System overview and characterization of a high-temperature, high-pressure, entrained-flow, laboratory-scale gasifier

Gasification is a technology that has seen increasing international interest as a cleaner method to convert the vast worldwide coal reserves into syngas for chemical and power production. In addition to coal, gasification can be used to convert biomass, municipal waste, and other alternative feedstocks into high-value products.¹ Gasification of these alternative feedstocks has the additional benefits of potentially net zero carbon dioxide release to the atmosphere and utilization of readily available low-cost fuels.² 

Entrained-flow gasifiers are a leading gasification technology³ as they are able to produce cleaner syngas with a higher overall efficiency. These advantages are realized from their ability to operate at temperatures and pressures that can exceed 1900 K and 50 bar.⁴ Other advantages of entrained-flow gasifiers include having a high throughput per reactor volume, flexibility in feedstock, and a relatively simple mechanical design. The laboratory-scale entrained-flow gasifier at the Colorado School of Mines (CSM) serves as a compact test bed for the gasification of carbonaceous materials (e.g., coal and biomass) over a wide range of operating conditions.

In addition to gasification for syngas production, the reactor has also been designed for studying the chemical kinetics important to gasification. Gasification of carbonaceous materials proceeds in two primary steps: the devolatilization of the feedstock followed by the slower char gasification reactions. These slower char gasification reactions are rate limiting and often determine the size, pressure, and temperature of the gasifier necessary to achieve complete conversion.⁵ At the high operating temperatures of coal gasifiers, char kinetics are known to be strongly influenced by surface and pore diffusion, meaning that the mass transport through the particle boundary layer and especially within the pore of the particle determines the rate of the char gasification reactions.⁶ In order to further advance the design of all gasifiers, a fundamental understanding of char kinetics at relevant process conditions is required. However, due to the extreme difficulty of collecting accurate char kinetics data in industrial scale gasifiers, the majority of investigations are performed on smaller, laboratory-scale or bench-scale gasifiers and reactors. Although the size and feed rates used in laboratory-scale gasifiers and reactors are typically orders of magnitude smaller than those found in industrial gasifiers (e.g., feed rates of kilograms per hour versus tons per hour), the data that are collected allow for a greater understanding of the coal gasification process which is directly applicable to all scales of gasification.

A significant amount of research has been devoted to char gasification in the previous half-century and has been summarized in numerous literature reviews on the kinetics of char gasification.^7–14 Thermogravimetric analyzers (TGAs) are often used to study char kinetics; however they are not well suited for studying char kinetics in the diffusion-limited regime due to the difficulty in achieving high temperatures and heating rates, the influence of buoyancy effects, and the inability to isolate mass diffusion to other particles.^15,16

The need for gasification studies at smaller scales, while still maintaining the ability to reach high temperatures and pressures, has led several groups to construct pilot-scale, laboratory-scale, and bench-scale reactors. A significant benefit of laboratory-scale gasifiers and reactors, in contrast to pilot-scale systems, is the associated operating costs. For the cost of operating a typical pilot-scale facility for a week, the CSM gasifier facility is able to operate for several months or longer.

For reference, the primary characteristics of some past and existing research reactors are summarized in Table I. These include facilities at Brigham Young University (BYU, USA), the University of Utah (USA), Sandia National Laboratories (USA), the Commonwealth Scientific and Industrial Research Organisation (CSIRO, Australia), the Central Research Institute of Electric Power Industry (CRIEPI, Japan), the Cooperative Research Center for New Technologies for Power Generation of Low-Rank Coal (CRC, Australia), the Korea Advanced Institute of Science and Technology (KAIST, South Korea), the University of Nottingham (UK), and the Georgia Institute of Technology (GA Tech, USA).

Although the physical size and temperature rating of the CSM gasifier are similar to the other past and existing laboratory-scale reactors, the primary characteristics that differentiate the CSM gasifier are the maximum operating pressure, the feed rate, and the ability to switch between flow regimes. As discussed earlier, industrial entrained-flow gasifiers can operate at pressures above 50 bar; therefore, laboratory-scale investigations must be performed at elevated pressures in order to produce data that are relevant. While lower feeding rates do not allow for the rapid generation of product gases, lower feeding rates do allow for highly dilute mixtures to be investigated. These dilute mixtures allow for the isothermal approximation to be applied appropriately inside the reactor and also allow for the gasification investigations to be performed in situ (measurements inside the reactor) opposed to ex situ (measurements only at the exit). The CSM gasifier also has the ability to operate in both laminar and turbulent flow regimes by switching between different injectors. Another advantage of the CSM gasifier is the ease with which investigators have control over multiple process inputs, allowing for an extensive range of operating conditions and environments to be achieved (e.g., coal-steam and coal-steam-oxygen). Additionally, the CSM gasifier was designed to allow for in situ measurement of gasification kinetics using laser diagnostics via three sets of optical access ports along the length of the reactor.

Optical diagnostics have long been used for studying chemical kinetics and have recently been applied in gasification environments. For example, in situ diagnostics based on calibration-free wavelength-modulation absorption spectroscopy^28,29 have been demonstrated to measure temperature and water concentration in harsh, high-temperature, and high-pressure environments.^30,31 More recently, the technique was demonstrated for temperature measurements in the reactor of an oxygen-blown pilot-scale coal gasifier, along with species concentration measurements in the post-quench downstream gases.²⁰ 

For data collected by a laboratory-scale reactor to be relevant or applicable, the reactor itself must be well-characterized. One primary reason for this is that char kinetics and other intricate studies are highly sensitive to reactor conditions; therefore any fundamental data acquired under unknown conditions are of little use to the gasification community. A well-characterized reactor also provides accurate boundary conditions for creating models as well as data for model validation.

Here, we present a description of the CSM gasifier system followed by initial characterization of the reactor using a coal feedstock.

The system diagram of the CSM gasifier and its primary subsystems is shown in Figure 1. Primary inputs to the gasifier are argon (for entrainment, purging, and quenching), oxygen, steam, and ground solid feedstock. Each input can be controlled independently allowing for a wide range of operating conditions to be achieved. The gasifier facility consists of five primary systems: the primary reactor vessel, the solids delivery system, the steam generator system, the LabVIEW-based control system, and gas processing systems—each of which will be described in detail in Secs. II B–II F.

The primary reactor vessel is a 38 cm diameter, 2 m tall shell-and-jacket pressure vessel that has been ASME certified (“U” designator) to 55 bar at 315 °C. The shell houses the internal components and the jacket is used to flow mineral oil which maintains the external temperature of the gasifier at approximately 50 °C for the pressure rating. The pressure vessel was manufactured and certified by Western Steel and Boiler Company.

An important feature of the pressure vessel is the inclusion of three sets of optical access ports (nine ports total). At three different axial locations (at the tip of the injector, 38 cm below the injector, and at the tip of the extraction probe), there are three optical access ports—two are on opposing sides of the vessel and the third is perpendicular to the first two. The two in-line optical access ports at each location will be utilized by the in situ laser diagnostic that is currently being developed. The laser diagnostic will allow for measurements of the gas temperature and water concentrations inside the reactor. In addition to the laser diagnostic, the orthogonal ports can be used to measure particle velocity fields using particle image velocimetry (PIV) under dilute feeding conditions. A cross-sectional view of the gasifier and the internal components is shown in Figure 2.

The gasifier pressure vessel is supported on an external frame coupled to a hydraulic lift and braking system that allows for the gasifier to be raised, lowered, and rotated. During typical operation, the gasifier is in a vertical orientation, whereas the gasifier is horizontal while undergoing maintenance. Access to the internal components of the gasifier is provided by removing either of the two end-caps (flanges) on the pressure vessel.

At the center of the pressure vessel is a 1.4 m long silicon carbide (SiC) reactor column. The reactor column has a 50 mm inner diameter and a 60 mm outer diameter. It is supported at the base by a stainless steel plate and at multiple axial locations by refractory insulation. In addition to providing support for the reactor column, the refractory insulation also divides the heated section of the pressure vessel into four 25 cm long independently heated zones, for a total heated length of 1 m. External to the reactor column, there is a 3 in. gap between the outside of the reactor column and the innermost layer of insulation where the heaters and thermocouples are located.

For the insulation, three different layers of high-performance refractory insulation supplied by Zircar Ceramics are used. The outermost layer is ALC (rated to 1500 °C), the next layer is AL-30 (rated to 1600 °C), and then the innermost layer is SALI (rated to 1700 °C). ZAL-45 (rated to 1650 °C) is used as supports for the reactor column and the heaters, which also act as baffles to divide the gasifier into four independently controlled heated zones.

To heat the reactor, 16 electrically powered SiC heaters are used which were supplied by I Squared R. Each independently controlled heated zone in the gasifier contains four SiC heaters and each zone has a maximum power input of 4 kW (absolute maximum of 6.6 kW but currently limited by the laboratory power supply). The heaters are capable of operating at 1650 °C in an inert environment (argon or helium are used) or 1370 °C in a reducing or nitrogen environment. In reducing environments, the protective layer of silicon dioxide is reduced and results in deterioration of the heating elements. In nitrogen environments above 1370 °C, nitrogen reacts with the SiC heaters to form silicon nitride which in turn creates hot spots that will damage the heaters—therefore care must be taken to purge all nitrogen (air) from the system before reaching operating temperatures.

Power is supplied to the four heaters in each zone through Spang 851 digital power controllers. Single phase 208 V is supplied to each power controller which then uses a silicon controlled rectifier (SCR) to adjust the voltage to maintain either the prescribed power or current setting for each zone.

Each zone contains two Omega Type C thermocouples which are used for measuring the temperature and for providing feedback to the power controllers for temperature control. The thermocouples in each zone are located on opposite sides of the reactor column and are redundant, allowing for operation if one of the thermocouples goes offline.

Additionally, two-color ratio pyrometers from Micro-Epsilon are used to directly measure the SiC reactor column wall temperature in zones 1, 2, and 4. Optical access to the reactor column is provided through three of the ports on the side of the gasifier using custom flanges with sapphire windows (see Figure 2). A $1 2$ in. diameter hole was drilled through the refractory insulation to provide line-of-sight to the reactor core.

Pressure in the system is measured using Keller America digital pressure transducers. Pressure is continually monitored at four critical locations: the main argon supply line, the solids hopper, the pressure external to the reactor column, and at the injector (assumed to be the approximate pressure inside the reactor column). Additional pressure transducers can be incorporated for pressure measurement at other locations such as in test stands for downstream processing. A proportional pressure relief valve is installed on the reactor vessel to prevent the system from reaching an over-pressurized state.

For the gas feed streams (argon and oxygen), four Bronkhorst mass flow controllers (MFCs) are used. The first MFC controls the argon used for entraining and delivering the feedstock particles to the injector and is rated at 50 standard liters per minute (SLPM). The second MFC controls the argon used for purging the gasifier and maintaining positive pressure on the reactor column and is rated at 100 SLPM. The third MFC controls the argon used for quenching the product stream before it enters the extraction probe and is rated at 10 SLPM. The final MFC controls the oxygen supplied to the injector and is rated at 35 SLPM.

At the top of the SiC reactor column is the multi-stage injector. In the current injector, at the center is the argon and solids feed which is then mixed with the oxygen (if desired) via a venturi mixer. External to the oxygen line is a water jacket which is used to keep the oxygen and the oxygen/solids/argon lines cold to prevent any reactions (i.e., combustion or the release of volatiles) from occurring in the injector. External to the water jacket is an insulating gap that separates the water jacket from the steam pre-heater (“annular steam super heater”). The steam pre-heater is heated by the SiC reactor column through conduction from zone 1. Steam enters the pre-heater at 200 °C and exits at approximately 800 °C. A diagram of the injector is shown in Figure 3. It should be noted that this injector is designed for rapid mixing of the reactants and therefore ideal for syngas generation. This injector creates a turbulent regime close to the injector which then transitions to a laminar flow further down the reactor (at typical operating conditions).

As stated earlier, the injector is not a permanent part of the gasifier and can be switched with another unit. Currently under development is a water-cooled laminar-flow injector. The injector consists of a central tube for the entrained feedstock surrounded by a water jacket (similar to the current injector), which is in turn wrapped in ceramic insulation. Secondary gases (e.g., argon, oxygen, and steam) will be delivered upstream of the injector and heated by the reactor column similarly to the way the “annular steam super heater” in the current injector is heated. To ensure laminar flow, ceramic flow straighteners will be installed around the injector to promote laminar flow in the secondary gases and only dilute mixtures will be passed through the injector (i.e., low solids feeding rates with minimal argon for entrainment).

After the feedstock, argon, and oxygen exit the injector, the mixture is exposed to the super heated steam and the heated SiC reactor column walls. The gasification processes occur in the reactor column, and the product stream (syngas) and any unreacted solids are extracted via the extraction probe. Similar to the injector, the extraction probe has multiple loops and channels. The product stream is extracted through the center of the probe and delivered out of the gasifier. External to the extraction line is an air loop that gradually cools the product stream which allows for any tars, ash, and other byproducts to remain hot enough so that they will not clog the extraction line. External to the air loop is an insulating gap that separates the hot return side of the air loop and the channel that delivers the argon used for quenching to the tip of the extraction probe. At the tip of the extraction probe, the cold argon passes through a porous Hastelloy X frit and then into the syngas. Finally, external to the argon quench channel is a water loop that is used to maintain the integrity of the extraction probe by cooling the Hastelloy X probe face and the external sheath of the probe.

On the outside of the extraction probe is ceramic insulation to keep the probe concentric with the reactor column and to prevent the probe from overheating. A ceramic probe cover sits on top of the probe tip to block thermal radiation from the reactor column. The ceramic cover has one axial hole on the top and six radial holes to allow the product stream to enter the extraction probe. A diagram of the extraction probe is shown in Figure 4.

Currently under development is a translatable extraction probe. This will allow for the product gases to be extracted from the gasifier at different axial positions in the reactor column, providing the capability to vary the residence time in the gasifier at a given temperature, pressure, and input flow and feed rates.

The current solids delivery system is capable of entraining between 2 and 20 g/min of feedstock to the injector. A diagram of the coal delivery system can be seen in Figure 5.

The solids delivery system consists of four primary components: the pressure vessel, the hopper, the auger system, and the eductor. The pressure vessel is similar to the pressure vessel for the gasifier—it has a main cylindrical chamber that is sealed on both ends by Class 600, 8 in. blank flanges and is ASME certified (“U” designator) to 58 bar at 50 °C. During operation, the vessel is pressurized to approximately 1 bar above the gasifier pressure vessel to drive flow from the solids delivery system to the gasifier injector. As mentioned previously, a digital pressure transducer is installed on the pressure vessel to continuously monitor pressure.

Inside of the pressure vessel is a 5 L solids hopper. The hopper can hold enough powdered feedstock to continually supply the gasifier for over 20 h at the nominal feeding rate of 3 g/min. At the center of the hopper is an auger manufactured by Auger Manufacturing Specialists, which is specifically designed for powdered feeds. The auger is driven by an electric motor that is controlled though a LabVIEW program that sets the motor rpm through feedback from an encoder on the drive shaft. In addition to the auger, there is a “wiper” that counter-rotates relative to the auger to help facilitate solids feeding and to prevent bridging of the feedstock inside of the hopper.

After the feedstock exits the funnel from the hopper, it enters the eductor where it is rapidly entrained in an argon vortex and then delivered to the injector via $1 8$ in. tubing. The eductor was fabricated in-house from a solid acrylic cylinder. Several iterations were designed and then simulated in SolidWorks Flow Simulation before fabricating the most effective design. Argon enters the eductor through the four side ports where it is tangentially injected to the conical chamber of the eductor, creating the entrainment vortex.

The solids delivery system has been characterized and the results are shown in Figure 6. The highly linear trend of feeding rate with motor rpm provides for very repeatable solids feeding. It should be noted that while the characterization of the solids delivery system was performed up to 45 g/min, the maximum practical feeding rate is 20 g/min due to entrainment limitations.

Currently under development is a low feed rate solids delivery system. The new system will utilize the current pressure vessel and eductor, but the solids hopper and auger system are being designed to provide feed rates below 1 g/min. The purpose of the low feed rate system is to enable very dilute feedstock mixtures to be injected into the reactor thus allowing for the isothermal approximation to be appropriately applied during char kinetics studies.

A dedicated steam generator system is used to generate steam to feed into the injector. A positive-displacement, micro-annular gear pump (supplied by HNP Mikrosysteme GmbH) is used to pump deionized water from a sterile tank, through a Bronkhorst Coriolis flow meter (to verify flow rates), to the physical steam generator unit.

The steam generator unit is comprised of a $3 8$ in. outer diameter, 8 in. long Omega cartridge heater (rated at 2 kW) that has been fixed inside a $1 2$ in. stainless steel tube with Swagelok fittings. The liquid deionized water passes through the annulus between the cartridge heater and the tubing, producing super heated steam that exits the generator unit at 500 °C. The temperature of the steam generator unit is controlled by an Omega heater controller that obtains temperature feedback from a Type K thermocouple that is in contact with the cartridge heater. The steam exit temperature is also measured with a Type K thermocouple (used for reference only).

After the steam exits the steam generator unit, it is delivered to the injector via $1 8$ in. stainless steel tubing. The steam typically reaches the injector at 200 °C (depending on flow rates) where it then passes through the annular super heater to reach 800 °C before entering the reactor column.

The majority of the gasifier system is controlled through a real-time LabVIEW virtual instrument (VI) that is deployed on a National Instruments PXI unit. The real-time LabVIEW VI controls and monitors the temperature in each zone, controls the power controllers, reads the pressure transducers, and controls the MFCs. There are two supporting VIs that are run on the desktop personal computer (PC), which control the water pump for the steam generator and control the auger rpm for the solids delivery system.

Supporting the PXI unit is a National Instruments SCXI unit that measures all of the thermocouple voltages, filters them, and converts them to temperature readings. The SCXI unit is currently reading in 13 thermocouple inputs but is capable of reading up to 32 thermocouple inputs.

To ensure uninterrupted operation of the gasifier system, the PXI is operated as a stand-alone unit. This allows for operation of the gasifier if communication is temporarily lost with the host PC. Additionally, the host PC, PXI, and SCXI are powered through a battery backup to allow for proper shut down of the gasifier system in the case of a power failure, preventing any potential damage to the gasifier and personnel.

There are several options available for processing the syngas after it has been extracted from the gasifier. These options include dropout tanks (for removal of tar and other byproduct), particulate filtering, sulfur removal, gas analysis, and test stands.

The dropout tanks were fabricated in-house, and these feature large expansion chambers to allow solids to drop out of entrainment and incorporate cooling lines and jackets to facilitate condensation of any residual steam remaining in the syngas. The particulate filtering is provided by a Norman 4300 Series in-line filter. The porous sintered stainless steel element that is currently installed has an absolute filter rating of 0.5 μm (0.2 μm nominal) and can collect up to 60 g of solid mass. A proprietary sulfur sorbent provided by TDA Research, Inc., is used for the removal of sulfur species in the syngas. The sorbent is used as a packed-bed system and typically ran in-line after the particulate filter. If downstream processing of the syngas is desired (e.g., hydrogen separation, sulfur removal, water-gas shift, and Fischer–Tropsch), test stands can be directly attached to the extraction probe or after the dropout tank.

An Agilent 490 Micro Gas Chromatograph (GC) is used to analyze the gaseous products. The GC has three channels installed (capable of four), two MolSieve 5 Å (MS5A) columns and one PoraPlot U (PPU) column, and uses ultra-high purity (UHP) grade nitrogen and helium as the carrier gases. The first MS5A channel uses nitrogen as the carrier gas and measures hydrogen concentration. The second MS5A channel uses helium as the carrier gas and measures argon, nitrogen, methane, and carbon monoxide. The PPU channel also uses helium as the carrier gas and measures carbon dioxide, hydrogen sulfide, carbonyl sulfide, and lower hydrocarbons (up to C₆). In addition to the micro GC, a Fourier Transform Infrared (FTIR) spectrometer with a long pass cell is available for measuring water concentrations.

In the current configuration, the GC can provide a full analysis in two minutes—measuring hydrogen, argon, nitrogen, methane, carbon monoxide, carbon dioxide, hydrogen sulfide, carbonyl sulfide, and up to C₃ hydrocarbons. The MS5A channels have a detection limit of 10 ppm and the PPU channel has a detection limit of 1 ppm. The relative standard deviation (RSD) for each channel is less than 0.5% at constant temperature and pressure.

For personnel safety, there are two RAE Systems gas monitors installed, a QRAE 3 and a ToxiRAE Pro. The QRAE 3 provides detection of carbon monoxide, lower-explosive limit (LEL), hydrogen sulfide, and oxygen concentration, while the ToxiRAE Pro provides detection of hydrogen. The primary purpose for these monitors is to detect any potential leaks coming from the gasifier or the product stream lines.

The capabilities of the CSM gasifier are summarized in Table II. The “total argon” parameter includes argon for coal entrainment, purging the vessel, and quenching the syngas. The values listed in the “Min” column are the lowest achievable values that are nonzero. The minimum and maximum “residence time” values are the residence times that are expected to be achievable using a translatable extraction probe. It should be noted that the “residence time” values are currently calculated assuming plug flow and therefore represent the residence times of the bulk gas. The “heater power” parameter is the power used by the SiC heaters in each zone (i.e., per four heaters).

The data presented in this section are a comparison of the syngas composition at a set pressure, temperature, and feed rate, while varying only the oxygen concentration. The coal used for this study was low-sulfur subbituminous coal from the North Antelope Rochelle Mine in the Powder River Basin (USA). An ultimate analysis of the coal used in the characterization studies is provided in Table III.

For reference, the oxygen used is taken to be a percentage of the oxygen required for complete combustion (CC), calculated from the coal feed rates and the composition based on the ultimate analysis. The steam value is similar—the baseline (100%) is taken to be the stoichiometric amount of steam required to completely gasify the coal; therefore a value of 33% excess would equate to 133% of the steam required for gasification using only steam. The parameters used for the characterization study are summarized in Table IV.

A graphical representation of the data is shown in Figure 7. It should be noted that the data presented have been normalized to only the primary components in the syngas (hydrogen, carbon monoxide, carbon dioxide, and methane)—all other gases, primarily argon, have been excluded. Also, each data point represents the average of a minimum of at least four GC analyses, and the error bars on the graph are the standard deviation at each data point.

The most notable trends for this data set are the increasing carbon dioxide concentration and simultaneous decreasing hydrogen and carbon monoxide concentrations as the oxygen concentration is increased. The addition of oxygen changes the reaction pathways through the increased heat release from partial combustion and the available reactants during the gasification reactions, thus changing the equilibrium composition. Methane concentrations are believed to be low throughout the oxygen sweep due to the excess of steam in the reactor, promoting the steam-methane reforming reaction.

At 30% oxygen, a significant increase in the standard deviation can be seen for the hydrogen, carbon monoxide, and carbon dioxide data points. The cause of the increase is believed to be due to a temporary pulsing in the solids delivery system which caused fluctuations in the coal feeding rate. Since the pulsing was continuous, it did not affect the average value for each of the data points.

The equilibrium gas-phase temperature is estimated to be approximately 1200 °C. This estimation was performed by comparing the actual composition of the product stream to the equilibrium composition predicted by STANJAN as a function of temperature.³² The computed equilibrium temperature is 200 °C below the reactor wall temperature. This lower temperature is expected due to several factors including the reactants are injected at low temperature (approximately 20 °C), the steam enters at 800 °C, and the developed flow is laminar (Re_D ≈ 80) leading to significant radial and axial gas temperature variations.

With a maximum operating temperature and pressure of 1650 °C and 40 bar, independent control of all process inputs, and facilities for downstream processing, the high-temperature, high-pressure, entrained-flow, laboratory-scale gasifier at the Colorado School of Mines is able to achieve a wide range of operating conditions and environments making it a highly flexible facility. As a compact test bed for the gasification of carbonaceous materials, the CSM gasifier is well suited for fundamental research studies, such as char kinetics, and also for downstream processing of syngas.

The CSM gasifier is fully functional and capable of simulating a broad range of gasification environments. A low-sulfur coal feedstock was supplied to the CSM gasifier with excess steam and oxygen concentrations corresponding to 0%–35% of oxygen required for complete combustion, producing syngas with expected concentrations of hydrogen, carbon monoxide, and carbon dioxide.

This material is based upon work supported by the Department of Energy under Award No. DE-NT0005202, a subcontract by Praxair, Inc., under Award No. DE-FE0004908, and the National Science Foundation under Award No. CBET-1336364. We would also like to thank Peabody Energy for providing the coal samples and Zybek Advanced Products for grinding the coal powders.