 high moisture content scenario had the highest acidification potential of 3.0x10^E-3 while the lowest
290 was for ECLM (9.3x10^E-4). Global warming potential was measured in kg CO2 eq. For every 1 MJ
291 of electricity produced in Maui using the thermochemical process results in GWP between 1.6x10^E-
292 1 and 2.9x10^E-2 kg CO2 eq. The trend looked similar for other environmental impacts where
293 banagrass had higher environmental impacts in comparison with Energy cane while high moisture
294 content had higher impacts in comparison with low moisture scenarios.
295
296 Table 4. Environmental impacts of different scenarios.
297
High Moisture Low Moisture
Impact category Unit Banagrass Energycane Banagrass Energycane
BGHM ECHM BGLM ECLM
Acidification kg SO2 eq. 3.0x10^E-3 1.2x10^E-3 1.9x10^E-3 9.3x10^E-4
Eco-toxicity CTUe 1.2x10^E+0 5.5x10^E-1 8.4x10^E-1 4.1x10^E-1
Eutrophication kg N eq. 8.1x10^E-4 4.1x10^E-4 5.3x10^E-4 3.0x10^E-4
Global Warming kg CO2 eq. 2.9x10^E-2 2.2x10^E-2 1.9x10^E-2 1.6x10^E-2
Carcinogenics 1.3x10^E-8 9.7x10^E-9 8.8x10^E-9 7.2x10^E-9
Non-carcinogenics 5.6x10^E-8 2.2x10^E-8 3.6x10^E-8 1.6x10^E-8
Ozone Depletion Kg CFC-11 eq. 7.1x10^E-8 2.6x10^E-8 4.6x10^E-8 1.9x10^E-8
Photochemical
kg O3 eq. 2.6x10^E-2 2.8x10^E-2 1.7x10^E-2 2.1x10^E-2
ozone formation
Resource depletion MJ surplus 8.8x10^E-1 2.3x10^E-1 5.7x10^E-1 1.7x10^E-1
Respiratory effects kg PM2.5 eq. 4.1x10^E-4 1.8x10^E-4 2.7x10^E-4 1.4x10^E-4
 Processes 2017, 5, x FOR PEER REVIEW 11 of 15

298 4. Discussion

299 4.1. Uncertainty analysis

300 The base case scenarios considered an annual processing capacity of 60,000 dry MT/year, which
301 is relatively small considering the plant sizes in mainland US. For the same, an uncertainty analysis
302 was carried out in identifying the CAPEX of the process under different capacities. The capacities
303 considered were between 30,000 and 360,000 dry MT/year. Figure 7 shows the uncertainty analysis
304 on the CAPEX for different scenarios evaluated in the base case. As a rule of thumb, as the capacity
305 increases, the CAPEX should go down by the economies of scale principle. To process 1-MT of
306 biomass (BGLM), if the plant had a capacity of 30,000 dry MT/year, it costs $390/MT. When the same
307 scenario had a capacity of 360,000 dry MT/year the costs got reduced to $200/MT. Increasing the
308 capacity of the plant by 12 times, decreased the CAPEX by 95% on a functional unit level i.e., $/MT.
309 It is worth to mention that the base case scenario processing 60,000 dry MT/year costs about $320/MT
310 (BGLM).

$100
Millions

ECLM
$80 ECHM
Cost ($)

$60

$40
BGHM
$20 BGLM
$0
0 100 200 300 400
Thousands

Capacity (dry MT/year)

311
312 Figure 7. Uncertainty analyses of different scenarios under different capacities between 30,000 –
313 360,000 dry MT/year.
314
$0.70
MESP
$0.60 Production
cost
Unit revenue,
$0.50
$0.27
Cost ($/kWh)

$0.40

$0.30

$0.20

$0.10

$-
BGHM ECHM BGLM ECLM
315
316 Figure 8. Minimum electricity selling price estimation in relation to production cost and selling
317 price for different scenarios in $/kWh.
318
 Processes 2017, 5, x FOR PEER REVIEW 12 of 15

319 4.2. Minimum electricity selling price

320 The minimum electricity selling price refers to the minimum costs at which the electricity needs
321 to be sold for a zero NPV. MESP also refers to the levelized cost of electricity (LCOE) that means the
322 cost it occurs over the plant life to produce the total amount of energy during the same period. Figure
323 8 shows the MESP along with the production cost and unit revenue generated for each of the four
324 scenarios. The scenarios including BGLM, ECHM needs to be sold at $0.42/kWh to meet a zero NPV
325 while for BGHM, it should be $0.60/kWh. The most profitable scenario ECLM needs to sell at 6 cents
326 higher than the current unit revenue generated to have a zero NPV i.e., $0.27/kWh. ECLM could
327 recover the investments after 11 years (payback period). The LCOE of using different technologies
328 including photovoltaics, wind, geothermal and nuclear that provide electricity in Hawaii islands
329 ranged between $0.1 – 0.5/kWh [21]. ECLM from this study would be on a median in comparison
330 with the data reported showing that further reduction in biomass costs could favour the
331 commercialization of biomass-based power for an island like Hawaii.

332 4.3. Interrelationship with other commodities

333 Relating the electricity with other commodities/energy would be an interesting to interpret how
334 electricity as an energy form would influence other energy types. For the same, a correlation was
335 built between the historical energy/commodity and crude oil prices from 2007-2016. This decade-long
336 comparison shows that the electricity prices weren’t fluctuating with the crude oil prices, whereas
337 other fuels including ethanol, jet fuel were fluctuating correspondingly. Ethylene was the most
338 volatile commodity that was fluctuating with the crude oil prices. It is also worth to mention that
339 crude oil prices fluctuated in two regions: 1. <$63/Barrel and 2. >$95/Barrel.

Crude oil ($/Barrel)

31.8 63.6 95.3 127.1
1.60

Ethylene
1.20
Cost ($/Unit)

0.80
Ethanol

0.40 Electricity
Jet Fuel
0.00
0.00 0.20 0.40 0.60 0.80
Crude oil ($/L)
340
341 Figure 9. A historical comparison of the unit cost of electricity, jet fuel, ethanol and ethylene in
342 relation with crude oil from 2007-2016. The primary horizontal axis shows crude oil as $/L while
343 secondary horizontal axis shows the crude oil as $/Barrel to ease understanding.

344 4.4. Comparison with literature

345 This study reported the production cost of energy between $270–470/MWh. Patel, et al. [22]
346 compared the different thermochemical technologies and reported that the production costs varied
347 between $ 80-600/MWh depending on the feedstock and technologies employed. It is worth to
348 mention that this study also fits within this range. In addition, the cost of biomass used in this study
349 was $ 80/dry MT which needs to be noted for a fair comparison with other literature reported. LCA
350 of power generation from forest biomass had GHG emissions between 7-15 gCO2/MJ while this study
 Processes 2017, 5, x FOR PEER REVIEW 13 of 15

351 reported between 16-29 gCO2/MJ using energy crops such as Banagrass and Energycane as a
352 feedstock [23]. The variation in the GHG emissions could be based on the difference in feedstock used
353 that sequesters the different amount of carbon from the atmosphere.
354
355 A study reported between 2-4% of the produced energy was consumed for the internal purposes
356 [24], whereas in our work the consumption of produced energy ranged between 7–15% which is
357 higher than the other studies reported, however, considering the small plant sizes higher losses and
358 more energy utilization could be justified. This work earlier mentioned that higher moisture content
359 of biomass results in higher emissions which were also confirmed by other works [25,26]. The plant
360 size considered in this study was 60,000 dry MT/year that was depending on the land available at
361 Maui which increased the overall emissions as higher plant size reduce the GHG emissions.

362 5. Conclusions

363 Thermochemical combustion for electricity production was evaluated on a holistic approach
364 including techno-economic analysis and lifecycle assessments using Banagrass and Energycane as a
365 feedstock under varying moisture content. The results suggest that drying Energycane on the field
366 was the most sustainable scenario in-terms of technology, economics and environmental impacts.
367 The most profitable scenario (ECLM) could yield the investments after 11 years that had a capacity
368 of 60,000 dry MT/year. Biomass cost ($80/dry MT) was identified as one of the main factors that
369 determine the profitability of the plant after the electricity selling prices. The GHG emissions from
370 different scenarios ranged between 16 – 29 gCO2/MJ for electricity from bioenergy crops. For an island like
371 Hawaii energy generation from biomass could be sustainable in technology, economics and environmental
372 impacts in comparison with fossil sources.

373 Acronyms:

374 CAPEX – Capital costs

375 GWP – Global warming potential
376 LCA – Lifecycle assessments
377 LCI – Lifecycle inventory
378 LCOE – Levelized cost of electricity
379 MESP – Minimum electricity selling price
380 NPV – Net present value
381 OPEX – Operational costs
382 PBP – Payback period
383 ROI – Return on investment
384 TEA – Techno-economic analysis
385

386 Supplementary Materials: The following are available online at www.mdpi.com/link, File 1-4: Superpro files of
387 four scenarios used in this study that can be accessed using the demo version of the software from
388 http://www.intelligen.com/.

389 Acknowledgments: The authors would like to thank the pilot facility in Hawaii that helped in field trial
390 agricultural data to carry out the lifecycle assessments.

391 Author Contributions: KR was responsible for the idea, model development, critical analysis, manuscript
392 preparation and editing.

393 Conflicts of Interest: The author declares no conflict of interests.

394

395 References
 Processes 2017, 5, x FOR PEER REVIEW 14 of 15

396 1. Stocker, T. Climate change 2013: The physical science basis: Working group i contribution to the
397 fifth assessment report of the intergovernmental panel on climate change. Cambridge University
398 Press: 2014.
399 2. Department of Business Economic Development and Tourism. Monthly energy trends.
400 http://dbedt.hawaii.gov/economic/data_reports/energy-trends/ (07/10/2017),
401 3. US Energy Information Administration. Short-term energy outlook.
402 https://www.eia.gov/outlooks/steo/report/global_oil.cfm (07 Apr 2017),
403 4. Rajendran, K.; Taherzadeh, M.J. Pretreatment of lignocellulosic materials. In Bioprocessing of
404 renewable resources to commodity bioproducts, Bisaria, V.S.; Kondo, A., Eds. 2014; pp 43-75.
405 5. Surles, T.; Foley, M.; Turn, S.; Staackmann, M. A scenario for accelerated use of renewable
406 resources for transportation fuels in hawai ‘i. University of Hawaii, Hawaii Natural Energy
407 Institute, School of Ocean and Earth Science and Technology. Accessed November 2009.
408 6. Phillips, V.D.; Singh, D.; Merriam, R.A.; Khan, M.A. Land available for biomass crop
409 production in hawaii. Agricultural Systems 1993, 43, 1-17.
410 7. Rajendran, K.; Drielak, E.; Varma, V.S.; Muthusamy, S.; Kumar, G. Updates on the
411 pretreatment of lignocellulosic feedstocks for bioenergy production–a review. Biomass
412 Conversion and Biorefinery 2017, 1-13.
413 8. Rajendran, K.; Rajoli, S.; Taherzadeh, M.J. Techno-economic analysis of integrating first and
414 second-generation ethanol production using filamentous fungi: An industrial case study.
415 Energies 2016, 9, 359.
416 9. Shen, Y.; Jarboe, L.; Brown, R.; Wen, Z. A thermochemical–biochemical hybrid processing
417 of lignocellulosic biomass for producing fuels and chemicals. Biotechnol. Adv. 2015, 33, 1799-
418 1813.
419 10. Schmidt-Rohr, K. Why combustions are always exothermic, yielding about 418 kj per mole
420 of o2. J. Chem. Educ. 2015, 92, 2094-2099.
421 11. Khanchi, A.; Birrell, S. Drying models to estimate moisture change in switchgrass and corn
422 stover based on weather conditions and swath density. Agricultural and Forest Meteorology
423 2017, 237–238, 1-8.
424 12. Kim, M.; Day, D.F. Composition of sugar cane, energy cane, and sweet sorghum suitable
425 for ethanol production at louisiana sugar mills. J. Ind. Microbiol. Biotechnol. 2011, 38, 803-807.
426 13. Lee, J.H.; Kwon, J.H.; Kim, T.H.; Choi, W.I. Impact of planetary ball mills on corn stover
427 characteristics and enzymatic digestibility depending on grinding ball properties. Bioresour.
428 Technol. 2017, 241, 1094-1100.
429 14. Ali Mandegari, M.; Farzad, S.; Görgens, J.F. Economic and environmental assessment of
430 cellulosic ethanol production scenarios annexed to a typical sugar mill. Bioresour. Technol.
431 2017, 224.
432 15. Ali Mandegari, M.; Farzad, S. Görgens jf. In Biofuels: Production and future perspectives, Singh,
433 R.S.; Panday, A.; Gnansounou, E., Eds. CRC Press: Boca Raton, 2016.
434 16. Kadhum, H.J.; Rajendran, K.; Murthy, G.S. Effect of solids loading on ethanol production:
435 Experimental, economic and environmental analysis. Bioresour. Technol. 2017, 244, 108-116.
436 17. Bare, J.; Young, D.; QAM, S.; Hopton, M.; Chief, S. Tool for the reduction and assessment of
437 chemical and other environmental impacts (traci). US Environmental Protection Agency,
438 Washington, DC 2012.
 Processes 2017, 5, x FOR PEER REVIEW 15 of 15

439 18. ISO Technical Committee. Environmental management: Life cycle assessment: Requirements and
440 guidelines. ISO: 2006.
441 19. Bare, J.C. Traci. J Ind Ecol 2002, 6, 49-78.
442 20. Bare, J. Traci 2.0: The tool for the reduction and assessment of chemical and other
443 environmental impacts 2.0. Clean Technol. Envir. 2011, 13, 687-696.
444 21. Hawaii State Energy Office. Hawaii energy facts & figures; Hawaii, USA, 2014; pp 1-27.
445 22. Patel, M.; Zhang, X.; Kumar, A. Techno-economic and life cycle assessment on
446 lignocellulosic biomass thermochemical conversion technologies: A review. Renew. Sust.
447 Energ. Rev. 2016, 53, 1486-1499.
448 23. Thakur, A.; Canter, C.E.; Kumar, A. Life-cycle energy and emission analysis of power
449 generation from forest biomass. Appl. Energ. 2014, 128, 246-253.
450 24. Wihersaari, M. Greenhouse gas emissions from final harvest fuel chip production in
451 finland. Biomass Bioenerg. 2005, 28, 435-443.
452 25. Whittaker, C.; Mortimer, N.; Murphy, R.; Matthews, R. Energy and greenhouse gas balance
453 of the use of forest residues for bioenergy production in the uk. Biomass Bioenerg. 2011, 35,
454 4581-4594.
455 26. Angus-Hankin, C.; Stokes, B.; Twaddle, A. The transportation of fuelwood from forest to
456 facility. Biomass Bioenerg. 1995, 9, 191-203.

457

458 © 2017 by the authors. Submitted for possible open access publication under the
459 terms and conditions of the Creative Commons Attribution (CC BY) license
460 (http://creativecommons.org/licenses/by/4.0/).

461