Data Sets

NRG 2017 Paper Study Overview

Investigator: Daniel P. Dupuis, R. Gary Grim, Eric Nelson, Eric C.D. Tan, Daniel A. Ruddy, Sergio Hernandez, Tyler Westover, Jesse E. Hensley, Daniel Carpenter

Link to study data

Citation

Daniel P. Dupuis, R. Gary Grim, Eric Nelson, Eric C.D. Tan,
Daniel A. Ruddy, Sergio Hernandez, Tyler Westover, Jesse E. Hensley, Daniel Carpenter,
High-Octane Gasoline from Biomass: Experimental, Economic, and Environmental Assessment,
Applied Energy,
Volume 241,
2019,
Pages 25-33,
ISSN 0306-2619,
https://doi.org/10.1016/j.apenergy.2019.02.064.
(https://www.sciencedirect.com/science/article/pii/S0306261919303617)
Article Included (See Attachment)
Link to publication: High Octane Gasoline from Biomass

I. Abstract

Five economically-advantaged biomass feedstocks identified in the United States Department of Energy’s 2016 Billion Ton Study were gasified and the syngas was reformed at the bench scale to study the feedstock price-performance relationship. The distribution of reformed syngas compositions, heating values, and yields were similar across the different feedstocks and blends thereof, which ranged from inexpensive residual wastes to more expensive and higher quality biomass, revealing that feedstock performance was mostly insensitive to its price. Custom blended feedstocks produced syngas with characteristics resembling linear combinations of syngas from single-component feedstocks, supporting the ability to customize and predict blended properties based on single-feedstock data. The experimental gasification data informed a techno-economic analysis of specific feedstock costs for producing high-octane gasoline, and the results showed that miscanthus and forest residues were the most cost-effective. A field-to-wheels life-cycle assessment of greenhouse gas emissions showed that forest residues was the most environmentally benign feedstock of those studied.

II. Method Visualization

Figure 1: Schematic of fluidized bed gasification system with downstream hydrocarbon reformer and acid-gas scrubber highlighting three analytical locations.

 

Figure 2: Simplified block flow diagram of the biomass-to-HOG blendstock conversion process. Adapted from Tan et al

 

Figure 3: Life-cycle system boundary for LCA evaluation.

 

III. Gasification – Figure 1

A schematic of the gasification system used in these experiments is shown in Figure 1. A dual-screw K-Tron hopper metered biomass at 0.6 kg h−1 to a 15.9 mm (OD) feed screw that continuously conveyed it into a 10.16 cm (ID) fluidized bed reactor maintained at 800 °C and 103 kPa. Steam was introduced through the bottom of the reactor at 0.6 kg h−1 to fluidize the olivine bed material, yielding a 1:1 (wt/wt) steam to biomass ratio. A series of two downstream cyclones and a hot gas filter at 500 °C and 98 kPa removed solids from the syngas stream before reforming. The fixed bed reformer, held at 900 °C and 90 kPa contained 7.86 kg of a commercial nickel reforming catalyst (Haldor-Topsoe R-67-7H) which upgraded the raw product gas, contributing to an increase in syngas yield, removal of hydrocarbons (Eq. (1)), and a shift in the relative H2 and CO concentrations (Eq. (2)).

Steam hydrocarbon reforming: CnH2n+2 + nH2O nCO + (2n + 1)H2 (1)

Water gas shift: CO + H2O CO2 + H2 (2)

After reforming, the syngas was cooled to approximately 30 °C and 69 kPa to condense water and any residual heavy hydrocarbons before finally passing through an acid-gas scrubber at 90°C and 55kPa to remove CO2 and H2S. The scrubbing fluid contained 30wt% mono- ethanolamine (MEA), 10 wt% methyl-diethanolamine (MDEA), and 60 wt% water. The syngas composition was measured at three points within the system as depicted in Fig. 1. An Agilent 7890A gas chro- matograph (GC) calibrated for permanent gases and hydrocarbon compounds up to C4 analyzed a slip-stream of raw and reformed syngas at locations one and two, respectively. A California Analytical Instru- ments nondispersive infrared analyzer (NDIR) and Nova Analytical Systems thermal conductivity detector (TCD) continuously monitored.

CO, CO2, O2, CH4, and H2 concentrations at the outlet of the acid-gas scrubber, location 3. The NDIR and TCD data was used to confirm steady-state operation, at which point data collection commenced. In total, gases were sampled over a minimum of three consecutive gas injections per stream at a rate of one sample every 20 min for a total of six datasets (three pre-reformer, three post-reformer).

IV. Techno-economic analysis – Figure 2

A simplified process flow diagram for the biomass-to-HOG conver- sion pathway used as the exemplary TEA case is shown in Figure 2. The detailed process description can be found in Tan et al. [16]. The diagram depicts the major processing steps for the conversion of biomass to syngas via indirect steam gasification, syngas cleanup, and sequential synthesis of methanol, dimethyl ether (DME), and high-octane hydro- carbons.The primary economic indicator for this analysis is specific feed- stock cost, which is defined as the cost of HOG production attributed to the feedstock cost, in dollars per gallon of gasoline equivalent ($/GGE), as illustrated in Equation (3). GGE is based on the lower heating value (LHV) of gasoline (32,356 kJ/L) obtained from Argonne National La- boratory’s GREET model software [17]; the LHV for HOG in this work is 31,094 kJ/L. All costs are adjusted to 2014 US dollars. The HOG yields as a function of feedstock compositions were determined from the Aspen Plus process model simulation [16].

Specific Feedstock Cost = Feedstock Cost /HOG Yield (3)

The assumed design capacity was 2000 dry tonnes per day (2205 dry US tons per day). With an expected 7884 operating hours per year (90% on-stream factor/availability), the annual feedstock requirement is approximately 657,000 dry tonnes per year (724,000 dry US tons per year). The assumed onstream factor allows approximately 36 days of planned and unplanned downtime per year.

The TEA reported here uses nth-plant economic assumptions, the key aspect of which is that a successful industry has been established with many operating plants using similar process technologies. The TEA model encompasses a process model and an economic model. For a given set of conversion parameters, the process model solves mass and energy balances for each unit operation. This data is used to size and cost process equipment and compute raw material and other operating costs. The capital and operating costs are then used for a discounted cash flow rate of return analysis [16]. The nth-plant economic assumptions are summarized in Table 1, while the capital costs and annual operating costs are presented in Tables 2 and 3, respectively.

V. Life-cycle assessment – Figure 3

A complete life-cycle assessment (LCA) study typically evaluates multiple impact categories, including global warming potential, acid- ification, eutrophication, resource depletion, and water use. While LCA can certainly be employed to evaluate a wide range of sustainability indicators, this study focuses only on the life-cycle GHG emissions. The scope of this attributional LCA study concentrated on the field-to-wheels (FTW) GHG emissions of the high-octane gasoline blendstock. GHG emissions are represented in grams of carbon dioxide equivalent (CO2e) using a 100-year GHG emission factor [18]. The system boundary for the LCA study is depicted in Figure 3. For this analysis, we account for the stages in the life cycle of the HOG fuel, including feedstock production and logistics, fuel production and transportation, and fuel consumption or vehicle operation. The functional unit is one megajoule (MJ) of fuel consumed. Additionally, the current LCA was performed to evaluate the entire supply chain or life-cycle GHG emissions of the high-octane gasoline product and is not intended to assess only on any particular unit operation (e.g., gasification).

The corresponding reference flows and life-cycle inventories are summarized in Table 4, which are based on material and energy inputs and outputs to the biorefinery estimated from the simulated conversion processes. The material and energy flows in the conversion step capture the impacts of input raw materials, and outputs, such as emissions, wastes, and coproducts as predicted by the process model. The copro- ducts (excess electricity) are treated as avoided products using the product displacement method [19]. Coproduct displacement is also known as system boundary expansion and is based on the concept of displacing the existing product with the new product. [17]. The GHG basis values for biomass feedstock production and logistics, fuel trans- portation and end use, and electricity are largely applied consistently with the values utilized in GREET [17]. Table 5 summarizes the GHG emission factors employed for the LCA evaluation. The GHG emission factors are used to convert the life-cycle inventory to the life-cycle GHG emissions to a common unit (g CO2e/MJ). It is assumed that carbon sequestered during biomass growth ends up in the fuel after the pro- duction step and returns to the atmosphere in the exhaust stream after the fuel combusts in the engine, so that biogenic carbon is treated as credit.

VI. Results

1. Feedstock properties

2. Syngas composition

3. Techno-economic analysis and life-cycle assessment