INTERIM FINAL TECHNICAL REPORT

September 1, 1996, through August 31, 1997

Project Title: DEVELOPMENT OF ACTIVATED CHAR FOR COMBINED SO2/NOx REMOVAL

ICCI Project Number: 96-1/2.1A-7

Principal Investigator: Anthony A. Lizzio, ISGS

Other Investigators: Sheila Desai, Gwen A. Murphy, Gwen L. Donnals, Mark P. Cal,

John M. Lytle, ISGS; Bret H. Howard, John L. Haslbeck, NOXSO

Project Manager: Ronald H. Carty, ICCI

ABSTRACT

The NOXSO Corporation has developed under the DOE Clean Coal Technology Program a dry, post-combustion flue gas treatment system that uses a regenerable sorbent in a fluidized bed to remove SO2 and NOx from coal combustion flue gas. The process is especially attractive for use with high sulfur coals because it generates a saleable sulfur by-product, and creates no new waste streams. The cost of the sorbent presently used in the NOXSO process, alumina beads impregnated with 5% sodium, is significantly greater than the projected cost of an activated char recently developed by the ISGS. Thus, an opportunity exists to incorporate ISGS activated char into the NOXSO flue gas treatment process.

The goal of this project, a cooperative effort between the ISGS and NOXSO, was to develop a low cost activated char from Illinois coal that can be used to effectively remove SO2 and NOx from coal combustion flue gas. In the first year of the project, activated chars were prepared from IBC-102 coal in both granular and powdered form. The SO2 capacity of steam activated IBC-102 char subjected to a liquid phase oxidation/thermal desorption treatment was greater than that of a commercial activated carbon (Centaur carbon). The NOx removal capability of this char was an order of magnitude greater than that of the Centaur carbon. Single (SO2 or NOx) and multicomponent (SO2/NOx) removal tests performed by NOXSO showed that this char worked better than the NOXSO sorbent in single component tests, whereas the NOXSO sorbent removed more NOx in multicomponent tests.

This year, combined SO2/NOx removal experiments on IBC-102 char were performed to examine the competitive effects of H2O and SO2 on low temperature NOx removal. The SO2 capacity of the char was not affected to any appreciable extent by NO in the flue gas. Both H2O and SO2 in the flue gas appeared to inhibit NO adsorption by activated char. To remedy this situation, several new new types of chars were prepared and tested with simulated flue gas (2500 ppm SO2, 500 ppm NO, 7% H2O, 5% O2). The best char removed 90% of the NOx for 6 h and 95% of the SO2 for 14 h at a space velocity of 3000 h-1. There is no commercial activated carbon on the market today that comes close to removing these amounts of SO2 and NOx simultaneously from coal combustion flue gas. NOXSO is presently testing this char and comparing its performance to that of the NOXSO sorbent.

Pages 8 through 26 contain proprietary information.

EXECUTIVE SUMMARY

Worldwide interest in carbon-based flue gas desulfurization (FGD) technology is growing and these processes have been proven successful at removing up to 95% of the SO2 and over 80% of the NOx from combustion flue gas. An activated carbon FGD process, typically placed after the precipitator and just before the stack, can be used alone or in conjunction with other methods of FGD to remove SO2/NOx from flue gas. This technology has been used in Europe and Japan for cleanup of flue gas from both coal combustion and waste incineration. Presently, no U.S. utility employs a carbon-based process to clean flue gas. Carbon based FGD systems can be integrated into both new and existing power plants. The retrofit of an existing utility boiler with such a FGD process could, in addition to improving SO2/NOx emissions, lower overall capital and operating costs compared to competitive FGD processes. One of the unique advantages of an activated carbon FGD process is that it removes nearly every impurity found in combustion flue gas including particulates, SO2/NOx, mercury, dioxins, furans, and other air toxics. No other existing FGD process has that capability. There are a number of research groups presently involved in the development of novel carbon based processes and materials for flue gas cleanup. The type of carbon used, more often than not, dictates the economic viability of a given process. A high quality carbon adsorbent for SO2/NOx removal should have a high adsorption capacity for SO2 and NOx, rapid SO2/NOx adsorption kinetics, low reactivity with oxygen, minimal loss of activity after regeneration, high mechanical strength, and low cost.

In September 1993, the ISGS initiated Phase I of a research program (Production and Use of Activated Char for Combined SO2/NOx Removal, Lizzio et al., 1994; 1995) to investigate the possibility of using activated char derived from Illinois coal to clean combustion flue gas. Activated chars were produced from IBC-102 coal under a wide range of pyrolysis and activation conditions and their SO2 adsorption capacities were measured. A novel char preparation method was developed to prepare chars with SO2 capacities significantly greater than those of commercial activated carbons. The work performed in Phase I attracted significant commercial interest, in particular, that of STEAG Aktiengesellschaft, a German firm seeking to market its technology in the U.S. for cleaning flue gas from incinerators. STEAG has pioneered flue gas cleanup in Europe using a low surface area (270 m2/g) activated carbon made from German brown coal. They are presently seeking out suppliers to provide them with a low cost (< $300/ton) activated char for their U.S. market. STEAG estimates a market potential of 80,000 tons/year of activated char (160,000 tons of coal) within five years, assuming 10% of U.S. incinerators adopt their technology, to meet needs emanating from anticipated regulation of emissions from existing incinerators. The ISGS agreed to provide 500 pounds of activated char to STEAG for tests in a demonstration unit to clean flue gas from a U.S. waste incinerator.

As part of Phase I activities, laboratory conditions were identified to produce a suitable low cost (< $350/ton) adsorbent from an Illinois Colchester No. 2 coal. With the assistance of Allis Mineral Systems (Milwaukee, WI), the production steps were carried through two levels of scale up, culminating in the production of 550 pounds of activated char in an 18 in. ID rotary tube kiln. A three step process, which included preoxidation, pyrolysis and activation, was used to produce an activated char with an SO2 adsorption capacity almost twice that of the char presently used by STEAG. The adsorbent was shipped to Germany and tested on a slip stream of flue gas from a commercial incinerator. Results indicated that the char was effective in removing more than 99.7% of the SO2, mercury, dioxins and furans from the incinerator flue gas.

The pilot scale production of low cost activated char by the ISGS attracted the attention of several organizations interested in utilizing inexpensive carbon to clean flue gas. One such company is the NOXSO Corporation which has developed under the U.S. DOE Clean Coal Technology Program a high efficiency, dry, post-combustion flue gas treatment system that uses a regenerable sorbent to simultaneously remove SO2 and NOx from coal combustion flue gas. The process has no impact on boiler performance, is compact, easy to retrofit, generates a saleable by-product (sulfuric acid, elemental sulfur, or liquid SO2), and creates no new waste streams, a distinct advantage over conventional FGD processes. The cost of the sorbent presently used in the NOXSO process, alumina substrate impregnated with 5% sodium, is $2,000/ton as compared to a projected $350/ton for ISGS activated char. The NOXSO process during normal operation for a 500 MW power plant typically consumes 2,000 tons of sorbent per year; thus, there could be a significant savings in operating costs if low cost activated char is utilized.

The goal of the second year project, a joint effort between the ISGS and NOXSO, is to integrate additional SO2/NOx control into the NOXSO process via low cost activated char, thereby creating a more cost effective and comprehensive pollution control system. The project encourages the economical and environmentally acceptable use of Illinois coal in two ways: first, as feedstock in the production of the activated char, and second, by improving the NOXSO process. The goal is to develop an activated char from Illinois coal that can be used in the NOXSO process, either alone or in conjunction with NOXSO sorbent, to effectively remove SO2and NOx and other air toxics from coal combustion flue gas. The project consists of seven tasks. In Task 1, activated chars will be prepared from lllinois coal (IBC-102) in both pellet and powder form; the physical/chemical properties of the chars will be optimized for combined SO2/NOx control. In Task 2, the ISGS will evaluate the combined SO2/NOx removal properties of these chars. In Task 3, selected chars will be further tested with simulated flue gas by NOXSO for SO2/NOx removal, regenerability, fluidization and attrition resistance. In Task 4, the physical/chemical properties of selected chars will be evaluated by NOXSO and ISGS. In Task 5, pound quantities of an optimized char will undergo parametric SO2/NOx removal tests; its performance will be compared to that of the NOXSO sorbent. In Task 6, if a suitable activated char is identified, process flowsheets for the production and use of the char will be developed. In Task 7, technical and management reports will be prepared and submitted to the ICCI.

In the first year of this project, a California pellet mill was used to make 20 pounds of 6 mm diameter cylindrical IBC-102 coal pellets. The pelletized steam activated char had excellent mechanical strength and an SO2 adsorption capacity (120oC) comparable to that of the NOXSO sorbent. To enhance both SO2 and NOx removal capabilities, the pelletized steam activated char was subjected to a liquid phase oxidation/thermal desorption treatment. The SO2 adsorption capacity of this char surpassed that of its granular counterpart and the catalytic carbon (Centaur carbon) recently developed by the Calgon Carbon Corporation. The NOx removal capacity (120oC) of the oxidized/thermally desorbed char was nearly twenty times greater than that of the Centaur carbon. NOXSO performed single and multi component SO2/NOx removal tests on several ISGS activated chars. A simulated flue gas containing either or both SO2 (1,200 ppm) and NO (400 ppm) was used. The oxidized/thermally desorbed char adsorbed more SO2 and NOx than the NOXSO sorbent in single component tests, whereas the NOXSO sorbent adsorbed more NOx in combined SO2/NOx removal tests. For the NOXSO sorbent, the presence of SO2 in the flue gas enhances the removal of NOx. SO2 adsorbs on the alumina substrate and forms a complex with sodium that facilitates the adsorption of NOx. On the other hand, it appears that SO2 and NOx compete for similar sites on the char surface and that SO2 adsorption is stronger. The challenge remains to design a carbon that can adsorb simultaneously both SO2 and NOxfrom coal combustion flue gas .

This year, the combined SO2/NOx removal capabilities of activated chars prepared from Illinois coal were determined. The SO2 capacity of the char was not affected to any appreciable extent by NO in the flue gas. On the other hand, both H2O and SO2 appeared to inhibit NO adsorption by activated char. The introduction of H2O into the flue gas after the char had adsorbed NO for several hours, desorbed a major portion of the adsorbed NOx. The addition of SO2 into the flue gas had a similar effect. To minimize these competitive effects and to gain better insight into the mechanism of combined SO2/NOx removal by carbon, several new types of activated char were prepared and tested with simulated flue gas (2500 ppm SO2, 500 ppm NO, 7% H2O, 5% O2). The following one or more char preparation methods were used: 1) thermal desorption of oxidized char, 2) metal catalyst added to coal by impregnation to incipient wetness or by ion exchange, 3) ammonia treatment, not necessarily in that order. The best char was able to remove 90% of the NOx for 6 h and 95% of the SO2 for 14 h at a space velocity of 3000 h-1, which outperforms by a wide margin all other commercial activated carbons on the market today. This char is now the leading candidate for use in the NOXSO process. NOXSO is presently testing this char and comparing its performance to the NOXSO sorbent. Efforts are also being made to simplify some of the processing steps in order to reduce production cost of the new carbon. Initial attempts at doing this were very successful as two of four processing steps were eliminated, and combined SO2/NOx removal performance was improved. This char is also being tested by NOXSO.

Other key results of this study can be summarized as follows: 1) SO2 adsorption capacity of ammonia treated oxidized char was more than 50% greater than that of Centaur carbon, 2) NOxremoval by thermally desorbed oxidized char was one order of magnitude greater than that of Centaur carbon, 3) H2 treatment of char helps to preserve active sites and increases NOx removal by a factor of two, 4) several new chars were produced that are able to remove both SO2 and NO from simulated flue gas, which was not possible in our previous work, 5) an Illinois coal char produced in 100 pound quantities for control of Hg emissions removed additional amounts of NO only when subjected to certain treatments mentioned above, by itself it could not remove NOx from simulated flue gas, 6) the catalytic effects imparted to the char by the above mentioned treatments seem to be additive.

During the next four months, NOXSO will test 20-100 g samples of optimized char in a fixed- bed adsorber system. The SO2/NOx removal performance of activated char will be compared directly with that of NOXSO sorbent. The fluidization properties and crush strength of the char will also be determined. The potential of activated char in carbon injection-like processes such as the one being developed by NOXSO and FLS milj a/s in Copenhagen, Denmark will also be investigated. NOXSO has the capability to perform large scale tests with activated char in both a 10 in. fluidized-bed adsorber and in a gas suspension adsorber, which simulates a carbon injection process with baghouse and recycle. For these tests, hundred pound quantities of activated char would be needed.

The remainder of this report contains proprietary information and is not available for distribution except to the sponsor(s) of this project.