Simple Parametric Model for Quick Assessment of IGCC Performance

The integrated gasification combined cycle (IGCC) power plant has been the most promising clean coal technology for more than two decades. Several large-scale commercial IGCC power plants have been in operation for many years to the point that the technology can safely be considered field proven [1,2]. Nevertheless, the large capital cost driven by the immense size, complexity, and construction time of IGCC power plants prevented the technology from a more widespread acceptance. Even so, IGCC's favorable emission characteristics (vis-à-vis the conventional coal-fired thermal power plants) and especially its amenability to carbon capture and sequestration (CCS) have so far ensured that research and development of the technology (as well as commercial interest in IGCC) is as active as ever. It is expected to remain so for the foreseeable future due to the abundance of coal as an affordable fossil fuel for electric power generation—as long as its detrimental effects on the environment can be eliminated in an economically and technically feasible manner.

Over the last two decades, many excellent IGCC studies have been published. These studies explored key competing technologies from a performance and cost trade-off perspective [3,4]. All published studies utilize highly complex chemical process and power plant heat balance software, including commercially available packages such as Aspen Plus and/or HYSYS [5], PRO/II [6], GateCycle [7], GT PRO [8], and in-house proprietary codes. Complex simulation software is vital for hardware design and performance evaluation (design as well as off-design). Some software packages also include detailed capital cost estimation add-ins to facilitate trade-off studies to establish IGCC's position vis-à-vis competing clean coal and other electric power generation technologies. An extensive list of recently published works on IGCC models by Perez-Fortes et al. [9] points to Aspen Plus as the most widely used software. (Even then, gasification modeling is mostly done via first principles encapsulated in computer models developed by the researchers using programming languages such as C++ or computing platforms such as Matlab or Microsoft Excel.) Recently, a combination of Aspen Plus (for the process plant) and GT PRO (for the power plant) was used by researchers to model the entire IGCC plant (e.g., Dennis et al. [10]). Note that either software is capable to model the entire IGCC power plant with varying fidelity for plant subsystems. Nevertheless, the primary strength of either, i.e., chemical process modeling for Aspen and gas turbine (GT) and combined cycle (CC) modeling for GT PRO, precludes in-depth study of design options pertaining to all subsystems on a single platform. In addition to commercial software packages, proprietary, in-house calculation platforms capable of simulating the entire IGCC plant are also available in published literature, exclusively from academic researchers (e.g., see Chiesa and Lozza [11] and references listed therein).

Regardless of the particular modeling platform(s) used in obtaining them, it is practically impossible to reproduce the published IGCC performance numbers. The complexity of the models, the large number of model icons and inputs (hundreds of the former and literally thousands of the latter), myriad techniques used in combining different platforms, and obtaining the iterative solutions preclude a full-blown description of the system in the limited space available. (This is in addition to the obvious problem that one would have to pay the significant fees to obtain the requisite commercial software and go through the very steep learning curve of using them for the task at hand.) The end result is the lack of transparency, wide variation in performance claims for nearly identical systems, and impossibility of making a comparison among different claims on a truly common basis. Recently, Field and Brasington [12] identified the same problem and proposed an Aspen Plus “baseline” model, which is made available to the public in its entirety for in-depth review and evaluation.

A simplified model combining key IGCC subsystems in a coherent manner, which is amenable to a parametric evaluation to highlight the performance characteristics of most (if not all) IGCC plant variants, is essential. Such a simplified mathematical model of the IGCC plant is described at a high level by Pruschek [13]. The structure is similar to what is proposed herein; however, no details on model implementation were provided. A similar performance calculation structure (including cost estimating models) was provided by Rubin et al. [14] along with all requisite formulae with the exception of steam turbine (ST) performance calculation. Zhu and Frey [15] proposed a simple parametric model for the CC power block with focus on the GT.

This paper describes a simplified parametric approach to the performance calculation of an IGCC power plant. The resulting model is intended for the rapid and reasonably accurate conversion of a known GT performance (e.g., from a detailed, rigorous model) into overall plant performance. The plant performance is expressed in terms of net plant output and efficiency. The model is not intended for detailed design or off-design calculations. It is intended for the following applications:

• competitive syngas-fired GT performance evaluation and comparison on a common overall plant basis

• syngas-fired GT design space exploration (i.e., compressor pressure ratio, PR, firing temperature, amount of air extraction, diluent injection, and others) on an IGCC basis

• syngas-fired GT performance optimization in a limited design space covering gasification, gas cleanup, and oxygen production technologies with different feedstocks

• sensitivity and/or uncertainty analysis using Crystal Ball [16] Monte Carlo simulations

The model assumes that detailed and rigorous GT design performance data are available, either in the form of fixed inputs or a “live” model from a commercially available software package or an in-house code. Using several key GT design data such as exhaust conditions, syngas heat consumption, etc., remaining plant performance drivers are estimated via simple ratios and transfer functions. The key philosophy underlying the simple modeling approach, described below in detail, is that, from a practical GT-centric point of view, the IGCC plant is simply a gas-fired GTCC power plant with a “giant fuel gas skid.” (The GTCC power plant and the “fuel skid” are commonly referred to as power and process blocks; e.g., see Fig. 1.)

In order to build the model in a logical and coherent manner and use it appropriately for the aforementioned purposes, one should be able to identify, quantify, and/or estimate the key subsystems of the process block and their functionalities. From a pure performance perspective (i.e., generation of net electric power and the associated thermal efficiency or, equivalently, fossil fuel consumption), the process block of an IGCC gas turbine has four key attributes:

• consumption of solid or liquid feedstock (e.g., coal, petcoke, oil shale, etc.)

• consumption of electric power, which is debited to the gross (generator) output of plant prime movers

• net consumption and/or generation of useful thermal energy, which, in the form of steam, is utilized by the steam turbine (ST) of the power plant for electric power generation

• generation of clean, gaseous fuel (which is referred to as synthesis gas or, in short, syngas)

The process block has three major components, which are bona fide plants in their own right:

• gasification plant (O₂ or air-blown; includes syngas cooler)

• gas cleanup plant

• oxygen plant (if the gasifier is O₂-blown) or air separation unit (ASU)

The gasification plant (GP), as shown in Fig. 1, has two subsystems:

1. (1)

   gasifier

2. (2)

   high-temperature gas cooling (HTGC)

   1. (a)

      radiant syngas cooler (RSC)

   2. (b)

      quench

Strictly speaking, the next three subsystems are parts of the gas cleanup plant (GCP):

1. (1)

   scrubber

2. (2)

   CO shift reactor

3. (3)

   low-temperature gas cooling (LTGC)

From a conceptual perspective, however, it is more convenient (at least in the opinion of the authors) to consider them an integral part of an extended GP. The gasifier converts the solid or liquid hydrocarbon feedstock to a gaseous mixture via partial oxidation. As such, the main product of this extended GP is production of raw syngas (SG) at a temperature suitably low for the GCP. The gasification process takes place at temperatures of 1200–1400 °C (2200–2600 °F) and pressures of 30–60 bar (435–870 psi). The resulting gas contains CO, H₂, CH₄, and other fuel constituents, which can be referred to as the desirable products (i.e., suitable to be used as GT fuels) in addition to the undesirable products such as CO₂, H₂S, NH₃, particulate matter (PM), and neutral products such as N₂ and H₂O vapor.

If the raw SG from the gasifier was not so hot and so “dirty,” it could be directly used in the GT combustor and one would not be compelled to write the current paper. Alas, environmental regulations and even more stringent GT hardware design and life requirements preclude the utilization of the raw SG as turbine fuel. Thus, it is first cooled to a more reasonable temperature either in the RSC (by making steam at a high pressure) or via quench and “scrubbed” with water to remove particulates. Scrubbing also removes some NH₃ and further cools the syngas to about 200 °C. (This is the so-called cold gas clean-up (CGCU) process. Alternative hot gas clean-up (HGCU) processes have been investigated in the past for their potential benefits (lower capital cost and higher efficiency) but they did not overcome their technical difficulties. For the most common variant of gasification technology for IGCC applications, slagging entrained-bed systems, the efficiency benefit was about 1 ppt [17].)

If the IGCC power plant is designed for CCS, the next step is the so-called “sour shift” process, where a catalyst-filled reactor converts a significant portion of the CO in the raw SG to CO₂. This reaction is exothermic and typically generates intermediate pressure (IP) steam, which is exported to the power block for power generation in the ST. It also increases the syngas H₂/CO ratio from less than 1.0 (for most gasifiers) to 2.0 or even 3.0 (suitable for substitute natural gas (SNG) production). In plants without CCS, a COS hydrolysis reactor is required to reduce the COS content for easier sulfur removal. In plants with CCS, this conversion takes place simultaneously with the shift reaction and no separate hydrolysis reactor is needed.

Whether CO shift and/or COS hydrolysis reactors are present or not, the raw SG is still too hot for the acid gas removal (AGR) process. In the LTGC section, still hot raw syngas is cooled down to a temperature suitable to the cleanup process (e.g., about 38 °C). A significant part of the heat recovered during the cooling process is utilized to generate low pressure (LP) or IP steam and sent to the power block (parts might be utilized in the process block).

While there is a variety of processes for AGR, either commercially available or under development, the most common technique involves absorption into a liquid with chemical (e.g., methyl diethanolamine (MDEA), etc.) or physical (e.g., Selexol, Rectisol, etc.) solvents. The selection of the optimal AGR technology in the IGCC is controlled by factors such as gas flow rate, concentration of acid gases, and the need for removal of CO₂ (i.e., a plant to be built with or retrofitted to CCS) as well as H₂S. Detailed cost/performance tradeoff is required for a final choice. However, the most likely choice for the high-pressure systems is a physical solvent process. Whatever the chosen technology, the process involves absorption (removal of acid gases from the syngas into the liquid absorbent) and solvent regeneration (removal of the acid gases from the liquid absorbent). The latter process requires significant energy, which is typically supplied by LP steam extraction from the CC bottoming cycle for the H₂S/CO₂ rich solvent. In the plants with CCS, staged flashing is the technique for regeneration of the CO₂ rich solvent from the CO₂ absorber, which is the second stage of the AGR. The primary unit system in this paper is the U.S. customary system, which is the system used in the authors' company. Equivalent SI unit values are provided in the text. Common conversion factors are provided in the nomenclature where they first appear. The equations are listed in their original form to prevent confusion and errors that may result from conversion to SI units (notwithstanding the needless complexity introduced by conversion factors). Therefore, when substituting parameters into the equations, one should use their original, i.e., U.S. customary, units.

The model is anchored to the gas turbine, i.e., performance, number of units in the plant (usually one or two), and the clean syngas heat input at the GT fuel skid inlet. If the syngas information is not available or if the model is intended for a generic study and/or “back of the envelope” type studies, reasonable estimates can be made for a given gasification technology. Representative SG properties for different gasifiers and feedstocks are provided in the literature [18,19].

Detailed calculation of the gas turbine should provide the key information such as output, syngas mass flow rate, exhaust flow, composition and temperature, air extraction, and diluent nitrogen flows (if applicable). For a plant-level performance, gasifier feedstock flow and heating value (lower/net, LHV, or higher/gross, HHV) are also required. Total plant heat consumption (HC) is the product of the two. Secondary information includes feedstock sulfur content (% by mass) and carbon capture effectiveness (as a %), which are used in the model for adjustment of key ratios. Finally, steam (import) and boiler feed water (export) properties are needed to estimate the power block ST power generation due to steam imported from the process block heat exchangers.

Key contributors to this item are dependent on the type of gasifier and associated feed system (e.g., “dry” versus “wet” or slurry systems). In wet systems, solid feedstock is preprocessed into slurry by fine grinding and water addition, which is pumped into the burner. In dry systems, pulverized and dried feedstock is pressurized in lock hoppers and fed into the gasifier with a transport gas by dense-phase conveying. Depending on the particular system, individual power consumers include coal grinding and slurry preparation, slurry feed pump, slag crusher, slag handling, etc. A reasonably good value for α_GSF is 2.0 kW/STPD, which is adequate for auxiliary power consumption bookkeeping purposes.

(Note that the contribution of natural gas (NG) or another fuel burned in the Claus unit furnace or the coal dryer (if present, e.g., as in Puertollano [10]) to the denominator of Eq. (5) is ignored.) Proper efficiency definition for coal-fired power plants is extremely important. Herein, per Eq. (5), IGCC net efficiency is defined on an HHV as received (ar) basis, which includes the moisture and ash content of the feedstock. In almost all practical applications, pure electric power generation or combined heat and power (CHP), no use can be made of the latent heat contained in the water vapor in combustion products. Thus, one can argue that using LHV (ar) in the denominator of Eq. (5) presents a more realistic efficiency number (even though the feedstock is purchased on an HHV basis). Another source of confusion stems from the widely used solid feedstock heating value reporting on a moisture and ash free (maf) or dry and ash free (daf) basis, which excludes the moisture and ash content of the fuel in per unit mass heating value definition. For good coals with low moisture and ash content, this does not introduce a significant error (i.e., the denominator of the fraction in Eq. (5) is approximately constant). However, for poor coals such as lignite with very high moisture content, especially for the dry-fed gasification systems, this can be a source of large error and/or confusion. In fact, with certain coal drying systems making use of the vapor released during the coal drying process, LHV-based efficiency definition can be misleading [21]. In order to prevent any confusion and unnecessary clarification, all formulations in this paper are based on as received feedstock heating values and flows.

This equation is very well suited to CC analysis of air-cooled GTs with the implicit assumption of a well-designed Rankine bottoming cycle based on today's most optimistic state-of-the-art (SOA) technology and aggressive equipment design specifications. Thus, it represents three-pressure reheat (3PRH) systems based on advanced F-class gas turbines with exhaust temperatures between 590 °C and 670 °C and heat recovery steam generator (HRSG) stack temperatures of about 82 °C. Typical IGCC bottoming cycles for gasifiers with RSC are two-pressure reheat (2PRH) systems with HRSG stack temperatures of about 250 °F. The reason for only two pressure levels is the large high pressure (HP) economizing and superheating duties in the HRSG associated with the RSC. This leaves relatively little exhaust gas energy for IP steam production (maybe for a few thousand pounds per hour). Economically, it makes sense not to build an IP section (superheater and evaporator) for that meager IP steam production opportunity. The high exhaust gas stack temperature is a direct result of the high-temperature condensate (∼230 °F) coming from the plant deaerator (to prevent sulfuric acid condensation on economizer tubes). Therefore, application of a reduction of 8 ppt to the SOA RBC exergetic efficiency is proposed to establish a suitable basis for IGCC applications. This should return a value of ∼65–68% at the GT exhaust temperatures typical for syngas-fired units.

This generic formulation assumes that there are several different net steam import (usually only two) streams from various syngas coolers at different pressures. Note that the “minus” sign is for the net steam export from the power block bottoming cycle. One example of a significant exergy export from the bottoming cycle is GT diluent steam injection. Typically, hot reheat or IP steam at 23 bar (absolute) and 345 °C is used as a diluent in the GT combustor for NO_x abatement. As can be seen from the above formula, this is detrimental to the IGCC performance due to the lost ST work. While less steam is needed than N₂ for the same level of NO_x reduction (about one-third of N₂ on a mass flow basis), in general, utilizing N₂ as a diluent is more advantageous in IGCC systems.

Gasifiers are classified based on their reactor types (i.e., moving or fluidized bed, entrained flow), gasifying medium (e.g., steam and air, steam and oxygen), and reaction temperatures (i.e., high temperature “slagging” or low temperature “dry ash”). Key characteristics of existing gasification technologies are given in Refs. [18,19]. For a comparison of oxygen and air-blown gasification systems with a GT perspective, a recent paper by Parulekar [27] is recommended. The simplified model developed herein is exclusively for electric power generation. As such, due to its dominance in the existing plants [1], only the entrained flow technology is considered. (One can certainly develop a parameter group pertinent to a particular gasification technology for specific studies.) From a purely performance impact perspective, differentiating features of major entrained flow gasification systems are type of coal feed, syngas cooling, and reactant, which determine the following key model parameters: (i) cold gas efficiency (CGE), (ii) gasification sensible heat recovery, and (iii) ASU power consumption, which is a strong function of the degree of integration with the GT.

The composition and heating value of the clean syngas are of prime importance to the GT combustor and turbine design. Typical values are cited in the literature [18,19]. The H₂/CO ratio is normally around 0.5 for dry-fed gasifiers and around 1.0 for the slurry-fed gasifiers (about 0.3 for the air-blown gasifier). In terms of the raw syngas at the scrubber inlet, syngas from the dry-fed gasifier has very low moisture, e.g., less than 5% (vol) vis-à-vis slurry-fed gasifier raw syngas with 20+% (vol). This is a disadvantage for the dry-fed systems with CCS due to the large steam need (about 2 for minimum steam/carbon ratio) for the water gas shift (WGS) reaction, which is diverted from the power block at the expense of ST power output.

The key gasifier metric of interest is the CGE. This is the ratio of the heating value of the clean syngas consumed by the GT to the heating value of feedstock consumed by the gasifier (i.e., IGCC plant heat consumption). In terms of absolute values, it is difficult to make a reliable assessment. Gasifier cold gas efficiency is a strong function of the coal type (HHV, sulfur content, etc.), feedstock delivery system, gasifier pressure, and other design features such as syngas or CO₂ recycle. For the entrained flow gasifiers with high gasification temperatures (∼1100 °C or higher), the energy supplied to the gasifier (mostly feedstock enthalpy, i.e., heating value) is roughly distributed as follows:

• raw syngas heating value (∼ 85%)

• sensible heat in the raw syngas (12–13%)

• heat losses and slag (2–3%)

Note that there is no universal agreement on the definition of $ηCG$⁠. In this discussion, $ηCG$ is defined with two key assumptions:

1. (1)

   The energy basis for syngas and feedstock is LHV.

2. (2)

   The numerator of the efficiency formula Eqs. (20) and (21) is clean syngas to the GT.

where ϖ′ is the HHV to LHV ratio of the clean syngas and ρ is the ratio of the total HHV energy content of the raw syngas stream to the clean syngas stream (typically, 1.07 with no CCS and 1.11 with 90% CCS). The energy content ratio ρ depends on different factors such as the gasifier feedstock's sulfur content (i.e., raw syngas H₂S content), gas cleanup process and effectiveness (i.e., with or without carbon capture), whether the clean syngas is moisturized or not. These are factors that impact the gas flow rate and heat content (note that H₂S has an LHV of 15,235 kJ/kg). In general, a good value for non-CCS and CCS systems is 1.02 and 1.10, respectively. Thus; for example, if a gasifier is quoted with 80% $η'CG$ with a bituminous coal feedstock (ϖ is 1.05), the corresponding $ηCG$ to be used in the current model is:

• 80%·1.05/(1.02·1.07) = 77.0% for a system with no CCS, and

• 80%·1.05/(1.10·1.11) = 68.8% for a system with 90% CCS

Thus, a reduction in CGE from its base value is to be expected when CCS is introduced. The actual value is dependent on the specific gasification and gas cleanup system (typically, 5 to 9 ppt).

This steam is utilized in the bottoming cycle of the CC plant to generate electric power. The power contribution can be estimated via Eq. (14) with $Q·GSF$ and $T¯stm$⁠, which is evaluated from the HRSG feed water supply and RSC steam return (usually saturated at 115–140 barg) conditions using Eq. (16). Feed water conditions (if not known) can be estimated using a 15 bar adder (to saturated return steam pressure) and 20 deg subcool. Depending on the gasification and gas cooling technology, a second (but much smaller) contribution to the power block ST output can be estimated from the LP (or IP) steam produced in LTGC heat exchangers. The net heat import and the ST power contribution can be evaluated in a manner similar to that for the gasifier via Eqs. (24). and (14) by replacing $θGSF$ with $θGC$⁠. Typical values for non-CCS systems are provided in Table 1.

The exact amount and quality of steam production via syngas cooling in multiple locations (e.g., gasifier, LTGC, gas shift reactor, etc.) are subject to system-level optimization (e.g., cost-performance tradeoff), heat exchanger design considerations such as fouling, and available power block hardware to make use of generated steam. (Note that heat import/export via condensate or HRSG feed water heating, heat rejection to plant cooling water, etc. are ignored due to their low exergy content.) Ideally, the values of $θGSF$ and/or $θGC$ should be determined from available information (e.g., technical papers, reports, published articles, or a detailed system model) on the particular gasification technology.

Note that there are two other uses for the raw gas sensible heat: clean syngas fuel heating and moisturization. (A third option is diluent nitrogen moisturization.) For suitably high gasifier pressures (e.g., 60–70 bar or higher), an SG expander can be considered in lieu of heat recovery. The numbers in Table 1 are commensurate with HTGC and LTGC heat recovery for dry syngas fuel at 175 °C to 200 °C at the GT fuel skid inlet with no SG expander. Fuel or diluent nitrogen (if available) moisturization is an effective use of low-grade waste heat for improved system efficiency and reduced combustor flame temperature to control NO_x emissions [28]. Adjustments to the heat recovery parameters listed in Table 1 can be made to account for SG expander, fuel moisturization and/or fuel heating to higher temperatures. This will be discussed later in the paper.

Steam (usually HP) is supplied to the WGS reactor in CCS systems. Especially dry-fed gasifiers require significant steam import from the power block because the raw syngas has insufficient H₂O vis-à-vis slurry-fed gasifiers. The exothermic shift reaction transfers the fuel heating value from CO to H₂ and converts the carbon from CO to CO₂. Heat generated during the exothermic shift reaction is recovered for IP or LP steam generation to be utilized in the ST. The net steam import resulting from these two mechanisms should be debited to the ST power generation in a manner similar to that for GT diluent steam injection.

Coals are ranked based on their energy and moisture/ash content. About 80% of the world's coal for power generation is bituminous or subbituminous. It is crucial to know the feedstock type and rank when using available gasifier data for calibration of the model parameters discussed herein. In general (but not always), especially for IGCC power generation applications, gasifier performance is quoted for a bituminous coal feedstock. However, lower rank subbituminous and lignite constitute nearly 50% of world coal reserves. The impact of coal quality on IGCC performance is strongest for slurry-fed gasifiers showing the largest variation in energy density of the gasifier feedstock and oxygen demand [21,29]. However, CCS performance loss in slurry-fed systems with quench is much less than in dry-fed systems due to the elimination of steam import (at the expense of ST output) for the shift reaction, i.e., about 4–6 ppt in net IGCC efficiency versus 8–10 ppt, respectively. Typical raw and clean syngas composition for a given gasifier and/or coal type can be obtained from the literature. For correction of a given performance for a different feedstock, data in Table 4 of Ref. [29] can be used.

The value of o is a feature of the gasifier technology and the particular feedstock. In general, for the same type of coal, slurry-fed gasifiers require more O₂ than the dry-fed ones since more heat is needed to vaporize all the water in the slurry. Typical values of o range between 0.7 and 1.2; for quick estimations a good default value is 0.8 for dry-fed and 1.0 for slurry-fed systems. (This parameter is subject to confusion and errors due to the definition of the oxygen and feedstock flows. Herein, the assumption is that (i) O₂ from the ASU is 95% pure and (ii) feedstock is on as received basis. When calibrating the model to published data care must be given to the definitions used by respective authors.)

ASU power consumption is the sum total of the power consumption of three compressors: main air compressor (MAC), oxygen compressor, and diluent nitrogen compressor. In general, it is a function of the ASU cold box pressure, gasifier pressure (to which the O₂ is compressed), GT pressure ratio, the diluent N₂ consumption of the GT (for the diluent N₂ compressor), and the GT air extraction (which will reduce the power consumption of the MAC). The ASU is the only process block system that has significant interaction with the GT via air extraction and diluent nitrogen streams and can significantly impact the GT and plant performance. Adjustments to the ASU power can be made for specified air extraction (AE) and diluent injection by simple ratios. A reasonable value for α_ASU is 250 kW per kpph of O₂ (∼2 MJ/kg) with no GT air extraction (i.e., all ASU air is supplied by the MAC) and N₂ sent to the GT as a diluent. This assumes a reasonably state-of-the-art compressor with 88% polytropic efficiency, 95% mechanical/electric efficiency, and an F-class GT with nominal compressor PR of 17–18. The other important consideration that goes into this number is related to the individual ASU compressor pressure ratios, which are functions of the ASU pressure, N₂ and O₂ pressures at the inlet to their respective compressors, gasifier pressure, and GT compressor PR. The number cited above (i.e., 250 kW/kpph = 2 MJ/kg) is based on the following assumptions:

• ASU pressure of 5 bar (75 psia)

• gasifier pressure of 45 bar (650 psia)

• N₂ and O₂ pressures equal to 90% of ASU pressure

• 95% O₂ purity

• all N₂ injected to the GT combustor (SOA F-class)

Typically, for systems without CCS, 100% of nitrogen from the ASU is about 1.2 to 1.4 in terms of diluent-to-fuel ratio, which corresponds to 2.6 to 3.1 in d. Air extraction cannot exceed about 15% to 20% of the GT compressor airflow. (For O₂-blown gasification systems, 10% AE would replace about 40% of the MAC airflow.) Previous studies and past experience have shown that the best GT-ASU integration strategy is a combination of air extraction and injection of all N₂ from the ASU [30]. This ensures the lowest NO_x emissions with maximum possible turbine output (to the extent hardware limits are not exceeded).

IGCC plants with air-blown gasifiers also include a small ASU to supply pressurized nitrogen, which is used for coal transport. Oxygen generated by the ASU is mixed with air extracted from the GT (100% of the oxidant) to the gasifier [27]. Power consumption is about 15% of the ASU for the same size of oxygen-blown gasifier per Eq. (27).

From an overall IGCC plant performance perspective, the gas cleanup plant (GCP) is a consumer of electric power (i.e., pumps and compressors) and heat energy (i.e., steam from the power block and/or various syngas coolers in the process block). In that sense, the following subsystems are of importance:

1. (1)

   acid gas removal (AGR)

   1. (a)

      H₂S removal

   2. (b)

      CO₂ removal or capture (with CCS)

2. (2)

   CO₂ compression (with CCS)

3. (3)

   Claus plant (sulfur recovery)

4. (4)

   tail gas treatment unit (TGTU)

As far as H₂S removal is concerned, the key performance metric is pumping power consumed per unit flow of raw or clean SG (or acid gas removed). The heat energy is usually small (68 MJ per kmol of acid gas removed for estimating purposes [18]) and of low exergy, i.e., the equivalent of a few megawatts of ST power output. While it is relatively easy to keep track of this ST power debit due to AGR regenerator steam load, ignoring it does not introduce a big error into the model within the framework of interest herein. (The values for θ_GC listed in Table 1 are net of the AGR steam load.) However, heat energy (i.e., steam) consumption can be significant for the CCS systems and should be tracked separately. Requisite adjustments are discussed in the following section.

A reasonably good value for α_H2S (Selexol-based system) is 4.0 kW/kpph (∼32 kJ/kg). This is for a typical U.S. bituminous coal from Illinois or Appalachia with sulfur content (by mass) of about 3%.

A reasonable default value for α_CO2 is 80 kW/pps (175 kJ/kg) for the solvent based capture unit via chemical absorption (e.g., see Hoffmann et al. [31]). Rigorous models of the actual system, based on chemical or physical solvents, can provide a more accurate number, which can be substituted (if available) but is unlikely to change the final answer by much. Thermal energy consumption by the stripper reboiler is about 600 kW/pps (usually low pressure steam) [31]. No separate accounting is made for the thermal energy. In the absence of specific information, using the values listed in Table 1, the following rules of thumb are recommended:

• Dry-fed gasifier: Set θ_GSF to zero, multiply non-CCS value of θ_GC by 3.5.

• Slurry-fed gasifier: Set θ_GSF to two-thirds of the non-CCS value, multiply non-CCS value of θ_GC by 2.5.

The value of CO₂ compressor power consumption and; consequently, α_CO2C is a function of storage pressure (typically ∼150 bar), the distance of storage location to the plant (which determines the pipeline and associated pressure loss allowance) and the compression system. For a typical application with intercooled gas compressors, a good value for α_CO2C, representing the SOA, is 105 kW per pps of CO₂ flow (∼230 kJ/kg). This assumes 85% stage efficiency for the compressor with three flashing pressure levels. This is adequate for auxiliary power consumption bookkeeping purposes, which is essentially insensitive to gas turbine optimization. For a single flash level, a good value is 135 kW/pps (∼300 kJ/kg). There are studies showing 20% reduction CO₂ compression power if the CO₂ is liquefied during the compression process and then pumped to the final pressure as a liquid [32]. Even further reduction is possible via refrigeration during liquefaction but the benefit in parasitic loss saving is offset by the lost ST power as a result of steam extraction required to drive the refrigeration cycle.

Utilizing the low-grade waste heat to heat the GT fuel is a standard CC feature to improve efficiency. Advanced F-class units in natural gas (NG) fired applications typically utilize intermediate pressure economizer feed water from the HRSG to heat the fuel gas to about 200 °C (about 400 °F). In IGCC plants, there is enough waste heat in the HTGC and LTGC sections to accomplish this level of fuel heating. Higher temperatures are possible, both in NG-fired CC and IGCC power plants, up to ∼540 °C (1000 °F) or even higher in the latter. However, studies have shown that a point of diminishing returns is reached at around 315 °C, beyond which no significant performance improvement is achieved to warrant the cost and complications associated with materials and design of piping, performance heater and GT fuel skid systems [33].

In Eq. (40), μ_SG is the clean syngas moisture content by volume at the exit of the fuel gas saturator and θ_SAT is the net heat required to moisturize the fuel gas as a fraction of the gasifier feedstock heat input (LHV). In a study involving fuel gas moisturization, θ_SAT from Eq. (40) should be subtracted from the θ_GC listed in Table 1 (unless air extracted from the GT can be utilized for the same purpose). Note that part of the moisturization water can be extracted from the HRSG downstream of the low pressure evaporator and making use of waste heat, which otherwise would be lost through the stack (i.e., no exergy penalty). This is limited by the allowable HRSG stack temperature to prevent sulfuric acid condensation on economizer tubes.

This level of fuel heating would require high pressure feed water from the HRSG or HP steam from the gasifier syngas coolers (e.g., θ_SAT is about 5% for 20% fuel moisture). Moisturization heat duty given by Eq. (40) should be debited to θ_GC. The remainder (if any) and the difference in θ_SAT between Eq. (41) and Eq. (40) should be debited to θ_GSF.

Air extracted from the GT (at more than 370 °C) is a high grade heat source. In IGCC plants with ASU-GT integration, it is almost always utilized to heat the diluent N₂ from the ASU. Remaining heat can be utilized to heat the fuel gas or nitrogen saturator circulating water, other feed water or for partial heating of the fuel gas. Possibilities are myriad and highly dependent on the GT and ASU N₂ compressor designs, which determines extraction air, diluent N₂, and moisturized SG flows and temperatures. Based on the GT simulation output, requisite adjustments can be easily made to the heat recovery parameters.

The parameters to be used in the simple IGCC model are listed in Table 2 for quick reference. In the absence of specific information pertaining to particular gasification and/or gas cleanup technologies, the suggested default values can be used. This should suffice for studies focusing on GT design modifications to adopt a natural-gas-fired unit for generic syngas applications. For more in-depth studies, especially those with novel process block technologies, the recommendation is to develop a specific parameter list from available data (e.g. from published studies, detailed process model runs, etc.).

Sensitivity of IGCC performance to selected model parameters in Table 2 is summarized in Table 3. The data in Table 3 confirms the well-known recipe to IGCC performance improvement, namely that GT is the key driver of performance and closely followed by gasifier CGE and ASU parasitic power consumption.

The simple IGCC model described in the preceding section is intended for specific applications outlined in the Introduction. It is not a bona fide design tool and, as such, it cannot be used to investigate the intricate design details of various IGCC subsystems. The model is primarily geared to engineers involved in the design of GTs for electric power generation. Typically, gas turbines are not designed from a blank sheet for IGCC or other low calorific value gases such as steel mill applications (where the GT fuel is a blend of blast furnace and coke oven gases). Gas turbines designed for natural gas fuel (mostly methane) are modified to fit the specific application (hardware as well as controls). From electric power generation perspective, the primary interest is in the following areas in terms of their impact on IGCC performance (output and efficiency):

• GT design and optimization (mainly firing temperature and PR)

• GT-ASU integration

• gasification technology (for a given GT)

• CCS

• gasifier feedstock (for a given GT and gasifier)

GT design optimization and integration with ASU impact the process block via requisite air extraction and/or diluent injection and to a lesser extent by the quantity of syngas. As far as the first four items are concerned, the model is expected to reproduce the qualitative performance trends dictated by the fundamental characteristics of the pertinent processes. In order to ascertain this capability, selected data from the published sources are compared with the model predictions (Refs. [4,10,35–45]). The IGCC plant data used herein cover the following range:

• quench and heat recovery (RSC) type gasifiers

• oxygen and air-blown gasifiers

• plants with and without CCS

• zero to ∼20% GT air extraction

• zero to full diluent N₂ injection

• diluent steam injection

• bituminous feedstock

Unfortunately, no plant data with enough detail could be found for low rank (e.g., lignite) coals. A few published plant data with low rank coals (also several others one should add) turned out to be severely inconsistent and deemed to be unreliable. In fact, this should be considered another confirmation of the usefulness of the model presented herein (i.e., weeding out unreliable performance data). Such cases are left out from the study because it is difficult to judge whether the inconsistencies are due to typos, unreasonable assumptions, or other types of inadvertent mistakes.

The model is applied with the default parameters summarized in Table 2 with the exception of the CGE, which is calculated from the published data. To the extent possible, per availability of detailed system data, requisite adjustments for SG expander, GT steam injection, and SG humidification are made. No system-specific adjustments are made beyond what is described in the main body of the paper. Only in some cases, where discrepancy was simply too big, significant effort has been made to tweak the parameters (strictly per published information) to close the gap between the model and published performance. In some instances, this proved unsuccessful and, as mentioned earlier, those cases with inexplicable inconsistencies have been left out from the comparison.

Similarly, published GT data from the listed references are used in the model. However, reasonable assumptions (for missing information) and corrections are made to ensure heat and mass balance integrity of the GT control volume. As highlighted by the sensitivity data in Table 3, apart from the GT, the two critical subsystems are the ASU and the ST. Model predictions of the ASU power consumption (via Eqs. (25)–(27)) and ST power generation (via Eqs. (8)–(19) and (24) and the data in Table 1) are shown in Figs. 2 and 3. Model predictions of the IGCC net output and efficiency are shown in Figs. 4 and 5. Absolute average difference between the model predictions and the published data are listed in Table 4.

The comparison summarized in Figs. 2–5 and Table 4 suggests that the model is adequate for high-level and consistent comparison of published performance data for a wide variety of IGCC systems on a common basis. The biggest discrepancy is introduced by the plant auxiliary power consumption, which can be traced to the wide variation in auxiliary load scope assumptions and optimistic boundary conditions (e.g., aggressively low steam condenser pressure, feed pump sizing, etc.). It is, therefore, recommended to compare published IGCC performances on an adjusted gross basis, including only the prime movers, SG expanders and the ASU (if present). Standard corrections for ambient conditions and the steam condenser pressure can be made using suitable correction curves.

A simple IGCC performance model is developed to facilitate syngas fired GT performance optimization and integration studies for IGCC power plants. The model contains simple parametric equations derived from the first principles and can be easily programmed into a spreadsheet. The model is also suitable to evaluation and comparison of published performance data on a common basis. Comparison with published data for a wide range of IGCC plants with and without CCS shows that, even with default parameters, the model is capable to reproduce the plant performance with reasonable error. This ability can be further enhanced by developing system specific parameter sets (for different gasification and gas cleanup technologies) for higher fidelity. The model can also be expanded to incorporate secondary IGCC plant features such as CO₂ recycle compressor or diluent nitrogen moisturization.

The authors would like to thank GE Energy for permission to publish.

Nomenclature

c_p =

constant-pressure specific heat (Btu/lb-R, 4.1868 kJ/kg-K)

d =

diluent nitrogen-to-oxygen ratio

e =

specific exergy (Btu/lb, 2.326 kJ/kg)

$E·$ =

total rate of exergy transfer (Btu/s or kW, 1.05506 Btu/s)

h =

specific enthalpy (Btu/lb)

$m·$ =

mass flow rate per unit or train (pps, 0.4536 kg/s, or kpph, 453.6 kg/h)

$M·$ =

total mass flow rate for all units or trains in the plant (pps or kpph)

N =

number of units or trains

p =

pressure (psi, 0.06895 bar)

ppt =

percentage points

$Q·$ =

rate of heat transfer (Btu/s or kW)

s =

specific entropy (Btu/lb-R)

S =

coal sulfur by weight (%)

T =

temperature (°F,  °C = [°F – 32]/1.8)

$T¯$ =

mean-effective temperature (°F)

$W·$ =

power (Btu/s or kW)

y_H2O =

GT exhaust gas moisture fraction (molar basis)

Greek Symbols

α_ASU =

oxygen plant power consumption per unit O₂ flow rate (kW/kpph)

α_CO2 =

carbon capture unit power consumption per unit captured CO₂ flow rate (kW/pps = 2.20462 kJ/kg)

α_CO2C =

CO₂ compressor power consumption per unit captured CO₂ flow rate (kW/pps)

α_GSF =

gasifier power consumption per unit feedstock flow rate (kW/STPD, 1 STPD is 2000 lbs or 907.2 kg per day)

α_H2S =

AGR power consumption per unit clean syngas flow rate, total for all GTs (kW/kpph)

$χCO2$ =

CO₂ flow rate (fraction of the total raw syngas flow rate)

ε =

bottoming cycle exergetic efficiency

ε′ =

exergetic conversion effectiveness

η =

efficiency or effectiveness

λ_fp =

boiler feed pump power as a fraction of ST power

μ =

moisture fraction (by volume)

o =

gasifier oxygen to feedstock mass flow ratio

π =

pressure ratio

θ =

heat duty as a fraction of gasifier feedstock input (LHV)

ρ =

raw-to-clean syngas HHV energy content ratio

ϖ =

gasifier feedstock higher to lower heating value ratio

ϖ′ =

clean syngas higher to lower heating value ratio

σ_SG =

clean-to-raw syngas mass flow ratio

ξ =

air extraction as a fraction of compressor airflow

Subscripts

AE =

air extraction

AUX =

auxiliary

BC =

bottoming cycle

Cap =

carbon capture

CC =

combined cycle

CG =

cold gas

exh =

GT exhaust

f =

saturated water

feed =

gasifier feedstock (coal, petcoke, etc.)

fw =

feed water

GC =

gas cooling

GCP =

gas cleanup plant

GSF =

gasifier

GT =

gas turbine

imp =

import (steam from process to power block)

RBC =

Rankine bottoming cycle

SAT =

saturator (i.e., “moisturizer”)

sat =

saturated (steam/water)

SG =

syngas

SGX =

syngas expander

stck =

heat recovery steam generator stack

stm =

steam