Catalyst Residence Time Distributions in Riser Reactors for Catalytic

Subscriber access provided by University of Newcastle, Australia

Article

Catalyst Residence Time Distributions in Riser Reactors for Catalytic Fast Pyrolysis: Part 2: Pilot-Scale Simulations and Operational Parameter Study Thomas D. Foust, Jack L. Ziegler, Sreekanth Pannala, Peter Nolan Ciesielski, Mark R Nimlos, and David J Robichaud ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.6b02385 • Publication Date (Web): 21 Feb 2017 Downloaded from http://pubs.acs.org on February 22, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Sustainable Chemistry & Engineering is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sustainable Chemistry & Engineering

Catalyst Residence Time Distributions in Riser Reactors for Catalytic Fast Pyrolysis: Part 2: PilotScale Simulations and Operational Parameter Study Thomas D. Foust*1†, Jack L. Ziegler1†, Sreekanth Pannala‡, Peter Ciesielski†, Mark R. Nimlos†, and David J. Robichaud† †National Bioenergy Center, National Renewable Energy Laboratory, 15013 Denver West Parkway, Golden, CO 80401 ‡Corporate Research and Development, SABIC Americas, 14100 Southwest Freeway, Sugarland, TX 77478 *To whom correspondence should be addressed [email protected] 1

Co-lead authors

KEYWORDS: catalytic fast pyrolysis, catalytic upgrading, riser reactor, multiphase flow simulation, catalyst residence time distribution

ABSTRACT: Using the validated simulation model developed in part one of this study for biomass catalytic fast pyrolysis (CFP), we assess the functional utility of using this validated model to assist in the development of CFP processes in fluidized catalytic cracking (FCC)

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 34

reactors to a commercially viable state. Specifically, we examine the effects of mass flow rates, boundary conditions (BCs), pyrolysis vapor molecular weight variation, and the impact of the chemical cracking kinetics on the catalyst residence times. The factors that had the largest impact on the catalyst residence time included the feed stock molecular weight and the degree of chemical cracking as controlled by the catalyst activity. Because FCC reactors have primarily been developed and utilized for petroleum cracking, we perform a comparison analysis of CFP with petroleum and show the operating regimes are fundamentally different.

Introduction In Part one of this study, we detailed the simulation development methodology for biomass catalytic fast pyrolysis (CFP) and compared against a validation case, Andreux et al.1, gaining confidence in our modeling methodology. In this second stage, we use the techniques developed in part one to assess the functional utility of using this validated model to assist in the development of CFP processes in fluidized catalytic cracking (FCC) reactors to a commercially viable state. Specifically, we examine the effects of the following on the catalyst residence time distributions (RTDs): mass flow rates, inlet boundary conditions (BCs), and pyrolysis vapor molecular weight variation. The flow rates include the ratios of pyrolysis vapor, fluidizing gases, and solid catalyst. Additionally, the non-reactive model developed in part one was enhanced to incorporate a reduced order chemical mechanism for CFP. With simplified reactive 2D and 3D simulations, this study demonstrates guidelines for improving the overall performance of the CFP in FCC reactors. CFP processes for the production of fuels can be described as a three-step process: preparation of the feedstock, rapid heating in the absence of oxygen (pyrolysis), and catalytic upgrading with


2

Page 3 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


a catalyst. Figure 1 provides a schematic diagram of the National Renewable Energy Laboratory (NREL) CFP pilot scale reactor (NCFP-PSR). The NCFP-PSR operates in an ex situ arrangement where the biomass is pyrolyzed in a bubbling bed reactor and the vapors are upgraded in the circulating bed riser reactor. At the top of the riser, the catalyst is separated from the upgraded vapors and sent through a regenerator removing coke, and then reintroduced into the riser reactor.

Figure 1. Schematic of NCFP-PSR. Computational Approach Model Framework Given these encouraging results for the Andreux et al., validation case detailed in part one of this study for the high mass flux case, our next step is to simulate the NCFP-PSR as described in the next section.


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 34

NCFP-PSR Model Reactor Dimensions and Operating Conditions The simulation model developed for this study is based on a pilot-scale riser reactor at NREL (NCFP-PSR) specifically designed and constructed to develop CFP to a commercially viable process. The riser section of the NCFP-PSR is ~3 meters tall, has a diameter of approximately one centimeter, and operates at 25 psi gage with a pressure drop of approximately –4 inches of water. The riser has three inlets in the bottom portion of the riser as shown, one inlet for the fluidizing N2, and/or steam, one inlet for the catalyst and one inlet for the biomass pyrolysis vapor. The design of the riser is that all three phases are thoroughly mixed in the bottom section of the riser. Simulation Approach Using the typical petroleum operating values as a guide scaled down to the NCFP-PSR, mass flow rates of 350–530 g/hour of pyrolysis feed supplemented with a 50% by volume additional nitrogen and a catalyst-to-feed mass flux ratio (cat/feed) of 11 was selected as the appropriate operating regimes. Additionally, 7.5% by feed weight of steam and 2.5% nitrogen are added as a fluidizing agent. Since the residence time of the vapor phase (pyrolysis vapors) is critical as well, the first step in the simulation is to set initial conditions for vapor phase flow rates and residence times in the NCFP-PSR. Since the vapor phase is less sensitive to boundary conditions the assumption of plug flow is a reasonable assumption for calculating vapor phase velocities and residence times. Using the plug flow assumption, the gas and solids flows from all inlets can be assumed to mix uniformly and have uniform velocity throughout the riser. This leads to the equations for the mass balance,


4

Page 5 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


= ∑ ,

(1)

= .

(2)

Where is the mas flux, ρ is density, V is velocity and A is area. Although these equations do not have a closed form solution since Vout is unknown, by using the additional simplifications of ignoring gravity, pressure drop, friction, heats of reaction, mole changes due to reactions, and kinetic energy, a thermal-energy balance can be used to find the mixture's final temperature, and hence density, through the ideal gas law, ∑ ℎ = ∑ = ∑ , = .

(3) (4)

Where h is the convective heat transfer coefficient, cp is specific heat at constant pressure, T is temperature, P is pressure and R is the universal gas constant. This can be solved for the equilibrium temperature and the mixture density can be calculated through the mixture ideal gas law, and a uniform velocity can be readily calculated. For the 500g/hr and 350g/hr pyrolysis vapor cases the bulk flow velocities are 106 cm/s and 81 cm/s respectively, which yields approximate residence times for the NCFP-PSR of 2.9 and 3.8 seconds respectively. Literature references2 indicate that most FCC risers are designed to operate with a bulk flow gas residence time of 2–3 seconds, hence these flow rates and the bulk flow approximation for the vapor phase roughly conform to this guidance. With these conditions as a starting point, a more detailed multidimensional setup of the problem was constructed. As a first pass, initial and boundary conditions were calculated for a simple fully mixed 2D one-inlet simulation. For this one-inlet simulation, the equilibrium values of temperature and a uniform equal gas and solids velocity were used. In order to avoid having


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 34

the solids coming in at a high velocity, which could lead to asymmetrical results, the gas volume fraction was set to 0.91 yielding gas and solid velocities of 115 and 10 cm/s, respectively. As a check of the physical validity of this result the pressure drop was measured and found to be 7,000 Ba, which is in the range of the 1–4 inches of water (2,500–10,000 Ba) for the NCFPPSR (1 inch of water = ~2,490.9 Ba = 249.09 Pa). Catalyst Properties and Chemistry For this study a, zeolite FCC catalyst3 is used. The physical parameters used for this study are a density of 1.56 g/cm3 and an average diameter of 80 micrometers, which are typical for an industrial zeolite alumina-sol-binded catalyst. This catalyst is of the Geldart A particle type4,5 yielding fluidization properties for "powders." Shown in Table 1 is a summary of our numerical parameter space. In order to employ the desirable no-slip wall (NSW) gas-phase BC and the Johnson and Jackson (J-J) partial-slip solids and the transport form of the granular energy equation (TGE) that produced the best results in part one of this study, the riser is modeled with a rectangular domain with a straight vertical outlet. We also modeled the riser as a physically realistic cylinder for comparison purposes using the Cartesian cut-cell method available in Multiphase Flow Interactions with Exchange (MFIX), which requires use of the algebraic granular energy (AGE) form of the granular energy equation. Table 1. Model Summary. 2-dimensional

3-dimensional

Grid type

Cartesian

Cartesian & Cart. cut-cell

Domain

rectangular

rectang. & cylindrical

Granular energy

algebraic transport

Gas wall BC

no slip

approx.,

full full transport no slip


6

Page 7 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Solids wall BC

partial slip

FSW, NSW, & partial slip

Roughness

specularity 0.001 & 0.05 specularity 0.001 & 0.05

Cells

96 x 5088

24 x 1476 x 24

Resolution

< 8 particle diameters

< 40 particle diameters

Chemistry

nonreactive & 3-step CFP nonreactive kinetic mechanism

NCFP-PSR Model Results Using the cases specified in Table 1, 2D and 3D simulations were conducted and studied extensively for different grid resolutions, wall BCs, and with and without the subgrid scale (SGS) models. Highly resolved non-SGS results in 2D were compared to lower resolved SGS results in 2D and 3D to gauge their expected performance and accuracy improvements in unresolved 3D simulations. Two-dimensional simulations were first conducted and “visual convergence” (defined as observing insignificant changes (

Catalyst Residence Time Distributions in Riser Reactors for Catalytic

Recommend Documents