Applied Energy 208 (2017) 495–510

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Techno-economic and greenhouse gas savings assessment of decentralized T

biomass gasification for electrifying the rural areas of Indonesia
⁎
Siming Youa, Huanhuan Tonga, Joel Armin-Hoilandb, Yen Wah Tonga,c, Chi-Hwa Wangc,
a
NUS Environmental Research Institute, National University of Singapore, 1 Create Way, Create Tower, #15-02, Singapore 138602, Singapore
b
All Power Labs, 1010 Murray Street, Berkeley, California, CA 94710, United States
c
Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117585, Singapore

H I G H L I G H T S

• The feasibility of decentralized gasification systems in Indonesia was explored.

• Two decentralized scenarios each were proposed for a village and palm oil mill, respectively.
• The global warming impact was evaluated using life cycle analysis (LCA).
• The economic feasibility was evaluated using cost-benefit analysis (CBA).
• The village and mill systems save up to 7700 g CO /kWh and 5.8 × 10 tons CO /year.
2-eq
4
2-eq

A R T I C L E I N F O A B S T R A C T

Keywords: This study explored the feasibility of decentralized gasification of oil palm biomass in Indonesia to relieve its
Biomass gasification over-dependence on fossil fuel-based power generation and facilitate the electrification of its rural areas. The
Biochar techno-feasibility of the gasification of oil palm biomass was first evaluated by reviewing existing literature.
Life cycle assessment Subsequently, two scenarios (V1 and V2, and M1 and M2) were proposed regarding the use cases of the village
Cost-benefit analysis
and mill, respectively. The capacity of the gasification systems in the V1 and M1 scenarios are determined by the
Decentralization
total amount of oil palm biomass available in the village and mill, respectively. The capacity of the gasification
systems in the V2 and M2 scenarios is determined by the respective electricity demand of the village and mill.
The global warming impact and economic feasibility (net present value (NPV) and levelized cost of electricity
(LCOE)) of the proposed systems were compared with that of the current practices (diesel generator for the
village use case and biomass boiler combustion for the mill use case) using life cycle assessment (LCA) and cost-
benefit analysis (CBA). Under the current daily demand per household (0.4 kWh), deploying the V2 system in
104 villages with 500 households each could save up to 17.9 thousand tons of CO2-eq per year compared to the
current diesel-based practice. If the electricity could be fed into the national grid, the M1 system with 100%
capacity factor could provide yearly GHG emissions mitigation of 5.8 × 104 ton CO2-eq, relative to the current
boiler combustion-based reference scenario. M1 had a positive mean NPV if the electricity could be fed into the
national grid, while M2 had a positive mean NPV at the biochar price of 500 USD/ton. Under the current
electricity tariff (ET) (0.11 kWh) and the biochar price of 2650 USD/ton, daily household demands of 2 and
1.8 kWh were required to reach the break-even point of the mean NPV for the V2 system for the cases of 300 and
500 households, respectively. The average LCOE of V2 is approximately one-fourth that of the reference sce-
nario, while the average LCOE of V1 is larger than that of the reference scenario. The average LCOE of M1
decreased to around 0.06 USD/kWh for the case of a 100% capacity factor. Sensitivity analysis showed that the
capital cost of gasification system and its overall electrical efficiency had the most significant effects on the NPV.
Finally, practical system deployment was discussed, with consideration of policy formulation and fiscal in-
centives.

⁎
Corresponding author.
E-mail address: [email protected] (C.-H. Wang).

http://dx.doi.org/10.1016/j.apenergy.2017.10.001
Received 28 June 2017; Received in revised form 8 September 2017; Accepted 1 October 2017
Available online 17 October 2017
0306-2619/ © 2017 Elsevier Ltd. All rights reserved.
S. You et al. Applied Energy 208 (2017) 495–510

1. Introduction considerable air pollutant emissions and limited energy efficiency.

Biomass gasification is an environmentally friendly alternative for
Bioenergy technologies have a great potential to help resolve urgent power generation from oil palm biomass. Prior studies (e.g., Ariffin
global challenges, such as lack of effective waste management and et al. [18], Atnaw et al. [19], Guangul et al. [20], Guangul et al. [21],
disposal, climate change, and energy and resource depletion. Extensive Ogi et al. [22]) have shown that oil palm biomass has a great potential
development of bioenergy systems has occurred in recent years, and a as feedstock for gasification for energy production. Moreover, gasifi-
wide variety of systems have been proposed, with the aim of sustain- cation is suitable for small-scale decentralized applications, which
ably processing various biomass types and benefitting users of different conforms well with the relatively small electricity demand for rural
social-environmental backgrounds. households [23,24]. Finally, biochar, a fixed carbon byproduct of the
For example, willow chips were converted into bioethanol via en- gasification process, can be used as a soil amendment to facilitate
zyme-catalyzed hydrolysis and fermentation, and electricity was gen- carbon sequestration and climate change mitigation [25].
erated using a biomass-fired integrated gasification combined cycle Environmental and economic evaluation must be conducted prior to
technology, both of which had more favorable environmental and en- the deployment of energy systems in order to consider the needs of
ergy performance than conventional fossil fuel-based energy sources various stakeholders, including policymakers, private investors, and
[1]. Crop residue-based gasification systems feature significant climate end users. First, policymakers are interested in the environmental
change mitigation benefits and short climate impact mitigation periods benefits (e.g. GHG mitigation) of the system relative to existing pro-
[2–4]. Rice straw was converted into syngas (also known as producer cesses, which can be estimated through life cycle assessment (LCA).
gas) via gasification for the synthesis of dimethyl ether, which can be Second, investors desire profitability, as their investment interests are
used as automotive fuel for diesel engines and a liquefied petroleum gas dependent on the commercial viability of the system, as evaluated using
supplement for household applications [5]. Mazzola et al. [6] showed cost-benefit analysis (CBA). Indeed, one of the major barriers to the
that utilizing woody biomass-based gasification could effectively re- success of existing decentralized bioenergy systems has been their
duce the levelized cost of electricity (LCOE) of isolated microgrids by commercial infeasibility [26]. Third, the electricity should be afford-
38%, relative to diesel engine-based systems. The greenhouse gas able to the end users, which is critical for the long-term viability of the
(GHG) emissions of cornstalk biomass briquette fuel was shown to be a project [27]. However, most of the existing studies evaluated the fea-
full order of magnitude lower than that of coal in China [7]. sibility of decentralized bioenergy systems using only environmental or
Indonesia is the largest energy consumer in Southeast Asia (SEA), only economic criteria. There has yet to be a comprehensive study
accounting for 36% of the region’s energy consumption [8]. Currently, evaluating both the techno-economic feasibility and the environmental
over 90% of electricity produced in Indonesia comes from fossil fuels, sustainability of bioenergy systems that are designed to address needs
with coal accounting for over 50% [9], making Indonesia one of the of all relevant stakeholders.
largest greenhouse gas GHG emitters in the world. The Indonesian In this work, we study the potential of decentralized gasification
government plans to reduce the country’s dependence on fossil fuels by systems in the disposal of oil palm biomass and the electrification of
increasing the share of renewable energy sources in the primary energy rural areas in Indonesia in terms of both their techno-economic feasi-
mix, which would contribute to reduce its GHG emission by 26% below bility and environmental sustainability. The techno-feasibility of the
the business-as-usual BAU value [9,10]. gasification technology is firstly reviewed based on existing studies.
Apart from its environmental concerns related to electricity gen- Then, two gasification-based system designs are proposed with respect
eration, Indonesia has also experienced great difficulty in expanding its to villages and mills and are compared with existing practices from
current national grid and related energy services to remote rural areas, environmental and economic perspectives using LCA and CBA, re-
largely due to its archipelago geography and forested countryside. Low spectively. The conditions supporting commercial viability of gasifica-
population density and electricity demand, along with low paying ca- tion-based systems are identified. The practical deployment of the
pacity of rural residents, make the long-distance transmission of cen- systems is also discussed.
tralized electricity prohibitively expensive [11]. As a result, there were
still over 10 million households without access to electricity in 2016
[12]. Decentralized power generation therefore represents perhaps the 2. Methodology
best solution to the country’s rural electrification dilemma. This de-
centralization could be accomplishe by transmitting electricity from 2.1. System and scenario design
distributed energy resources to surrounding households via “mini-
grids”. Such decentralized systems are particularly suitable in the use This work considers a representative palm oil mill that is supported
cases of remote, mountainous villages where electricity access is a by 9000 ha (on average) of plantations, distributed throughout the
central economic and social issue [13]. However, for the electrification surrounding villages [28]. The average population of a village was 1217
of rural areas, power generation systems based on the fossil fuels are [29], while the average size of a household was 4.3 people [30], which
generally less economically feasible than those based on renewable suggests an average of ca 300 households per village. To consider the
energy resources [14]. Overall, decentralized renewable energy systems variation of household number, 300 and 500 households per village are
should therefore be proposed for electrifying rural areas. considered and the whole village shares a single gasification system.
Indonesia has the highest biomass energy potential in the SEA re- According to the “PIR Trans” smallholder oil palm farming program,
gion, with oil palm biomass being the dominant biomass source [15]. each household owns a 2-ha plot [31,32]. The production of oil palm
The country produces more palm oil than any other nation, accounting biomass in Indonesia is summarized in Table 1. Sung [33] estimated
for 52% of global production in 2012–2013 [16]. Gravimetrically, only
10% of a palm tree will be converted to oil products, while the rest of a Table 1
tree becomes waste biomass [17]. Solid oil palm biomass waste includes Production of oil palm biomass in Indonesia [10].
oil palm fronds and trunk (OPF and OPT), produced from pruning and
Oil palm biomass Production (dry ton ha−1 yr−1)
felling during field plantation operations, as well as empty fruit bunches
(EFB), palm kernel shells (PKS), and palm mesocarp fibers (PMF), Field biomass OPF 11
generated as a byproduct of the palm oil production process in mills. OPT 2.8
Mill processing biomass EFB 1.6
This oil palm biomass has not been fully utilized for power generation,
PKS 1.1
and the current biomass waste combustion-based method of power PMF 1.7
generation causes problems typical to biomass combustion, such as

496
S. You et al. Applied Energy 208 (2017) 495–510

Fig. 1. A schematic of the (a) reference and (b) proposed scenarios. Green circles denote villages and the gray triangle denotes the palm oil mill. In the reference scenario, the villages are
powered by diesel generators while the mill is powered by a combustion-based system whose capacity is determined by either the amount of biomass available (Table 1) or the daily mill
electricity demand. In the proposed scenario, the villages are powered by gasification systems whose capacity is determined by either the amount of biomass available (Table 1) or the
daily village electricity demand, and the mill is powered by a gasification system whose capacity is determined by either the amount of biomass available (Table 1) or the daily mill
electricity demand. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

that up to 0.9–1.5 ton ha−1 yr−1 OPF and OPT could be removed from a electricity demand (the aggregate of household electricity demand in
mature oil palm plantation to maintain an acceptable soil nutrient the village). In the second village sub-scenario (V2), the capacity of the
condition without adding extra fertilizers. We consider that an amount system is designated by the daily village electricity demand. Hence, the
equivalent to that of OPT is available for power production from the system in the V2 scenario always has a capacity factor of 100%. All of
field. the village scenarios are without connection to the national grid and
A schematic diagram of the reference and proposed scenarios are operate for 6 h per day for 365 days per year. Since energy storage
shown in Fig. 1. Reflecting current practice in the study area, in the systems (e.g., batteries) are relatively capital intensive [34], they are
reference scenario (Fig. 1(a)), the villages were powered by diesel not included in the system design.
generators, while the mill was powered by boiler combustion of oil The first mill sub-scenario (M1) has a capacity designated by the
palm biomass. In the village reference scenario (D), each household has amount of biomass available (i.e. EFB, PKS, and PMF). Without the
a 2 kW diesel generator that is operated for 6 h per day for 365 days per national grid, the capacity factor of the system is dependent upon the
year. In the mill reference scenario, two sub-scenarios (R1 and R2) are electricity demand of the mill. However, if the national grid is avail-
considered: the capacity of R1 is determined by the amount of biomass able, which is true for some mills located in relatively large towns, the
available, while the capacity of R2 is determined by the daily electricity system could have a capacity factor of 100% with the electricity being
demand of the mill. The systems of R1 and R2 are operated for 12 h per fed into the grid. In the following analysis, we will consider a special
day for 264 days per year. case in which the system has a 100% capacity factor to fully utilize the
The proposed village and mill scenarios (Fig. 1(b)) have two sub- biomass available for biochar production, but only electricity is only
scenarios, respectively. In the first village sub-scenario (V1), the capa- used to satisfy the daily mill demand. The second sub-scenario (M2) has
city of the system is designated by the amount of OPT available, while a capacity designated by the electricity demand of the mill. The mill
the actual electricity generation is reflected by the capacity factor, scenarios are operated for 12 h per day for 264 days per year. The
which is defined as the ratio between the energy output and the ca- differences between the considered scenarios are listed in Table 2.
pacity of the system. A 100% capacity factor means that the system
consumes all the biomass available. The system in the V1 scenario has 2.2. Gasification and biomass
the ability to cater to future increase in electricity demand. In the fol-
lowing analysis, we will consider a special case in which the system has There are three major types of gasifiers: fixed bed, fluidized bed,
a 100% capacity factor to fully utilize the biomass available for biochar and entrained flow. Fixed bed gasifiers are further divided into down-
production, but electricity is used only to satisfy the daily village draft and updraft architectures. Guangul et al. [21] compared the

Table 2
A comparison of scenario considerations.

Symbol System capacity designation Electricity generation National grid

Village V1 Amount of available biomass Satisfying village demand No

V2 Daily village electricity demand Satisfying village demand No
D# Daily village electricity demand Satisfying village demand No

Mill M1 Amount of available biomass Satisfying mill demand or 100% capacity factor with electricity fed into the grid No/Yes
M2 Daily mill electricity demand Satisfying mill demand No
R1 & Amount of available biomass Satisfying mill demand or 100% capacity factor with electricity fed into the grid No/Yes
R2 Daily mill electricity demand Satisfying mill demand No

#
‘D’ denotes the diesel generator-based reference scenario for the case of village.
&
‘R’ denotes the boiler combustion-based reference scenario for the case of mill.

497
S. You et al. Applied Energy 208 (2017) 495–510

Table 3
Chemical compositions of oil palm biomass.

Oil palm biomass OPF PKS EFB EFB OPT PMF

Proximate Volatile 85.1 87.82 87.37 83.5 75.20 68.8

analysis (wt matter
% dry Fixed 11.5 16.12 9.47 15.2 10.04 15.2
basis) carbon
Ash 3.4 4.44 3.16 1.3 4.79 10.2

Ultimate Carbon 42.4 49.65 42.08 44.58 41.88 43.19

analysis (wt Hydrogen 5.8 6.13 7 4.53 5.98 5.24
% dry Oxygen 48.2 43.33 49.93 48.80 43.24 49.79
basis) Nitrogen 3.6 0.41 0.99 0.71 3.76 1.59
Sulphur – 0.48 0.07 0.35 0.19

HHV (MJ/kg) 18.04 16.14 19.35 17.02 17.52 19.0

Moisture content after drying 4 5–11 11/11.3# 5.18 9.97 –
(wt%)
Reference [91] [92] [93] [94] [95] [96]

#
Values correspond to the EFB pellets of the diameter of 6 and 8 mm, respectively.

Fig. 2. A schematic diagram of a generic downdraft fixed-bed gasification system. moisture content for gasifiers is generally suggested to be lower than
25 wt%, because high moisture content can adversely affect gas yields,
carbon conversion efficiency, and overall gasification efficiency
different types of gasifiers for processing OPF and found that fixed bed
[24,43]. OPF, OPT, EFB, and PMF must undergo a drying pretreatment
downdraft gasifiers were the best choice for use with OPF feedstock,
process prior to gasification. The moisture content of OPF could be
based on seven criteria, including fabrication cost, ease of operation, tar
effectively reduced to 16 wt% after 20-day air drying [21]. Air drying is
content, and cold gas efficiency. Downdraft gasifiers also have the ad-
adopted as an economical and feasible method, in view of the abundant
vantage of being suitable for the small-scale decentralized power gen-
solar energy and space in the rural areas of Indonesia.
eration [35]. Hence, downdraft gasifiers are the type considered in this
The chemical and energy properties of oil palm biomass are shown
work.
in Table 3. OPF, PKS, and EFB could have high volatile contents, up to
A typical downdraft fixed bed gasification system is illustrated in
80 wt% on a dry basis. The high volatile content suggests that it has a
Fig. 2. The feedstock is introduced into the hopper and pretreated in the
high reactivity (rapid volatilization) during the gasification process.
drying bucket by the heat from the hot producer gas. The motorized
The resulting producer gas may contain high tar concentrations, which
screw feeder moves the feedstock to the heat exchanger, where the
could be mitigated in a downdraft gasifier [44]. One method is filtra-
drying and pyrolysis processes of gasification take place. The combus-
tion, which can remove tar from the gas stream prior to its utilization in
tion and reduction processes take place at the bottom of the reactor.
a genset for power generation. The ash content of oil palm biomass is
The vacuum from the engine pulls the gas through the system into the
generally less than 10 wt% on a dry basis [18,20,21,36,45], which is
engine, lowering the pressure in the reactor below atmospheric pressure
favorable for mitigating some ash-related problems, such as slagging
and drawing the ambient air into the combustion zone of the reactor.
and clinkering. The carbon and energy contents of the oil palm biomass
The producer gas from the reactor is first cleaned with a cyclonic se-
are comparable to those of some other commonly-used biomass gasi-
parator to remove particulate matter and then goes around the drying
fication feedstock, such as corn stalk and wheat straw, suggesting that
bucket to heat up the feedstock. The producer gas then goes through a
oil palm biomass waste is an energy efficient source for power gen-
filter (to remove tar), after which it goes into the gas engine for power
eration from gasification.
generation. To initiate the gasification process, a few mL of auxiliary
Table 4 lists the existing gasification experiments on oil palm bio-
fuel (kerosene) is used to ignite the reactor bed through an ignition port
mass with fixed bed downdraft gasifiers. The LHV of producer gas
on the side wall of the reactor. The biochar is obtained at the bottom of
ranges from 3.75 to 5.9 MJ/Nm3, which is comparable to that of syngas
the reactor. The power output is calculated as Moghadam et al. [36]
from other types of biomass such as coconut shells, hazelnut shells, and
P = FR × LHVfeedstock × CGE × EF (1) woody biomass [46]. The carbon conversion efficiency and CGE are
larger than 70% and 50%, respectively. A higher CGE suggests a greater
where FR is the feeding rate (kg/h) and EF is the electrical efficiency of power generation potential from the biomass. The biochar yield ranges
the gas engine. LHV is the lower heating value and estimated based on from 5.2% to 29.13% and the carbon content of biochar from the ga-
the higher heating value (HHV), as shown in the following Table 3 sification of OPF, PKS, and EFB is 91%, 81%, and 75%, respectively
under an assumption of 10 wt% moisture content with the empirical [47]. When biochar produced by gasification systems is used as soil
equation [37]: LHV = HHV-0.212MH-0.0245 Mm-0.008MO. MH, Mm, amendment, it serves as a carbon sink and facilitates carbon abatement
and MO are the weight percentage of hydrogen, moisture, and oxygen, [25]. Gasification biochar can also improve soil structure, nutrient and
respectively. Based on Table 3, the average value of cold gas efficiency water retention, and increase crop productivity [48,49]. However,
(CGE), 58%, is used. EF is set at 42% for a typical gas engine [38]. studies of the effects of the gasification biochar specifically from oil
Hence, the overall electrical efficiency is around 24%, which is con- palm biomass is still lacking.
sistent with values reported by previous studies [39,40]. Electricity is
also consumed by the gasification system itself because of the use of
motors, the automated controls, and the electricity loss in the system. 2.3. LCA
This onboard consumption is termed auxiliary electricity consumption
(AEC). AEC is considered to be 10% of the total electricity generation A LCA was conducted to identify the most environmental-friendly
[27]. power generation strategies for the village and mill use cases. The LCA
Raw OPF and OPT have a moisture content of approximately 70 wt boundary (gate-to-gate) is illustrated in Fig. 3. Inventory data was
% [41,42], while EFB, PKS, and PMF have a moisture content of 65 wt collected regarding the stages of infrastructure construction, oil palm
%, 10 wt%, and 40 wt%, respectively [10]. The acceptable feedstock biomass transportation, thermal conversion (gasification or

498
S. You et al. Applied Energy 208 (2017) 495–510

Table 4
A comparison of the conditions and results of the gasification of oil palm biomass.

Oil palm biomass OPF OPF OPF PKS OPF OPF EFB# EFB* EFB

Gasifier power (kW) 250 250 250 – 50 50 – – 106

Gasifying agent Air Air Air – Air Air Air Air Air
Temperature (°C) – – 700–900 700 – – – – 900

Gas composition CO 22.49 24.98 17.54 14 18.07–28.89 22.78 17 17.40 –

(vol%) CH4 1.98 2.49 1.15 5 0.91–1.76 2.02 1.90 1.50 –
H2 9.67 13.58 9.13 9 8.47–12.59 8.47 13.50 12.90 –
CO2 – – 12.91 15 7.58–15.43 11.81 14.50 13.70 –
N2 – – 59.28 – 51.40–55.81 – 53.3 55.0 –

Syngas LHV (MJ/Nm3) 4.88 5.9 3.75 4.84 4.95 4.66a 4.3 4.1 4.7
Gas yield (Nm3/kg) 1.95 1.8 2.51 – – 1.91 1.8–2.1 2.1–2.5 –
Carbon conversion efficiency 83.8 87.3 91.18 – 93.94 74.4 – – –
(%)
Cold gas efficiency (%) 56.1 62.5 60.37 51 – 51.5 – – 64
Biochar yield (%) 9.3 8.4 9.4 29.13 5.2 – – – –
Reference Atnaw et al. [19], Atnaw et al. [19], Atnaw Ariffin Atnaw et al. Guangul Erlich and Erlich and Ariffin
Guangul et al. [20], Guangul et al. [20], et al. [19] et al. [18] [46] et al. [21] Fransson [93] Fransson [93] et al. [18]
Guangul et al. [21] Guangul et al. [21]

a
The number is with respect to HHV.
#
Denotes 6 mm pellet.
* Denotes 8 mm pellet.

combustion), exhaust gas purification, ash management, and avoided effects of other factors on oil palm biomass could hardly be observed
emissions by the generated products upon substituting existing mate- [51,52]. Given that the choice of treatment strategy on oil palm bio-
rials and energy carriers. The key parameters in life cycle inventory mass has a limited relevance on the practice of plantation, removing
(LCI) are shown in Table 5, and detailed explanations about infra- palm plantation from the considered scenarios in the gate-to-gate
structure construction are given in Table S1 (Please see the boundary does not adversely affect the comparison of results and even
Supplementary Material). The GHG emissions from palm plantations allows a “zoom-in” differentiation of the various utilization options of
are outside of the LCA boundary in this study. It is commonly agreed oil palm biomass.
that the global warming (GW) contribution of the upstream operation The CO2-eq emissions during the thermal conversion of oil palm
(mainly plantation) to the farm gate is predominant for a cradle-to-gate biomass are assumed to be carbon-neutral and not to contribute to the
boundary [50]. The impacts of palm plantation, especially the trans- GW impact, since the amount of CO2-eq released during biomass utili-
formation of tropical rain forests to plantation, is so large that the zation is offset by the CO2-eq eliminated from the atmosphere by

Fig. 3. The LCA boundary (dash lines) for the reference scenarios (a) and new scenarios (b). T denotes transportation.

499
S. You et al. Applied Energy 208 (2017) 495–510

Table 5
LCI.

Item Value Source

Diesel-electric generator
Diesel required for 1 MJ electricity 0.0667 kg Ecoinvent 3.3
GHG emissions for 1 MJ electricity 0.25 kg CO2-eq Ecoinvent 3.3

OPT
Fertilizing value of 1 ton OPT 27 kg CO2-eq Calculated based on OPT element composition

Gasification
Kerosene usage 10 L/year/gasifier Assumed
Fly ash production 8.941 g/kg DM Ecoinvent 3.3 data for softwood
Biochar production 123 g/kg DM
Carbon content in the biochar 80%
Recalcitrant carbon in biochar 80% Galinato et al. [57]
Syngas production rate 2 Nm3 syngas/kg solar dried
material
Syngas composition 20% CO, 2% CH4, 11% H2 Average value from literature
GHG emissions from syngas engine 0.05 g/Nm3 syngas Calculated based on methane concentration in syngasa

Combustion
Combustion ash production rate 7% of the DM of feedstock Doka [97]
K% in mill solid residue 1.89% Sung [33]
K transfer coefficient from feedstock to 41.2% Ecoinvent 3.3 data for incineration plant
combustion ash
Wastewater (m3), average 1.49 × 10−5 m3/1 kg dry feedstock Ecoinvent 3.3 dataset for wood chips burning in the boiler with a capacity of 6667 kW is
Waste mineral oil 6.20 × 10−5 kg/1 kg dry feedstock utilized to describe the emissions of burning mill solid residue.
Municipal solid waste 6.20 × 10−5 kg/1 kg dry feedstock
Chlorine, gaseous 6.20 × 10−6 kg/1 kg dry feedstock
Ammonia, liquid 1.55 × 10−7 kg/1 kg dry feedstock
Water (kg), decarbonised 1.49 × 10−2 kg/1 kg dry feedstock
Sodium chloride, powder 7.75 × 10−5 kg/1 kg dry feedstock
Lubricating oil 6.20 × 10−5 kg/1 kg dry feedstock
Chemical, organic 1.11 × 10−4 kg/1 kg dry feedstock
Dinitrogen monoxide 4.35 × 10−5 kg/1 kg dry feedstock
Methane 0.95 kg/1 kg dry feedstock
Transportation
Lorry 16–32 metric ton, EURO3 0.168 kg/tkm Ecoinvent 3.3

a
For the modelling of syngas combustion, it is assumed that CO is converted completely to CO2, and that CO2 does not react in the combustion process and is emitted as it is. CH4 is
considered as ‘natural gas’ and modelled according to the emissions of the process ‘natural gas burned in industrial boiler > 100 kW (emissions only)’. H2 is converted to water.

photosynthesis during the growth of the biomass from which the The functional unit for the mill scenarios is set as the treatment of
emissions are released [53]. The gasification systems in the villages are all the oil palm biomass available in the mill (EFB, PKS, and PMF). In
assumed to occupy the shrubland in the vicinity of the villages, while the reference scenario of mill (R2), all the biomass residues are sun-
forest cleared to accommodate the gasification (or boiler combustion) dried before combustion, and the bottom ash is utilized as K fertilizer
system of the mill. Shrubland soil and tropical forest soil were reported [58]. The inventory data about fly ash disposal, wastewater treatment
to have a soil organic carbon (SOC) content of 54 ton C/ha [54] and and gaseous pollutant emissions during combustion is based on the
62 ton C/ha [55], respectively, which is lost during the land transfor- Ecoinvent database. The generators are correspondingly dimensioned
mation. One kg C released to the atmosphere from the SOC deficit is to produce sufficient electricity to power the palm oil milling process.
equivalent to 3.67 kg CO2-eq GHG emissions, and the resulted GW im- In scenario R1, the national grid is available to the mill. Upon satisfying
pact is allocated to the electricity produced from the land during the the electricity demand of the mill, extra electricity could be used to
whole lifetime of the power system (25 years). displace Indonesian grid electricity, which is based on 50% lignite coal,
The functional unit for the village scenarios is set as the generation 29% petroleum products, 17% natural gas, 8% hydropower and 6%
of 1 kWh electricity. In the diesel generator reference scenario of the renewables [59]. Based on this power generation mix, 1048 g of CO2-eq
village (D), the OPT from pruning and replanting is left in the field as GHG emissions could be avoided by feeding 1 kWh of electricity from
fertilizer, while the electricity in the households is provided by a diesel the gasification system at the mill into grid. In M2 scenario, the capacity
generator. Inventory of the generator manufacture and pollutant of the gasification system matches the current mill power consumption,
emissions during diesel burning are sourced from Ecoinvent data. In the and the remaining un-gasified portion of the biomass waste is processed
proposed scenarios, chemical fertilizer is added to the field to make up through combustion to yield ash fertilizer, without energy recovery.
the nutrient loss caused by the removal of OPT. Based on the element The detailed LCI is shown in Table 5 (more explanations on LCI are
composition, the fertilizing value of 1 ton OPT is equal to 3.8 kg N, given in Table S1). We use GaBi LCA software and ReCiPe 1.08 Mid-
1.1 kg P2O5, 3.1 kg K2O fertilizer on a dry basis, corresponding to point impact categories for the analysis.
37.3 kg CO2-eq if transportation is also taken into account. For the
syngas and biochar yields, and CGE, the average values calculated 2.4. CBA
based on Table 4 are used. The GHG emission from the engine running
on syngas is calculated based on the measurement of Ahrenfeldt et al. A list of the cost and benefit information is given in Table 6. The
[56]. The fly ash collected from the syngas cleaning process is con- cost of gasification and combustion systems is given for 2006 and 2007,
sidered to be disposed of in a landfill. The carbon in the biochar is respectively. The Chemical Engineering Plant Cost Index (CEPCI) is
classified as recalcitrant, with a sequestration rate of 80% of the total used to update these costs to the current year:
carbon contained in the biochar (−2.35 kg CO2-eq/kg biochar) [57].
Costi = Costj (CEPCIi/ CEPCIj ) (2)

500
S. You et al. Applied Energy 208 (2017) 495–510

Table 6
List of system parameters for LCA and CBA.

Item Unit Value Reference

System life time Year 25 Balamurugan et al. [98]

Gasification system USD/kW 1500 & Abe et al. [61], Suramaythangkoor and Gheewala [62]
O & M (gasification system) % 16.8* You et al. [81]
CGE % 58 Table 2
EF % 42 François et al. [38]
Combustion system USD/kW 1200 Malek et al. [63]
O & M (Combustion system) % 4* Malek et al. [63]
Overall EF (Combustion system) % 15 González et al. [42]
Kerosene USD/L 0.9 Dufo-López et al. [99]
Kerosene usage L//gasifier/year 10 This work
Diesel generator USD/kW 250 Thangavelu et al. [95]
O & M (Diesel generator) USD/kW/year 18 Thangavelu et al. [95]
Overall EF (Diesel generator) % 27 Thangavelu et al. [95]
Diesel USD/liter 0.32 Blum et al. [23]
Heating rate of diesel MJ/L 35 Rodríguez-Fernández et al. [100]
Mini-grid cost USD/meter 20 Adaramola et al. [101]
Transmission length Meter 25 Adaramola et al. [101]
Biochar USD/ton 0, 500, 2650 Choppala et al. [102], Liu et al. [103], Meyer et al. [104], Ahmed et al. [67]
Biochar yield % 12.3 Table 2
Electricity tariff USD/kWh 0.1–0.3 This work

&
The cost includes all components of the system (i.e. gasifier, syngas cleaning, and engine).
* The value denotes the ratio between the O & M cost and capital cost.

where i and j denote the most recent year (2016) and base year (2007 NPV and LCOE are used as the indicators in the CBA. NPV is cal-
and 2006, respectively). The annual values of CEPCI for 2006, 2007 and culated as
2016 are 499.6, 525.4 and 541.7, respectively. The scale dependence of LT
facility cost is considered by Jenkins [60] Cit
NPV = ∑ (1 + r )t
−C0
t (4)
Costk = Costi (Sk / Si ) f (3)
where Cit is the net cash inflow in a year t; C0 is the total initial in-
vestment; LT = 25 years denotes the lifetime of facilities; r is the dis-
where Sk and Si denote the designed facility capacity and base facility
count rate and is set as 10% [68,69]. The LCOE represents the minimum
capacity (9 kW [61] and 1 MW [62] for small-scale village and medium-
electricity tariff for the break-even point of the project over its lifetime.
scale mill gasification systems; and 10 MW [63] for combustion sys-
It is calculated as
tems, respectively). f is the scaling factor and is set to be 0.7 [64].
The benefits in the gasification-based scenarios derive from sales of LT LT
Cct Et
electricity and biochar. The national electricity tariff (ET) was around LCOE = ∑ (1 + r )t
∑ (1 + r )t
t t (5)
$0.11/kWh (current ET) in July 2016 [12] which is similar to In-
donesia’s feed-in-tariff (FiT) rate for biomass-generated renewable en- where Cct and Et is the overall cost and energy generation in a year t. To
ergy [65]. To explore the effect of ET on the economics of the various account for the uncertainty in the data regarding cost and benefit, the
scenarios, a range of 0.1–0.3 USD/kWh is considered in the analysis. On cost and benefit parameters in Table 6 are assumed to follow triangular
average, a remote rural household has an average daily electricity distributions widely employed in CBA [70,71]. The values in Table 6
consumption of around 0.4 kWh (current household demand) con- serves as the modes, while 150% and 50% of the nominal values are the
sidering a major electricity consumption period between 18 pm – 24 am upper and lower limits of triangular distributions. Monte Carlo simu-
(6 h) [23]. To explore the effect of household electricity demand, a lation is used to model the triangular distributions in the analysis.
range of 0.4–6 kWh is considered, which corresponds to a gasification
capacity range of 22.2–333 kW and 37–555 kW for the cases of 300 and 3. Results and discussion
500 households, respectively, in V2. In V1, the full load capacities of the
system are calculated to be 0.8 and 1.4 MW for the cases of 300 and 500 3.1. LCA
households, respectively. In addition, the current household demand of
0.4 kWh corresponds to a capacity factor of 2.8% for the system in V1. 3.1.1. Village
To study the effect of household electricity demand for sub-scenario V1, The net GW impact of the reference scenarios and proposed alter-
the capacity factor is varied from 2% to 40%. The mill has a power natives, as well as the contribution of each factor of the scenarios, per
consumption rate of 16 kWh per ton of fresh fruit bunches (FFB) pro- kWh of electricity generated, are shown in Fig. 4. The process emis-
cessed [66]. A FFB production rate of 20.78 ton/ha/yr (backward cal- sions, the manufacturing of diesel generator, and the use of diesel ac-
culated based on the EFB production rate) and a total number of counts for 60.8%, 28.5%, and 10.7% of the net GW burden (1261.6 g
9000 ha plantations leads to an electricity demand of ca 11,000 kWh, CO2-eq/kWh) of Dl (the reference scenario under the current daily
based on 12 h of operation per day. For sub-scenario M2, the daily household electricity demand of 0.4 kWh), respectively. In contrast, the
electricity demand is varied from 10,000 to 20,000 kWh to study the proposed gasification-based scenarios (V2l/300 and V2l/500) have a net
effect of electricity demand. For sub-scenario M1, the current electricity CO2-eq reduction, due to the strong carbon sequestration effects of using
demand of 11000 kWh per day means a capacity factor of 7.4%. The biochar as a soil amendment. The GW benefit of biochar is −272.4 g
capacity factor is varied from 5% – 100% to study its effect on elec- CO2-eq/kWh, which has the potential to fully offset the GHG emissions
tricity demand. The global average biochar price was reported to be associated with gasification-based electricity production. Generally, the
2650 USD/ton [67]. However, the biochar market is still not mature in dominant GW contributor to the gasification-based scenarios is the
Indonesia, despite consistent growth. The benefit of biochar is con- manufacturing of the gasification system, followed by fertilizer com-
sidered by setting the biochar price at 0, 500, and 2650 USD/ton. pensation (i.e. the extra fertilizer application due to removal of OPT

501
S. You et al. Applied Energy 208 (2017) 495–510

Fig. 4. The global warming (GW) impacts of the reference

scenarios and proposed alternatives for a village. Dl: the
diesel generator (reference) scenario with the current
household electricity demand (0.4 kWh); V2l/300: V2
(0.4 kWh and 300 households); V1l/300: V1 (2.8%, 0.4 kWh,
and 300 households); V1Fl/300: V1 (100%, 0.4 kWh, and 300
households); V2l/500: V2 (0.4 kWh and 500 households);
V1l/500: V1 (2.8%, 0.4 kWh, and 500 households); V1Fl/500:
V1 (100%, 0.4 kWh, and 500 households). The percentages
denote the capacity factors.

from fields) and mini-grid construction. The GW burdens from the ga- CO2-eq/year), while 2.0% is ascribed to the transportation of combus-
sification process emissions, kerosene use, and land usage are in the tion bottom ash back to the fields as fertilizer. Substituting the bottom
range of 10−2–10 CO2-eq/kWh, which are orders of magnitude smaller ash for K fertilizer offers an emissions saving of 209.6 ton CO2-eq/year.
than the major emissions contributors, and are thus negligible. Feeding the gasifier-generated electricity into the national grid dis-
However, the scenarios V1l/300 and V1l/500 had a net GW burden, as places the emissions produced by electricity generation from fossil fuel
the manufacturing of gasification system (i.e. the energy and raw ma- in power plants. The R1 system could provide 1.7 × 107 kWh electricity
terials during the manufacturing of gasification system) contributes to the grid per year and earn a GHG emissions credit of −2.5 × 104 ton
640.3 g CO2-eq/kWh, which overtakes the GW benefit of biochar. Only CO2-eq/year, which renders a net environmental GHG mitigation of
when the gasification capacity is fully utilized and all of the available 2.4 × 104 ton CO2-eq/year in R1.
OPT waste is converted to energy and biochar, does the GW impact of The GW impact from the construction of the gasification system for
the gasification system become net carbon negative, as shown in V1F M1 is much higher than that of the reference scenarios (R1 and R2). The
scenarios. Although extra removal of field OPT increases the chemical GW burden from the gasification process emissions is 336.2 ton CO2-eq/
fertilizer input, with a GW impact being sixty times higher than that in year, which is only 60% of that from the combustion process. This is
V1, significant saving from biochar could offset all the GW burdens, driven by far lower NOx emissions for gasification (6 mg per ton dry
leading to a great environmental benefit around −7700 CO2-eq/kWh. biomass) compared to combustion (43.5 mg per ton dry biomass).
The comparison between the two cases with different numbers of Indeed, it is commonly agreed that NOx emissions are in general not a
households (300 vs. 500) revealed that increasing the number of problem for syngas systems. First, less NOx is present in the produced
households only slightly decreases the net GW impact, mainly because synthetic gas due to its lower operation temperature (< 900 °C) and
the environmental impact is dominated by system construction and deficient oxygen condition in gasifier, compared with the combustion
carbon sequestration of biochar, which are assumed to have a linear process (> 1000 °C and sufficient air supply) [72]. Second, the syngas
relationship with the quantity of feedstock and are not affected by the engine is commonly operated with a lean burn (high air:fuel ratio) to
capacity of the system. Among all the scenarios, V1F1/500 (V1 with all increase efficiency, and the excessive air absorbs heat and lowers the
OPT removal to meet 500 household power demanding of 0.4 kWh) has combustion temperature, which reduces the NOx formation [56].
the best performance (−7703.0 g CO2-eq/kWh), followed by V1F1/300 Carbon sequestration by biochar receives a credit of −1.1 × 104 ton
(−7699.3 g CO2-eq/kWh), V2l/500 (−208.5 g CO2-eq/kWh), V2l/300 CO2-eq/year and plays an important role in alleviating the overall GW
(−204.5 g CO2-eq/kWh), V1l/500 (428.0 g CO2-eq/kWh), and V1l/300 impact from the gasification system. M1 realizes the highest net GW
(431.7 g CO2-eq/kWh). If a scenario similar to V1F1/500 was deployed in mitigation by both displacing grid electricity and producing and se-
104 villages, 17.9 thousand fewer tons of CO2-eq could be emitted into questering biochar. Compared to R1, the deployment of a M1 system
the atmosphere per year compared to the current diesel-based practice. provides yearly GHG emissions mitigation of 5.8 × 104 ton CO2-eq.
In M2l, it is assumed that a small-scale gasifier is constructed
3.1.2. Mill corresponding to budget limitations. The credits received from bio-
The GW impacts of the reference scenarios and proposed alter- char (−759.0 ton CO2-eq/year) and bottom ash-based fertilizer sub-
natives for a mill is shown in Fig. 5. Fig. 5 shows that a net environ- stitution (−195.6 ton CO2-eq/year) can fully compensate all the GHG
mental saving is achieved in all the scenarios except for R2. For R2, emissions, leading to a net environmental saving of 222.7 ton CO2-eq/
M1Nl and M2l, the electricity is just satisfying the mill demand and not year for M2l. The results again show that incorporating even a small-
fed into the national grid, and thus the electricity displacement of GHGs scale gasifier in oil palm biomass management can reduce GHG
is considered to be zero. The combustion process emissions (559.4 ton emissions, with a large contribution from biochar production.
CO2-eq/year) accounts for ca 77.9% of the GW burden for R2. The Overall, the results clearly demonstrate that the gasification systems
dominant process emissions contribution is attributed to N2O emissions, have three advantages over the mills’ current biomass combustion
which is mainly formed by the reaction between N2 and O2 at elevated systems: 62% higher electricity output, less than half of the GHG
temperatures both in the boiler and engine. In addition, 20.1% of the emissions during thermal conversion, and tremendous benefits of
total GHG emissions derives from the system construction (144.3 ton carbon sequestration in the produced biochar.

502
S. You et al. Applied Energy 208 (2017) 495–510

Fig. 5. The GW impacts of the reference scenarios and

proposed alternatives for a mill. R1: the electricity could be
fed into the national grid. R2: the electricity just satisfies to
the mill. M1: 100% capacity factor and electricity fed into
the national grid; M2l: M2 under the mill electricity demand
of 11,000 kWh; M1Nl: M1 with 100% capacity factor but
without the national grid connection. Note that the scales of
y-axis between the upper and lower sub-plots are different.

3.2. CBA NPV, suggesting the impact of economies of scale. The biochar pro-
ductivity ranges from a few tons to hundreds of tons for villages, and a
3.2.1. Base case few tons to thousands of tons for the mill. Despite the lack of robust
The base case results are shown in Table 7 in terms of cost and data on the biochar market in Indonesia, the vast agricultural areas
benefit components for the proposed scenarios. In general, the O & M suggests a huge demand in the country for biochar-based agricultural
cost serves as the largest cost contributor for both the village and mill products (e.g., soil conditioner and fertilizer). Existing data from other
scenarios, followed by the system cost, mini-grid cost, and kerosene countries with major economies have shown significant increase on
cost. However, under the current household electricity demand, the biochar demand worldwide, which will serve to facilitate the interna-
mini-grid cost overtakes the system cost in V2. The income from elec- tional biochar market [73].
tricity sales generally dominates the income from biochar. Under the
biochar price of 500 USD/ton and current electricity consumption de-
3.2.2. Monte Carlo simulation
mands, the systems of V1 and M1 have positive NPVs when the capacity
3.2.2.1. Village – NPV. The profitability conditions of the system can be
factor is 100%. We also considered a special case related to V1Fl/
described using NPV contours as shown in Fig. 6. For V1, it is clear that
300––determining the capacity of the system by the amount of biomass
a relatively high (or low) capacity factor is accompanied by a relatively
available and the capacity factor is 100% with electricity consumption
low (high) ET to maintain the same NPV level. At the capacity factor
corresponding to the current daily household electricity demand
level corresponding to the current household demand (2.8%), the ET
(0.4 kWh). In this case, the biochar income overtakes the electricity
must be significantly larger than 0.3 kWh to achieve the break-even in
income and the mean NPV is 1.3 × 104 USD under the biochar prices of
the average NPV. Electricity that is too expensive, however, suggests a
500 USD/ton and current ET (0.11 USD/kWh). In a similar case for M1,
great financial pressure on the end-users. It is important that the
where there is a 100% capacity factor while the electricity income still
electricity price stays within an affordable price range for consumers.
corresponds to the consumption demands, a mean NPV of 7.8 × 105
Past experience from India has shown that some gasification-based
USD resulted under the biochar price of 500 USD/ton and current ET
power generation projects in rural areas closed after few months of
(0.11 USD/kWh).
operation because the low paying capacity of the end users precluded
For V2, although the capacity factor is 100%, it still bears a negative
the economic viability of the project [27]. The monthly net income

Table 7
Cost and benefit components and biochar production for the proposed scenarios.

Scenarios Conditions Cost and benefit components Biochar NPV (USD) LCOE
production (USD/
per system kWh)
ET Demand Capacity Biochar Number of System O&M Mini-grid Kerosene Electricity Biochar (ton/year)
(USD/ (kWh) factor price households cost cost cost cost income income
kWh) (USD/ (USD) (USD) (USD) (USD) (USD) (USD)
ton)

V1 0.11 0.4 2.8% 500 300 3.2 × 105 1.4 × 106 1.5 × 105 225 1.2 × 105 7.0 × 104 5.6 −9.0 × 105 2.16
V1 0.11 0.4 100% 500 300 3.2 × 105 1.4 × 106 1.5 × 105 225 1.2 × 105 2.6 × 106 208 1.3 × 104 0.06
V2 0.11 0.4 100% 500 300 2.6 × 104 1.1 × 105 1.5 × 105 225 1.2 × 105 7.6 × 104 6.08 −1.5 × 105 0.46
M1 0.11 11,000 7.4% 500 – 9.4 × 106 4.0 × 107 492.3 225 8.0 × 106 4.5 × 106 360 −2.0 × 107 0.83
M1 0.11 11,000 100% 500 – 9.4 × 106 4.0 × 107 492.3 225 8.0 × 106 6.1 × 107 4880 7.8 × 105 0.06
M2 0.11 11,000 100% 500 – 1.6 × 106 6.6 × 106 492.3 225 8.0 × 106 4.9 × 106 392 6.9 × 105 0.15

503
S. You et al. Applied Energy 208 (2017) 495–510

Fig. 6. The contours of mean NPV with respect to the capacity factor and ET for V1 ((a) and (b)), and the household demand and ET for V2 ((c) and (d)). (a) and (c) correspond to 300
households. (b) and (d) correspond to 500 households. The blue, red, and green lines denote the biochar prices of 0, 500, and 2650 USD/ton, respectively.

from oil palm plantations was reported to be around 500 USD/month better chance of being profitable. The study by Mangoyana and Smith
per household [30,74]. Considering that the electricity bill accounts for, [26] also indicated that the commercial viability of decentralized sys-
at most, less than 5% of the income [75], the corresponding ET under tems could be enhanced by integrating the bioenergy production system
the current level of household demand should be less than 2.1 USD/ with other production systems or forming a closed-loop waste produc-
kWh. For the case of villages with 300 households: At a biochar price of tion and re-usage system. On the other hand, there is a trend that the
500 USD/ton, the system can reach the NPV break-even point at the household electricity demand will increase with respect to time. Batih
current ET level with a capacity factor of around 44% and a and Sorapipatana [76] showed that the average daily electricity con-
corresponding daily electricity consumption of 6.4 kWh. At a biochar sumption was around 3.4 kWh and 5.4 kWh for the urban households of
price of 2650 USD/ton, the capacity factor would need to be 16% and the lowest (< 31 USD) and second lowest (31–73.5 USD) monthly in-
the electricity consumption would be 2.3 kWh to reach the break-even come per capita in the major cities of Indonesia. When the household
point at the current ET level. For the case of villages with 500 demand in rural areas reaches the urban level for the second lowest
households: At a biochar price of 500 USD/ton, the NPV break-even income families, the designed V2 system will have a high chance of
point is at a 38% capacity factor. With biochar prices at 2650 USD/ton, becoming economically viable under the current ET. Otherwise, sig-
the 500-household break-even capacity factor is 14%. nificant fiscal incentives are needed to ensure the commercial viability
Similarly, for V2, a relatively high (or low) household demand is of the designed system under the current ET and demand.
accompanied by a relatively low (high) ET to maintain the same NPV V2 is more financially viable (has a higher NPV) than V1 under the
level. In the case of 300 households (500 households), a daily house- same conditions (i.e. ET and biochar price) because it could achieve the
hold demand of 2 kWh (1.8 kWh) is required to reach the break-even break-even NPV point at significantly lower household demands. The
point in the average NPV at the current ET and highest biochar price. power generation ability of the V2 system may be limited and primarily
When the ET increases to 0.3 USD/kWh, the break-even of NPV could depends upon expanding the gasification system’s operating hours to
be achieved at the household demand of 0.8 kWh (0.6 kWh) at zero increase power output (at most, a four time increase with 24 h per day
biochar price for the cases of 300 households (500 households). Fig. 6 operation). The V1 system is more flexible and has a greater potential to
also shows that the larger the number of households considered, the cover unexpected future power demand increases. Detailed relation-
easier for the system to be profitable. The biochar price has a greater ships between the NPV and ET, capacity factor, and household demand
impact on the NPV at higher capacity factors (i.e. higher household for V1 and V2 could be found in the Supplementary Material (Figs. S1
demands). This is related to the facts that (1) the capacity factor and and S2).
household demand directly affect the biochar production for both V1
and V2, and (2) the relative effect of biochar income to electricity in- 3.2.2.2. Village – LCOE. The distributions of LCOE for V1 and V2 are
come shrinks at high ET levels. shown in Fig. 7. The mean and standard deviations of the LCOE
A high electricity demand, e.g. by introducing additional productive distributions decrease with increases in capacity factor (V1) and
uses, is desirable so that the correspondingly designed system has a household demand (V2). The average LCOEs in the V1 scenario are

504
S. You et al. Applied Energy 208 (2017) 495–510

Fig. 7. The distributions of LCOE for V1 and V2 for the cases of 300 ((a), (c), and (e)) and 500 ((b), (d), and (f)) households. (a) and (b), (c) and (d), and (e) and (f) correspond to the
household demands (capacity factor) of 0.4 (2.8%), 1.9 (13.1%), and 5.5 kWh (38.0%), respectively for V2 (for V1).

Fig. 8. The contours of mean NPV with respect to the capacity factor and ET for (a) M1, and with respect to the mill demand and ET for (b) M2. The blue, red, and green lines denote the
biochar price cases of 0, 500, and 2650 USD/ton, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

always larger than the current ET, 0.11 USD/kWh, while the average stand-alone PV system was reported to be 0.66 USD/kWh [77], which is
LCOEs in the V2 scenario are smaller than the current ET under the higher than that of the V2 system and the V1 system with a capacity
household demand of 5.5 kWh. Blum et al. [23] estimated the LCOEs of factor larger than 10%. The LCOE of grid-connected PV system was
renewable (micro-hydropower and solar PV + battery) and hybrid 0.17–0.24 USD/kWh for rural areas [78], which is more competitive
(renewable + conventional) energy systems. These LCOEs were compared to V2 under the current household demand, while less
generally larger than 0.5 USD/kWh, except for the micro- competitive under the household demands of 1.9 and 5.5 kWh. A
hydropower-based system (23.4 kW), which had a LCOE less than 0.2 system based on a combination of solar, biomass, and conventional
USD/kWh. Hence, under the current household demand, the average power in remote areas of Indonesia had a LCOE of about 0.6 USD/kWh
LCOE of V2 proposed in this work is less than that of the renewable and [79], which is less competitive compared to the V2 system under the
hybrid systems proposed by Blum et al. [23], except for the micro- current household demand level (i.e. (a) and (b)). The variation of the
hydropower-based system. The necessity of adding a battery to the solar LCOE with respect to the household demand for the diesel generator
system to provide the power usage at night likely caused the higher reference case (please see Fig. S3 in the Supplementary Material) shows
LCOEs for the systems proposed by Blum et al. [23]. The LCOE of a that the LCOE of diesel generator decreases as the household demand

505
S. You et al. Applied Energy 208 (2017) 495–510

increases. Under the same household demand, the average LCOE of V2 USD/kWh, B = 16.8%, C = 24.4%, D = 0.11 USD/kWh, E = 2.8%
is about one-fourth that of the reference scenario, while the average (V1), E = 7.4% (M1), E = 0.4 kWh (V2), E = 11000 kWh (M2),
LCOE of V1 is higher than that of the reference scenario, with the F = 250 USD/ton, and the low and high levels of the factors are −20%
difference being reduced with increases in the capacity factor. and +20% of the nominal values. The main effects and interactions are
calculated by
3.2.2.3. Mill – NPV. The contours of mean NPV for M1 and M2 are 64
1
shown in Fig. 8. The contour patterns are similar to those in the village Eff =
25
∑ ± NPVi,j
scenarios (Fig. 6); yet, a high capacity factor (or mill demand) and ET j=1 (6)
underpin the system's profitability. Fig. 8(a) shows that M1 will reach where ± corresponds to the (+/−) signs of each main effect and in-
the break-even point of the mean NPV at the ET of 0.3 USD/kWh when teraction for each response. More details on the methods could be found
the capacity factor is around 25%, 20%, and 12%, for the biochar prices in the study by You et al. [81]. The main effects and interactions of the
of 0, 500, and 2650 USD/ton, respectively. M2 could reach the break- factors are estimated and their significance is examined using a normal
even point at the current ET and electricity demand and at a biochar probability plot (Fig. 10) wherein, if a factor or an interaction has more
price higher than 500 USD/ton. M1 will have a positive average NPV at significant effect on the NPV, it will deviate farther away from the
the current ET and zero biochar price if its capacity factor is over 70%. straight line.
If the biochar price increases to 500 USD/ton, a capacity factor around Fig. 10 shows that V1, V2, and M1 have similar significant inter-
44% (or 18% for 2650 USD/ton biochar) is needed to reach the break- actions of the factors which are based on E (capacity factor or electricity
even point of the average NPV for M1. demand) and F (biochar price). These suggest the critical roles of bio-
The NPV contours of the reference scenarios (R1 and R2) are shown char marketing and electricity demand in the commercial viability of
in Fig. S8. At the current ET, R1 could not reach the break-even point of these scenarios. For M2, the biochar price (F) and some of its interac-
the mean NPV, even under a full load capacity. When the ET increases tions with factors B to D play a moderate role in M2, which suggests
to 0.3 USD/kWh, R1 will have a positive mean NPV when the capacity that the commercial viability of the M2 system should be relatively
factor is larger than 47%. Similarly, R2 remains unprofitable up to a sensitive to the variation of biochar price, compared to the other sce-
mill demand of 2.0 × 104 kWh under the current ET. If the ET increases narios. For V1 and M1, the most significant main effects are A (cost of
to around 0.24 USD/kWh, R2 could reach the break-even point under gasification system) and C (overall electrical efficiency). For V2 and
the current daily demand per household (0.4 kWh). M2, the most significant main effects are A (cost of gasification system)
and D (ET), followed by B (O & M cost) and C (overall electrical effi-
3.2.2.4. Mill – LCOE. The LCOE distributions (Fig. 9) indicate that the ciency) playing a more moderate role. This means that the cost of the
average and standard deviation of LCOE decreases as the capacity gasification system is always a critical determinant of the system’s
factor increases for M1. The average LCOEs for M1 and M2 are profitability. This is coincident with a recent survey by Aghamo-
respectively larger and smaller than that of grid-connected PV system hammadi et al. [82], which showed that capital investment was one of
(0.17–0.24 USD/kWh), but the average LCOEs for M1 are smaller than the most important factors in the decisions of mill owners to adopt oil
that of stand-alone PV system (0.66 USD/kWh) as reported by [78]. palm waste-based renewable energy business, followed by the attrac-
However, the average LCOE decreases to approximately 0.06 when the tiveness of electricity tariff and biomass supply chain consistency. So-
M1 system operates at a 100% capacity factor (not shown here), re- vacool and Bulan [83] also found that the tariff level and capital ex-
emphasizing the benefit of feeding electricity into the national grid. The penditure critically affected the performance of a renewable energy
LCOEs of M1 and M2 are smaller than those of V1 and V2 under the development plan. For the systems with the capacity determined by the
current electricity demand conditions of village and mill, respectively. amount of available biomass (V1 and M1), more focus should be put to
The average LCOE of R1 (Fig. S9 in the Supplementary Material) are enhance their overall electrical efficiency, whereas adjusting the ET
smaller than that of M1. Hence, better profitability of M1 than R1 as would be desirable for the systems with the capacity determined by the
shown by the contours of mean NPV should be cause by the higher electricity demands. For a large-scale system where an integrated ga-
benefit for M1. When the system capacity is determined by the elec- sification combined cycle is suitable, the net electricity efficiencies
tricity demand of the mill, the LCOE of gasification-based M2 is around could reach up to 30–40% [84]. The adjustment of ET will be subjected
60% of combustion-based R2. to the paying capacity of the consumers. Past experiences from India
have shown that some gasification-based power generation projects in
3.2.3. Sensitivity analysis rural areas closed after few months of operation due to the low paying
To explore the effects of CBA parameters on the NPV for the pro- capacity of the end users [27].
posed scenarios (i.e. V1, V2, M1, and M2), sensitivity analysis was
conducted based on the design-of-experiments (DOE) [80]. Six para- 3.3. Overall discussion
meters, including the unit cost of the gasification system (A), O & M cost
(B), the overall electrical efficiency (i.e. CGE × EF ) (C), ET (D), the Community participation and support are critical for the success of
capacity factor for V1 and M1 or electricity demand for V2 and M2 (E), decentralized bioenergy systems [26]. The proposed scenarios in this
and the biochar price (F), are considered, and thus it is a 26 factorial work pay special attention to the electricity demands of affiliated
design. In the analysis, the setting of the nominal values is A = 1500 communities and serve to enhance the communities’ self-reliance. This

Fig. 9. The distributions of LCOE for M1 and M2. For M1, (a), (b), and (c) correspond to the capacity factors of 7.4% (lower), 10.1% (medium), and 13.4% (higher), respectively. For M2,
(a), (b), and (c) correspond to the daily mill demand of 11,000, 15,000, and 20,000 kWh, respectively.

506
S. You et al. Applied Energy 208 (2017) 495–510

Fig. 10. Sensitivity analysis shown in normal probability plots for (a) V1, (b) V2, (c) M1, and (d) M2.

study works to design decentralized gasification systems with a target reported that the gasification of straw had a GW impact of 80 g CO2-eq
to cater to the demands of all the stakeholders (i.e., consumers, in- per kWh of electricity, which is smaller than that of V1l/300 and V1l/500,
vestors, and policymakers), and is thus of great practical value for but higher than other village scenarios in Fig. 4. Their LCA also con-
guiding the deployment of bioenergy systems in rural areas. In the fu- sidered the impacts of the removal, collection, pre-processing, and
ture, the hybrid LCA- and CBA-based evaluation scheme could be ex- transportation of straw, which played a major role in the calculated GW
tended to the cases of various other biomass types and regions or impact. Similar to our study, Nguyen et al. [88] also found that the
countries for developing a global view concerning the economic and more environmentally friendly nature of gasification, relative to com-
environmental impacts of decentralized gasification systems for bio- bustion, was contributed to by its higher electricity efficiency, lower
mass waste disposal. exhaust emissions, and higher recalcitrant content in the solid residue.
Utilizing oil palm biomass waste for bioenergy could effectively In addition to their environmental and economic viability, the
reduce the GHG emissions and benefit the environmental sustainability successful implementation of decentralized power systems in rural
of the region. Shi and Yamaguchi [85] showed that most of the CO2 areas is critically affected by policy formulation and fiscal incentives
emissions in Indonesia was coming from biomass burning. Note that the [13,86]. Policy formulation is important in (1) effective dissemination
external cost of the reduction of GHG is not accounted for in the CBA. and operation of the systems, and (2) developing education mechan-
However, under a carbon price of 1.34 (Clean Development Mechanism isms for rural communities to understand the importance of renewable
market) or 9.20 (voluntary carbon market) USD/ton CO2-eq. [86], the energy systems in creating employment, increasing incomes, and im-
yearly external benefit by switching from diesel generator (Dl) to ga- proving living standards [12]. Kardooni et al. [89] found that public
sification (V2l/300) is 9667.3 or 64621.3 USD, which increases the acceptance of renewable energy technology is an essential element in
average NPV from −2.2 × 104 USD to 6.6 × 104 USD or 5.6 × 105 the dissemination and development of renewable energy resources. The
USD at the biochar price of 2650 USD/ton. CBA shows that, under the current ET and demands, the proposed
The LCA results of this work is consistent with that of Yang and scenarios could hardly be economically viable unless the electricity
Chen [3] in terms of the largest GHG-emitting components of a gasifi- could be fed into the national grid for M1 and there is a high biochar
cation project, i.e. the operation and construction stages corresponding price (e.g., 500 USD/ton) for M2. As a result, innovative fiscal and fi-
to the consumption of crop residue, electricity and steel. As mentioned nancial incentives should be developed to finance the relatively large
above, the plantation of oil palm biomass is outside the LCA boundary capital costs, initially with acceptable loans, and increase the com-
in this work, but may serve as an important GHG emissions source for a merciality of the market. Indonesia has introduced a feed-in-tariff (FiT)
cradle-to-gate LCA boundary. For example, Silalertruksa et al. [5] found scheme, with the aim of increasing the share of renewable energy in the
that the plantation of rice straw accounts for around 50% of the total country’s total energy mix [65]. The FiT scheme allows renewable en-
GHG emissions of rice straw bio-DME production via gasification. The ergy developers to have a guaranteed benefit by exporting electricity to
production of cereal biomass was also found to dominate the GW im- the main power grid at a fixed, stable price. However, in Indonesia, the
pact for a decentralized, combustion-based plant [50]. Logistics lack of appropriate transmission line infrastructure makes electricity
(transportation) of rice straw was found to account for almost 50% of export to the main grid difficult; the FiT policy could thus hardly be
the GW impact for a gasification-based, large-scale polygeneration enjoyed for the mills without increased access to the national grid. In
(power, ethanol, heating and cooling) plant [87]. Nguyen et al. [88] 2013, the government proposed a ceiling price region to facilitate the

507
S. You et al. Applied Energy 208 (2017) 495–510

development of solar PV power by attracting IPP Independent Power performance of biomass cogeneration technologies. Appl Energy
2015;154:122–30.
Provider investment [90]. Setting a similar ceiling price for bioenergy [5] Silalertruksa T, Gheewala SH, Sagisaka M, Yamaguchi K. Life cycle GHG analysis of
could potentially make the proposed scenarios economically viable, rice straw bio-DME production and application in Thailand. Appl Energy
even under current market conditions. On the whole, a benign inter- 2013;112:560–7.
[6] Mazzola S, Astolfi M, Macchi E. The potential role of solid biomass for rural
action between government, small-holding farmers, and companies electrification: a techno economic analysis for a hybrid microgrid in India. Appl
needs to be maintained for the ultimate success of the business and Energy 2016;169:370–83.
industry. [7] Wang Z, Lei T, Yang M, Li Z, Qi T, Xin X, He X, Ajayebi A, Yan X. Life cycle
environmental impacts of cornstalk briquette fuel in China. Appl Energy
2017;192:83–94.
4. Conclusions [8] Azam M, Khan AQ, Bakhtyar B, Emirullah C. The causal relationship between
energy consumption and economic growth in the ASEAN-5 countries. Renew
Sustain Energy Rev 2015;47:732–45.
Gasification is a technologically feasible means for the disposal of
[9] ADB, Asian Development Bank. Achieving universal electricity access in
oil palm biomass waste in Indonesia, with the ability to relieve the Indonesia: < https://www.adb.org/sites/default/files/publication/182314/
country's over-reliance on fossil fuel-based power generation and fa- achieving-electricity-access-ino.pdf > [accessed at April-12-2017].
cilitate the electrification of its rural areas. Overall, under the current [10] Paltseva J, Searle S, Malins C. Potential for advanced biofuel production from palm
residues in Indonesia, Washington, DC: The International Council on Clean
ET and electricity demand, V2 (village sub-scenario with the capacity Transportation; 2016. Retrieved from < http://www.theicct.org/advanced-
determined by the village electricity demand) is superior to V1 (village biofuel-productionfrom-palm-residues-indonesia > .
sub-scenario with the capacity determined by the amount of biomass [11] Pode R, Diouf B, Pode G. Sustainable rural electrification using rice husk biomass
energy: a case study of Cambodia. Renew Sustain Energy Rev 2015;44:530–42.
available), in terms of both GW impact and economics, unless sig- [12] ADB, Asian Development Bank. Fossil fuel subsidies in Indonesia trends, impacts
nificantly more biochar is produced in the case of V1. For the case of a and reforms: < https://www.adb.org/sites/default/files/publication/175444/
mill, M1 (mill sub-scenario with the capacity determined by the amount fossil-fuel-subsidies-indonesia.pdf > [accessed at April-12-2017].
[13] Kaundinya DP, Balachandra P, Ravindranath N. Grid-connected versus stand-alone
of biomass available) is both environmentally and financially better energy systems for decentralized power – a review of literature. Renew Sustain
than M2 (mill sub-scenario with the capacity determined by the mill Energy Rev 2009;13(8):2041–50.
electricity demand) when the national grid is accessible or significantly [14] Chakrabarti S, Chakrabarti S. Rural electrification programme with solar energy in
remote region-a case study in an island. Energy Policy 2002;30(1):33–42.
more biochar is produced. Under the current daily demand per
[15] Ahmed T, Mekhilef S, Shah R, Mithulananthan N, Seyedmahmoudian M, Horan B.
household (0.4 kWh), deploying the V2 system in 104 villages with 500 ASEAN power grid: A secure transmission infrastructure for clean and sustainable
households each could mitigate up to 17.9 thousand tons of CO2-eq per energy for South-East Asia. Renew Sustain Energy Rev 2017;67:1420–35.
[16] Ng DK, Ng RT. Applications of process system engineering in palm-based biomass
year compared to the current diesel-based practice. If the electricity
processing industry. Curr Opin Chem Eng 2013;2(4):448–54.
could be fed into the national grid, the M1 system with 100% capacity [17] Yang H, Yan R, Chen H, Lee DH, Liang DT, Zheng C. Pyrolysis of palm oil wastes
factor provides yearly GHG emissions mitigation of 5.8 × 104 ton CO2- for enhanced production of hydrogen rich gases. Fuel Process Technol
2006;87(10):935–42.
eq, compared to R1. For V2, daily household demands of 2 and 1.8 kWh
[18] Ariffin MA, Wan Mahmood WMF, Mohamed R, Mohd Nor MT. Performance of oil
are required to achieve the break-even of the average NPV at the cur- palm kernel shell gasification using a medium-scale downdraft gasifier. Int J Green
rent ET (0.11 kWh) and the biochar price of 2650 USD/ton for the cases Energy 2016;13(5):513–20.
of 300 and 500 households, respectively. In terms of the average NPV, [19] Atnaw SM, Kueh SC, Sulaiman SA. Study on tar generated from downdraft gasi-
fication of oil palm fronds. The Scientific World Journal; 2014.
M1 will be economically feasible if the electricity can be fed into the [20] Guangul F, Sulaiman S, Ramli A. Temperature profile and producer gas compo-
national grid, while M2 will be economically feasible at the biochar sition of high temperature air gasification of oil palm fronds. IOP Conference
price of 500 USD/ton. The average LCOE of V2 is about one-fourth that Series: Earth and Environmental Science. IOP Publishing; 2013. p. 012067.
[21] Guangul FM, Sulaiman SA, Ramli A. Gasifier selection, design and gasification of
of the diesel generator-based reference scenario, while the average oil palm fronds with preheated and unheated gasifying air. Biores Technol
LCOE of V1 is larger than that of the reference scenario. The average 2012;126:224–32.
LCOE of M1 decreases to around 0.06 USD/kWh when the full load [22] Ogi T, Nakanishi M, Fukuda Y, Matsumoto K. Gasification of oil palm residues
(empty fruit bunch) in an entrained-flow gasifier. Fuel 2013;104:28–35.
capacity is reached. Sensitivity analysis shows that the most significant [23] Blum NU, Wakeling RS, Schmidt TS. Rural electrification through village grid-
main effects on the NPV are the cost of the gasification system and the s—assessing the cost competitiveness of isolated renewable energy technologies in
overall electrical efficiency. Indonesia. Renew Sustain Energy Rev 2013;22:482–96.
[24] Vijaya S, Ma A, Choo Y, Nik Meriam N. Life cycle inventory of the production of
crude palm oil-a gate to gate case study of 12 palm oil mills. J Oil Palm Res
Acknowledgement 2008;20(June):484–94.
[25] Lehmann J, Gaunt J, Rondon M. Bio-char sequestration in terrestrial ecosystems-a
The authors acknowledge the funding support from the National review. Mitig Adapt Strat Glob Change 2006;11(2):395–419.
[26] Mangoyana RB, Smith TF. Decentralised bioenergy systems: a review of opportu-
Research Foundation (NRF), Prime Minister’s Office, Singapore under nities and threats. Energy Policy 2011;39(3):1286–95.
its Campus for Research Excellence and Technological Enterprise [27] Palit D, Malhotra R, Kumar A. Sustainable model for financial viability of decen-
(CREATE) program. Grant Number R-706-001-101-281, National tralized biomass gasifier based power projects. Energy Policy
2011;39(9):4893–901.
University of Singapore. [28] Yuliansyah AT, Hirajima T. Development of the Indonesian palm oil industry and
utilization of solid waste. J Min Mater ProcessInst Jpn 2009;125(12):583–9.
Appendix A. Supplementary material [29] Gatto M, Wollni M, Asnawi R, Qaim M. Oil palm boom, contract farming, and rural
economic development: village-level evidence from Indonesia. World
Development; 2017.
Supplementary data associated with this article can be found, in the [30] Lee JSH, Ghazoul J, Obidzinski K, Koh LP. Oil palm smallholder yields and in-
online version, at http://dx.doi.org/10.1016/j.apenergy.2017.10.001. comes constrained by harvesting practices and type of smallholder management in
Indonesia. Agron Sustain Dev 2014;34(2):501–13.
[31] Paoli G, Schweithelm J, Gillespie P, Kurniawan Y, Aurora L, Harjanthi R. Best
References management practices in the Indonesian palm oil industry: Case studies. Daemeter
Consulting: Bogor, Indonesia; 2014. p. 9.
[32] Pauli N, Donough C, Oberthür T, Cock J, Verdooren R, Abdurrohim G, et al.
[1] González-García S, Iribarren D, Susmozas A, Dufour J, Murphy RJ. Life cycle as-
Changes in soil quality indicators under oil palm plantations following application
sessment of two alternative bioenergy systems involving Salix spp. biomass:
of ‘best management practices’ in a four-year field trial. Agr Ecosyst Environ
bioethanol production and power generation. Appl Energy 2012;95:111–22.
2014;195:98–111.
[2] Field JL, Tanger P, Shackley SJ, Haefele SM. Agricultural residue gasification for
[33] Sung CTB. Availability, use, and removal of oil palm biomass in Indonesia.
low-cost, low-carbon decentralized power: An empirical case study in Cambodia.
< http://www.theicct.org/sites/default/files/publications/Teh_palm
Appl Energy 2016;177:612–24.
%20residues_final.pdf > [accessed at April-12-2017].
[3] Yang J, Chen B. Global warming impact assessment of a crop residue gasification
[34] Ho WS, Khor CS, Hashim H, Lim JS, Ashina S, Herran DS. Optimal operation of a
project—a dynamic LCA perspective. Appl Energy 2014;122:269–79.
distributed energy generation system for a sustainable palm oil-based eco-com-
[4] Djomo SN, Witters N, Van Dael M, Gabrielle B, Ceulemans R. Impact of feedstock,
munity. Clean Technol Environ Policy 2015;17(6):1597–617.
land use change, and soil organic carbon on energy and greenhouse gas

508
S. You et al. Applied Energy 208 (2017) 495–510

[35] Martínez JD, Mahkamov K, Andrade RV, Lora EES. Syngas production in down- optimum size. Biores Technol 2010;101(14):5609–21.
draft biomass gasifiers and its application using internal combustion engines. [65] MEMR, Ministry of Energy and Mineral Resources. Feed in Tariff: < http://ebtke.
Renew Energy 2012;38(1):1–9. esdm.go.id/regulation/9/feed.in.tariff > [accessed at April-07-2017].
[36] Moghadam RA, Yusup S, Uemura Y, Chin BLF, Lam HL, Al Shoaibi A. Syngas [66] Yusoff S. Renewable energy from palm oil-innovation on effective utilization of
production from palm kernel shell and polyethylene waste blend in fluidized bed waste. J Clean Prod 2006;14(1):87–93.
catalytic steam co-gasification process. Energy 2014;75:40–4. [67] Ahmed MB, Zhou JL, Ngo HH, Guo W. Insight into biochar properties and its cost
[37] Channiwala S, Parikh P. A unified correlation for estimating HHV of solid, liquid analysis. Biomass Bioenergy 2016;84:76–86.
and gaseous fuels. Fuel 2002;81(8):1051–63. [68] Ertürk M. The evaluation of feed-in tariff regulation of Turkey for onshore wind
[38] François J, Abdelouahed L, Mauviel G, Patisson F, Mirgaux O, Rogaume C, et al. energy based on the economic analysis. Energy Policy 2012;45:359–67.
Detailed process modeling of a wood gasification combined heat and power plant. [69] Manioğlu G, Yılmaz Z. Economic evaluation of the building envelope and opera-
Biomass Bioenergy 2013;51:68–82. tion period of heating system in terms of thermal comfort. Energy Build
[39] Arena U, Di Gregorio F, Santonastasi M. A techno-economic comparison between 2006;38(3):266–72.
two design configurations for a small scale, biomass-to-energy gasification based [70] Barrett SR, Yim SH, Gilmore CK, Murray LT, Kuhn SR, Tai AP, et al. Public health,
system. Chem Eng J 2010;162(2):580–90. climate, and economic impacts of desulfurizing jet fuel. Environ Sci Technol
[40] Aziz M, Kurniawan T, Oda T, Kashiwagi T. Advanced power generation using 2012;46(8):4275–82.
biomass wastes from palm oil mills. Appl Therm Eng 2017;114:1378–86. [71] Withers MR, Malina R, Gilmore CK, Gibbs JM, Trigg C, Wolfe PJ, Trivedi P, Barrett
[41] Bocci E, Sisinni M, Moneti M, Vecchione L, Di Carlo A, Villarini M. State of art of SR. Economic and environmental assessment of liquefied natural gas as a sup-
small scale biomass gasification power systems: a review of the different typolo- plemental aircraft fuel. Prog Aerosp Sci 2014;66:17–36.
gies. Energy Procedia 2014;45:247–56. [72] Maya DMY, Sarmiento ALE, de Sales CAVB, Oliveira EESL, Andrade R. Gasification
[42] González A, Riba J-R, Puig R, Navarro P. Review of micro-and small-scale tech- of municipal solid waste for power generation in brazil, a review of available
nologies to produce electricity and heat from Mediterranean forests wood chips. technologies and their environmental benefits. J Chem 2016;10:249–55.
Renew Sustain Energy Rev 2015;43:143–55. [73] Guo M, Uchimiya SM, He Z. Agricultural and environmental applications of bio-
[43] Jarungthammachote S, Dutta A. Equilibrium modeling of gasification: Gibbs free char: Advances and barriers. Soil Science Society of America, Inc.; 2016.
energy minimization approach and its application to spouted bed and spout-fluid [74] Feintrenie L, Chong WK, Levang P. Why do farmers prefer oil palm? Lessons learnt
bed gasifiers. Energy Convers Manage 2008;49(6):1345–56. from Bungo district Indonesia. Small-scale Forest 2010;9(3):379–96.
[44] Atnaw SM, Sulaiman SA, Yusup S. Syngas production from downdraft gasification [75] Wijaya ME, Tezuka T. A comparative study of households' electricity consumption
of oil palm fronds. Energy 2013;61:491–501. characteristics in Indonesia: A techno-socioeconomic analysis. Energy Sustain
[45] Lahijani P, Zainal Z. Fluidized bed gasification of palm empty fruit bunch using Develop 2013;17(6):596–604.
various bed materials. Energy Sources, A: Recov, Utilizat, Environ Effects [76] Batih H, Sorapipatana C. Characteristics of urban households electrical energy
2014;36(22):2502–10. consumption in Indonesia and its saving potentials. Renew Sustain Energy Rev
[46] Atnaw SM, Sulaiman SA, Yusup S. Influence of fuel moisture content and reactor 2016;57:1160–73.
temperature on the calorific value of syngas resulted from gasification of oil palm [77] Veldhuis A, Reinders A. Potential and cost-effectiveness of off-grid PV systems in
fronds. Sci World J; 2014. Indonesia-An evaluation on a provincial level. In: Photovoltaic Specialist
[47] Mahmood WMFW, Ariffin M, Harun Z, Ishak N, Ghani JA, Ab Rahman MN. Conference (PVSC), 2014 IEEE 40th. IEEE; 2014. p. 3562–67.
Characterisation and potential use of biochar from gasified oil palm wastes. J Eng [78] Veldhuis A, Reinders A. Reviewing the potential and cost-effectiveness of grid-
Sci Technol, vol. 10(Spec. Issue on 4th International Technical Conference (ITC) connected solar PV in Indonesia on a provincial level. Renew Sustain Energy Rev
2014); 2015. p. 45–4. 2013;27:315–24.
[48] Carter S, Shackley S, Sohi S, Suy TB, Haefele S. The impact of biochar application [79] da Fonseca MB, Poganietz W-R, Gehrmann H-J. Environmental and economic
on soil properties and plant growth of pot grown lettuce (Lactuca sativa) and analysis of SolComBio concept for sustainable energy supply in remote regions.
cabbage (Brassica chinensis). Agronomy 2013;3(2):404–18. Appl Energy 2014;135:666–74.
[49] Hansen V, Müller-Stöver D, Ahrenfeldt J, Holm JK, Henriksen UB, Hauggaard- [80] Montgomery DC. Design and analysis of experiments; 1991.
Nielsen H. Gasification biochar as a valuable by-product for carbon sequestration [81] You S, Wang W, Dai Y, Tong YW, Wang C-H. Comparison of the co-gasification of
and soil amendment. Biomass Bioenergy 2015;72:300–8. sewage sludge and food wastes and cost-benefit analysis of gasification-and in-
[50] Sastre C, Maletta E, González-Arechavala Y, Ciria P, Santos A, del Val A, et al. cineration-based waste treatment schemes. Biores Technol 2016;218:595–605.
Centralised electricity production from winter cereals biomass grown under cen- [82] Aghamohammadi N, Reginald SS, Shamiri A, Zinatizadeh AA, Wong LP, Nik
tral-northern Spain conditions: global warming and energy yield assessments. Appl Sulaiman NMB. An investigation of sustainable power generation from oil palm
Energy 2014;114:737–48. biomass: a case study in Sarawak. Sustainability 2016;8(5):416.
[51] Jungbluth N, Chudacoff M, Dauriat A, Dinkel F, Doka G, Faist Emmenegger M, [83] Sovacool BK, Bulan L. Energy security and hydropower development in Malaysia:
et al. Life cycle inventories of bioenergy. Final report ecoinvent data v2. 0; 2007. the drivers and challenges facing the Sarawak Corridor of Renewable Energy
p. 17. (SCORE). Renew Energy 2012;40(1):113–29.
[52] Wiloso EI, Bessou C, Heijungs R. Methodological issues in comparative life cycle [84] Belgiorno V, De Feo G, Della Rocca C, Napoli R. Energy from gasification of solid
assessment: treatment options for empty fruit bunches in a palm oil system. Int J wastes. Waste Manage 2003;23(1):1–15.
Life Cycle Assess 2015;20(2):204–16. [85] Shi Y, Yamaguchi Y. A high-resolution and multi-year emissions inventory for
[53] Eksi G, Karaosmanoglu F. Combined bioheat and biopower: a technology review biomass burning in Southeast Asia during 2001–2010. Atmos Environ
and an assessment for Turkey. Renew Sustain Energy Rev 2017;73:1313–32. 2014;98:8–16.
[54] i Canals LM, Muñoz I, McLaren S, Brandão M. LCA methodology and modelling [86] Sparrevik M, Lindhjem H, Andria V, Fet AM, Cornelissen G. Environmental and
considerations for vegetable production and consumption. CES Working Paper 02/ socioeconomic impacts of utilizing waste for biochar in rural areas in Indonesia-a
07; 2007. systems perspective. Environ Sci Technol 2014;48(9):4664–71.
[55] Guillaume T, Damris M, Kuzyakov Y. Losses of soil carbon by converting tropical [87] Jana K, De S. Environmental impact of biomass based polygeneration – a case
forest to plantations: erosion and decomposition estimated by δ13C. Global study through life cycle assessment. Biores Technol 2017;227:256–65.
Change Biol 2015;21(9):3548–60. [88] Nguyen TLT, Hermansen JE, Nielsen RG. Environmental assessment of gasification
[56] Ahrenfeldt J, Jensen TK, Henriksen U, Schramm J. Investigation of continuous gas technology for biomass conversion to energy in comparison with other alter-
engine CHP operation on biomass producer gas. SAE Technical Paper natives: the case of wheat straw. J Clean Prod 2013;53:138–48.
2005:0148–7191. [89] Kardooni R, Yusoff SB, Kari FB. Renewable energy technology acceptance in
[57] Galinato SP, Yoder JK, Granatstein D. The economic value of biochar in crop Peninsular Malaysia. Energy Policy 2016;88:1–10.
production and carbon sequestration. Energy Policy 2011;39(10):6344–50. [90] Hirsch B, Burman K, Davidson C, Elchinger M, Hardison R, Karsiwulan D, et al.
[58] Maschowski C, Zangna MC, Trouvé G, Gieré R. Bottom ash of trees from Cameroon Sustainable energy in remote indonesian grids: accelerating project development.
as fertilizer. Appl Geochem 2016;72:88–96. National Renewable Energy Laboratory; 2015.
[59] Itten R, Frischknecht R, Stucki M, Life cycle inventories of electricity mixes and [91] Sa S, Atnaw SM, Ao M. Feasibility study of gasification of oil palm fronds. J Mech
grid; 2012. Eng Sci 2015;9:1744–57.
[60] Jenkins BM. A comment on the optimal sizing of a biomass utilization facility [92] Samiran NA, Jaafar MNM, Ng J-H, Lam SS, Chong CT. Progress in biomass gasi-
under constant and variable cost scaling. Biomass Bioenergy 1997;13(1–2):1–9. fication technique-With focus on Malaysian palm biomass for syngas production.
[61] Abe H, Katayama A, Sah BP, Toriu T, Samy S, Pheach P, et al. Potential for rural Renew Sustain Energy Rev 2016;62:1047–62.
electrification based on biomass gasification in Cambodia. Biomass Bioenergy [93] Erlich C, Fransson TH. Downdraft gasification of pellets made of wood, palm-oil
2007;31(9):656–64. residues respective bagasse: experimental study. Appl Energy
[62] Suramaythangkoor T, Gheewala SH. Potential alternatives of heat and power 2011;88(3):899–908.
technology application using rice straw in Thailand. Appl Energy [94] Mohammed M, Salmiaton A, Azlina WW, Amran MM, Fakhru’l-Razi A. Air gasi-
2010;87(1):128–33. fication of empty fruit bunch for hydrogen-rich gas production in a fluidized-bed
[63] Malek AA, Hasanuzzaman M, Rahim NA, Al Turki YA. Techno-economic analysis reactor. Energy Convers Manage 2011;52(2):1555–61.
and environmental impact assessment of a 10 MW biomass-based power plant in [95] Thangavelu SR, Nutkani IU, Hwee CM, Myat A, Khambadkone A. Integrated
Malaysia. J Clean Prod 2017;141:502–13. electrical and thermal grid facility-testing of future microgrid technologies.
[64] Sultana A, Kumar A, Harfield D. Development of agri-pellet production cost and Energies 2015;8(9):10082–105.

509
S. You et al. Applied Energy 208 (2017) 495–510

[96] Chiew YL, Iwata T, Shimada S. System analysis for effective use of palm oil waste [101] Adaramola MS, Agelin-Chaab M, Paul SS. Analysis of hybrid energy systems for
as energy resources. Biomass Bioenergy 2011;35(7):2925–35. application in southern Ghana. Energy Convers Manage 2014;88:284–95.
[97] Doka G. Updates to life cycle inventories of waste treatment services-Part II: Waste [102] Choppala G, Bolan N, Megharaj M, Chen Z, Naidu R. The influence of biochar and
incineration. Zürich, Switzerland: Doka Life Cycle Assessments; 2014. black carbon on reduction and bioavailability of chromate in soils. J Environ Qual
[98] Balamurugan P, Ashok S, Jose T. Optimal operation of biomass/wind/PV hybrid 2012;41(4):1175–84.
energy system for rural areas. Int J Green Energy 2009;6(1):104–16. [103] Liu X, Zhang A, Ji C, Joseph S, Bian R, Li L, et al. Biochar’s effect on crop pro-
[99] Dufo-López R, Bernal-Agustín JL, Yusta-Loyo JM, Domínguez-Navarro JA, ductivity and the dependence on experimental conditions-a meta-analysis of lit-
Ramírez-Rosado IJ, Lujano J, et al. Multi-objective optimization minimizing cost erature data. Plant Soil 2013;373(1–2):583–94.
and life cycle emissions of stand-alone PV–wind–diesel systems with batteries [104] Meyer S, Glaser B, Quicker P. Technical, economical, and climate-related aspects
storage. Appl Energy 2011;88(11):4033–41. of biochar production technologies: a literature review. Environ Sci Technol
[100] Rodríguez-Fernández J, Lapuerta M, Sánchez-Valdepeñas J. Regeneration of diesel 2011;45(22):9473–83.
particulate filters: effect of renewable fuels. Renew Energy 2017;104:30–9.

510