FINAL TECHNICAL REPORT

September 1, 1997, through August 31, 1998

Project Title: DEVELOPMENT OF NOVEL SORBENTS FOR HOT GAS DESULFURIZATION IN IGCC

ICCI Project Number: 97-1/5.2A-2

Principal Investigator: Javad Abbasian, Institute of Gas Technology

Other Investigators: Rachid B. Slimane, James R. Wangerow, and Minoo K. Zarnegar, Institute of Gas Technology

Project Manager: Ronald H. Carty, ICCI

ABSTRACT

The overall objective of this project was to develop highly reactive and attrition resistant copper oxide- and manganese oxide-based sorbents and to evaluate the effectiveness of the best sorbent formulation in a "life-cycle" test, under process conditions similar to those of the Sierra Pacific (Piñon Pine) Clean Coal Technology IGCC Demonstration Program.

To achieve this objective, a total of 62 formulations were prepared which included 21 copper-based and 41 manganese-based sorbents. Nine (9) copper-based and thirteen (13) manganese-based sorbent formulations that met the minimum physical strength requirement were evaluated for their sulfur removal efficiency and effective capacity in packed-bed tests.

The results indicate that the manganese-based sorbent formulations prepared in this project have excellent reactivities at 550C. However, the reactivities of these sorbents are considerably lower at 450C. Furthermore, a starting temperature of 750C was required to properly regenerate manganese-based sorbents. Three (3) copper-based sorbents were shown to exhibit excellent sulfur removal efficiency and effective capacity at 450C. Based on sulfidation reactivity, regenerability, and attrition resistance, IGTSS-326A was selected as the best copper-based sorbent formulation. IGTSS-326A has extremely high attrition resistance that is 5 times higher than that of a typical FCC catalyst, and 20 times higher than the commercial UCI-4169 zinc titanate sorbent. The long-term durability of the IGTSS-326A sorbent was assessed in a "life-cycle" test comprised of 50 sulfidation/regeneration cycles. The results obtained indicate IGTSS-326A can maintain excellent sulfur removal efficiency (< 1 ppmv H₂S in the cleaned fuel gas) as well as excellent effective sulfur capacity during cycling.

These results strongly suggest that the IGTSS-326A sorbent should be regarded as a strong candidate for demonstration in the Sierra Pacific (Piñon Pine) Clean Coal Technology IGCC Demonstration Program.

EXECUTIVE SUMMARY

Coal gas desulfurization to sufficiently low levels at elevated temperatures (i.e. >350C) has been recognized as crucial to efficient and economic coal utilization in advanced Integrated Gasification Combined Cycle (IGCC) and Integrated Gasification Fuel Cell (IGFC) power generation processes. The implementation of hot coal gas desulfurization technology heavily relies on the development of regenerable sorbents that can efficiently remove hydrogen sulfide (H₂S).

The focus of much of the current research on hot coal-derived fuel gas desulfurization for IGCC processes has been on the use of zinc-based sorbents. Although these sorbents have been the subject of extensive pilot-scale and process development work, zinc-based sorbents have been shown to suffer from sulfate formation and zinc volatilization, leading to sorbent degradation over multicycle use. Therefore, investigation of non-zinc based sorbents, such as copper-based and manganese-based sorbents, is a logical approach to develop an effective alternative to zinc-based sorbents.

In a previous ICCI-funded project, a new class of novel copper-based sorbents was identified and produced. This class of sorbents was shown to have much higher reactivity, effective sulfur capacity, and attrition resistance compared to the commercial zinc titanate sorbents as well as the copper chromite sorbents developed earlier at IGT. Furthermore, unlike most zinc titanates, the reactivity of IGT's copper-based sorbents gradually and consistently improved with cycling.

Another alternative to zinc-based sorbents is manganese-based sorbents, which have recently been shown to withstand high temperature operation and also maintain structural and reactive integrity over many cycles. One of the investigators of this project developed and patented a process for the manufacture of manganese-based pellets with remarkable effectiveness for coal gas desulfurization above 750C. More recently, limited testing at IGT indicated manganese-based sorbents hold considerable promise for lower temperature applications that are of current interest.

The main goal of this project was to produce highly reactive and attrition resistant copper-based and manganese-based sorbents and to evaluate the effectiveness of the best sorbent formulation in a "life-cycle" test, under process conditions similar to those of the Sierra Pacific (Piñon Pine) Clean Coal Technology Program.

In this program a total of forty-one (41) manganese-based and twenty-one (21) copper-based sorbent formulations were prepared. The effects of several parameters on the sulfidation performance, regenerability, and chemical and physical characteristics of the resulting sorbents were determined. The parameters that were investigated included (a) the addition of organic and inorganic binders; (b) the addition of minor amounts of reactive metal oxides; and (c) the variation of the "reactive" and the non-reactive support materials. The effect of sorbent induration (thermal treatment) conditioning on the crush strength or attrition resistance of each sorbent was also determined by employing different temperatures in a practical range for the specified formulation.

A single-hole attrition test unit was used to make a preliminary determination of the relative physical strength of the sorbents. Thirteen (13) Mn-based and nine (9) Cu-based sorbents which met the physical strength requirement were evaluated for their sulfur removal efficiency and effective capacity in packed-bed tests. The physical and chemical characteristics of eight (8) sorbent formulations were also determined.

The results indicate that the majority of manganese-based sorbent formulations prepared in this project have excellent reactivities at 550C, achieving effective sulfur capacities exceeding 20 g S/100 g sorbent. However, the reactivities of these sorbents are considerably lower at 450C, requiring an "activation" step to enhance their reactivities to approach those at 550C. In addition, a starting temperature of 750C was required to properly regenerate manganese-based sorbents in fluidized-bed tests employing dilute O₂-N₂ mixtures.

To lower the required regeneration temperature, a different regeneration route was also investigated which utilized a reducing gas mixture (i.e., H₂-H₂O-N₂) as a regeneration gas in the temperature range of 550 to 700C. The results of these tests indicate that manganese-based sorbents could not be successfully regenerated with the reducing gas mixture in the temperature range investigated. Based on the activation requirement for sulfidation and the high temperature requirement for regeneration, further investigation of manganese-based sorbents was not pursued in this program.

For the copper-based sorbents, nine (9) formulations met the physical strength requirement and were evaluated for their sulfur removal efficiency and effective capacity in packed-bed tests at 450C. The results of packed-bed evaluation indicated that three (3) copper-based sorbents (IGTSS-320A, -322A, and -326A) had sulfidation reactivity, sulfur capacity, and attrition resistance properties that exceeded those of the baseline IGTSS-179 sorbent formulation.

To improve the fluidization properties of the copper-based sorbents, parametric studies were conducted in a fluidized-bed reactor. The parameters studied included the fluidized-bed composition (fraction of inerts), reduction/sulfidation temperature (250 to 650C), superficial gas velocity (20 to 45 cm/s), and the gas composition in both reducing and oxidizing environments. The sorbents used ranged in size between 40 and 180 microns. The results of these parametric studies indicate that the fluidization behavior of the sorbents significantly improves with increasing gas velocities or addition of other materials such zinc-based sorbents or sand.

Following extensive evaluation in packed- and fluidized-bed reactors, the copper-based IGTSS-326A sorbent was selected jointly by IGT and ICCI as the best formulation. A 50-cycle "life-cycle" test series was completed with a mixture containing 50 wt.% IGTSS-326A and 50 wt.% zinc-based material. The goal of these series of tests was to demonstrate the physical durability and reproducibility of the sulfur removal efficiency of this sorbent mixture with cycling. The results obtained indicate this selected formulation maintained excellent sulfur removal efficiency (< 1 ppmv H₂S in the cleaned fuel gas) as well as excellent effective sulfur capacity throughout the 50 sulfidation cycles completed.

The IGTSS-326A sorbent was also shown to have extremely high attrition resistance, as determined in accordance with the ASTM D5757-95 procedure. A comparison of the results obtained in attrition tests conducted with different sorbents indicates that the IGTSS-326A sorbent has an air jet index (AJI) or 5-hour loss that is 5 times lower than that of a typical FCC catalyst, and 20 times lower than the commercial UCI-4169 zinc titanate sorbent.

To assess the cost of the IGTSS-326A fluidized-bed sorbent, several different commercial sorbent manufacturers were contacted, of which the United Catalyst Inc. (UCI) and Contract Materials Processing, Inc. (CMP), were interested in discussing possible future manufacturing of the IGTSS-326A sorbent. According to these sorbent manufacturers, the cost of the sorbent for commercial application will heavily depend on the projected commercial market demand for the product and the timeline for the market consumption of the material. However, given the nature of the metal oxides and the support materials in the IGTSS-326A sorbent, the sorbent cost is expected to be similar to those projected for zinc titanate sorbents.

Although high reactivity, superior attrition resistance, and excellent durability of the IGTSS-326A have been demonstrated in this program, commercialization of the sorbent will require evaluation of the sorbent at the Sierra Pacific Power Company (SPPC) IGCC Clean Coal Demonstration Project (Piñon Pine facility). To move the sorbent towards this demonstration stage, the U.S. DOE, the M. W. Kellogg (MWK), and the Sierra Pacific Power Company (SPPC) require that the sorbent be produced by a commercial sorbent manufacturer and evaluated according to a precise test protocol developed by USDOE/FETC specifically for evaluation of candidate sorbents for Piñon Pine application on a consistent basis.

To satisfy the requirements for commercialization of the sorbent outlined by US DOE/FETC, a two-phase commercialization plan was initiated. A proposal was submitted to DCCA/OCDM to carry out the work required for Phase I, which consists of making the necessary arrangements for production of sufficient quantities of the IGTSS-326A sorbent (about 10 to 25 kg) by a commercial sorbent manufacturer according to IGT and the USDOE specifications and evaluation of the sorbent in the bench-scale fluidized-bed reactor according to USDOE/FETC test protocol. The work in Phase II of the commercialization plan will be carried out through a CRADA arrangement according the US DOE/FETC recommended strategy for " Road Map To Demonstration of Transport Desulfurization Sorbents" in coordination with DCCA/ICCI, USDOE/FETC, M. W. Kellogg Co., and a commercial sorbent manufacturer.