Project Title: ORGANIC SULFUR AND HAP REMOVAL FROM COAL WITH SUBCRITICAL WATER

ICCI Project Number: 95-1/1.1D-1

Principal Investigator: Chris M. Anderson, UND EERC

Other Investigators: Ronald C. Timpe, UND EERC

Project Manager: Ken Ho, ICCI

To date, no economically feasible organic sulfur and hazardous air pollutant (HAP) precursor removal process has been developed; however, an effective sulfur and selected HAP removal process is needed to enhance the utilization of high-sulfur coals and to comply with increasingly stringent regulations. Subcritical water has been shown by the Energy & Environmental Research Center (EERC) researchers on this project to be an extremely effective fluid for the removal of organic sulfur from coals.

A multigram reactor designed and built at the EERC for supercritical water extraction was used to scale up from milligram-sized samples to 10-20 grams of coal charge. Work performed during this project year resulted in production of low-sulfur (as low as 0.5% S) extracted coal first at supercritical conditions, i.e., 450C and 400 atm (5880 psig), but then at conditions below the critical conditions, i.e., 420C and 156 atm (2300 psig). Still milder conditions of 400C and 156 atm (2300 psig) resulted in sulfur values similar to those of obtained under the supercritical conditions. IBC-102 extracted with supercritical water had a sulfur value of 0.7 wt%. Extraction of IBC-102 at subcritical conditions of 420C and 156 atm (2300 psi) resulted in a sulfur content of 0.49%. The tar obtained from the extracted coal had sulfur values ranging from 1.4 to 6.5 wt%, and when treated by catalytic desulfurization, tar was quantitatively recovered with a sulfur value of 0.6 wt%. Float-sink physical cleaning of IBC-102 with Certigrav 1.4 reduced the sulfur content of the coal to 1.5 wt% in a recovered float fraction of 83.3%. Approximately 300 lb of IBC-102 was obtained for use in preparing 100 lb of low-sulfur fuel. Float-sink cleaning on a sample of this "new" coal returned 87.1 wt% as float fraction, with 1.7 wt% sulfur. 158 lb of physically cleaned IBC-102 was used for the continuous process test on the pilot scale. An additional 150 lb of physically cleaned coal slurry was received from Dr. R. Honaker of Southern Illinois University (SIU) for testing in the continuous unit.

The hydrothermal process development unit (PDU) in the continuous mode was used to prepare 100 lb of low-sulfur fuel from physically cleaned IBC-102 at three temperatures (300, 325 and 340C) and a pressure of 2300 psig. Residence time was controlled by flow rate. Treating a swelling coal such as IBC-102 required a slurry with low solids loading to prevent plugging due to expansion during residence in the reactor. In this preliminary test with physically cleaned IBC-102, the sulfur content was decreased by as much as one-third, few volatiles were lost, and mercury and selenium levels were decreased by 87% and 46%, respectively.

The solution to a major problem still eludes the coal utilization industry: the lack of cost-effective methods for precombustion sulfur and selected hazardous air pollutant (HAP) removal from coal. With the diminishing supply of low-sulfur coal and with more restrictions imposed on the use of high-sulfur coal, sulfur removal from coal prior to combustion has become important.

Despite numerous attempts to remove sulfur from coal for low-sulfur, clean coal combustion applications, physical methods (e.g., selective agglomeration, flotation, or magnetic separation) have enjoyed some success for pyritic sulfur removal. Similar physical separation methods are ineffective for organic sulfur and organically bound HAP removal from coal. Cost-effective and technically sound methods for removing organic sulfur from coal are not available. While, in principle, efficient organic sulfur and HAP removal can be achieved through the use of expensive, high-concentration chemical reagents (e.g., molten caustic leaching method), little attention has been given to economic feasibility and potential environmental impact.

The main objective of this investigation at the Energy & Environmental Research Center (EERC) is to develop an economically feasible method for the removal of organic sulfur along with chlorine and selected air toxic trace metallic elements from Illinois Basin high-sulfur coals for clean coal and niche-market applications.

The so-called multigram unit, a 40-cubic-centimeter reactor, which was built as an extraction vessel and is typically charged with 10-20 grams of coal during experimentation, was used for scoping tests leading to determination of operating conditions for the larger pilot-scale process development unit (PDU). Effluent extraction fluid (water) containing coal tar was collected. This tar was subsequently removed from the water by liquid-liquid extraction for analysis. Additional tar was collected from an air-cooled condenser. This "dry" tar was analyzed for sulfur content. The extracted coal was dry when collected and was analyzed as recovered. The total effluent gas was metered and collected and was analyzed by gas chromatography.

Two major improvements in the batch hydrothermal desulfurization process were achieved as a result of the bench-scale testing. The first improvement was a reduction in effective desulfurization temperature and pressure. The sulfur content of IBC-102 bituminous coal was successfully decreased at temperatures as low as 400C and subcritical pressures of 156 atm (2300 psig). Sulfur levels in the desulfurized coal were consistently less than 0.80%, and values as low as 0.49% were achieved. The second major improvement was the addition of on-line catalytic tar desulfurization, which reduced the sulfur level in the tar from 1.4-3.6 wt% to <1.0 wt%. It was also demonstrated that the desulfurized coal could be added back to the solids, resulting in an increase of volatiles and improved mass balance.

Experiments designed to determine mass balance from multigram tests gave results for extractions of IBC-102 that were >90%, and the closures of the duplicates were 96 and 97 wt%, based on coal input and product output. Sulfur content of IBC-102 solids ranged from 0.77 to 0.46 wt%; tar was 1.4 wt% before catalytic desulfurization, and gas was ~2 wt%.

Scaleup from the multigram unit to the pilot scale was accomplished by reproducing the batch process of the multigram unit in the pilot unit, resulting in an approximately 1000-fold increase in coal capacity, complete with on-line catalytic tar desulfurization and a module for testing fuel reconstitution from desulfurized tar and solids. The EERC demonstrated the use of an on-line desulfurization system with the larger-scale system, reducing tar sulfur from 1.6% to 0.7%. The EERC then reconfigured the reaction system to carry out a continuous process. In the final test of the project year, 100 pounds of desulfurized fuel was produced in a continuous process from physically cleaned IBC-102 coal. The volatiles content of the product remained essentially unchanged, and the sulfur level was decreased by as much as one-third over a residence time approaching 20 minutes in a temperature range of 300-340C. In addition, the mercury content was 87% less than in the feed, and the selenium content was 46% less than in the feed.