Regenerative Activated Coke Technology (ReACT) is an advanced multi-pollutant technology that uses activated coke (AC) to reduce SO2 emissions, with a coincident reduction of NOx, mercury, and other pollutants, said expert testimony filed Sept. 17 at the Wisconsin Public Service Commission by Wisconsin Public Service Corp.
The filings are in support of the utility’s plan to install a ReACT system on the coal-fired Weston Unit 3. ReACT technology has been successfully implemented on large coal-fired boilers in Japan, was tested last decade on the coal-fired Valmy power plant in Nevada, and is commercially available worldwide, wrote H. James Peters, employed by Hamon Corp. as Executive Vice President of Strategic Planning and Business Development.
The utility, a subsidiary of Integrys Energy Group (NYSE: TEG), filed its initial application for approval of this project on May 7.
ReACT involves three process stages: adsorption, regeneration and by-product recovery.
- In the adsorption stage, flue gas from the unit passes through a slowly moving bed of activated coke pellets, which are made from coal, in the presence of ammonia. SO2, NOx, and mercury are simultaneously removed from the flue gas through a combination of adsorption and reduction mechanisms as the flue gas passes through the adsorber. From the adsorber, cleaned flue gas is directed to the stack and activated coke pellets with adsorbed pollutants are directed to the regeneration stage.
- In the regeneration stage, the accumulated pollutants are removed from the activated coke pellets by thermal desorption. The AC is then cooled and returned to the adsorption stage. Some new activated coke pellets are added to replace a small amount of AC that is lost in the regeneration process. Desorbed pollutant-rich gases from the regeneration stage are directed to the third stage, by-product recovery.
- In by-product recovery, the sulfur rich gases from desorption are used as a feedstock for the on-site production of industrial grade sulfuric acid.
“The ReACT system provides a simple but efficient contacting scheme between the flue gas and the AC in the adsorber, and provides ample margin for operating at full load,” Peters wrote. “Other expected operating scenarios, including low load operation, cycling operation, start-up and shutdown, and normal maintenance outages on upstream equipment are considered in the design and should pose no problems for the system.”
The main limitations for the ReACT process are related to its application for high-sulfur coals (which requires an increased adsorber and regenerator size) and its use with very high temperature flue gases (which limits sorption efficiency), neither of which are concerns at Weston Unit 3, Peters added.
ReACT can reduce emissions of sulfur oxides (SO2 and SO3) by more than 90%, NOx from 20% to 60% (depending on design conditions and the ammonia injection rate), and mercury emissions by more than 90%. The moving beds also provide for a net reduction of particulates due to interception and impaction mechanisms.
Process mostly used at this point in Japan
Peters noted that in the 1970s, Mitsui and Foster Wheeler further developed the existing ReACT process in conjunction with the Foster Wheeler Resox process at a demonstration plant at the coal-fired Scholz station in Florida, which ultimately did not become commercial. Mitsui continued to develop the process during the 1980s and together with Electric Power Development (now J-Power), Sumitomo and other Japanese industrial participants advanced the technology’s development in a project under a Japanese Ministry of Trade program. By the 1990s, Electric Power Development (currently known as J- Power) had fully commercialized the process as an advanced generation multi-pollutant control technology for coal-fired boilers and implemented the system on its own coal-fired fleet. Soon after, J-Power acquired the patents and intellectual property from Mitsui for the process, established a joint venture with Mitsui to develop activated coke, and formed a subsidiary called J-Power EnTech Inc. to market ReACT worldwide.
ReACT has been installed on at least fourteen commercial units outside the U.S., including three coal-fired utility units in the J-Power system and other industrial flue gas applications, most notably on large sinter plants in the steel industry. The majority of commercial installations are located in Japan.
J-Power Entech worked with the Electric Power Research Institute and a number of sponsoring U.S. utilities to evaluate ReACT. This program culminated in a demonstration test at the Valmy coal plant in northern Nevada in 2007. The project demonstrated the performance of the ReACT process and its components on typical U.S. low-sulfur western coals, specifically, sub-bituminous and western bituminous coals.
The Valmy project involved a series of tests to establish baseline conditions on sub-bituminous and western bituminous coals, as well as a series of parametric tests on the two types of coal to assess ReACT’s performance over a range of operating parameters and conditions. Demonstration testing revealed that the ReACT process was reliable over the five-month test period and achieved removal efficiencies comparable to those previously reported by J-Power Entech. The ReACT demonstration unit at Valmy reduced SO2 by more than 98%, NOx by up to 48%, mercury by up to 99.6% and had PM2.5 emissions in the range of 0.007 to 0.01 lb/MMBtu.
ReACT is a technology that offers some unique advantages for utilities that burn low-sulfur coals, and who have not yet installed SO2 control systems, Peters noted about its prospects for deployment in the U.S. “The potential advantages derive from the capability of the process as a flexible multi-pollutant control technology that can satisfy increasingly stringent emissions regulations, its near zero water use or impact, its ability to be retrofit into existing coal-fired boilers, and its minimization of residual wastes,” he added.
For Weston Unit 3, ReACT has a number of advantages that are either site specific and/or derive from the technology itself. As a retrofit installation to the existing power plant, the relatively small footprint for the adsorbers and regenerators fits just behind the existing fabric filters. Also, the operation of the existing upstream equipment is minimally effected, as the ReACT system will be located downstream of the existing fabric filters, Peters noted.
Peters described ReACT’s advantages over more traditional wet and dry flue gas desulfurization (FGD) systems. For Weston Unit 3, one serious drawback associated with using a wet FGD is that it creates a mercury-laden wastewater discharge stream. The plant abuts the Wisconsin River, and it is his understanding that the Wisconsin Department of Natural Resources has listed the river as being impaired for mercury, Peters wrote. So any effluent from a wet FGD would likely have to be treated before being discharged into the river, which would require the extra cost of a wastewater treatment facility. Also, there are other operational considerations that weigh against a wet FGD system at this unit, such as:
- the higher amount of power and water required;
- the resultant heavy steam plume that would be visible at the stack exit during all ambient conditions;
- the need for a new chimney to be constructed to prevent corrosion of the chimney liner and to meet stack exit velocity requirements; and
- the operating and maintenance requirements that are more involved.
Among the other factors that weigh against other systems are that both WFGD and DFGD processes consume significant quantities of water in the process, and WFGD processes have large liquid bleed streams that must be further processed, Peters said. ReACT minimizes wastes, creates a saleable by-product and has only minor water usage. ReACT also can achieve NOx reductions, which are not available from WFGD or DFGD systems, and provides high level co-benefit mercury reduction, where DFGD and WFGD systems may have to incorporate activated carbon injection systems or may have issues with mercury re-release.
There are four coal fired units and two gas fired combustion turbines at Weston. The coal units are Weston Unit 1 (57.2 MW), Unit 2 (79.8 MW), Unit 3 (323.9 MW) and Unit 4 (551 MW). The combustion turbines are Weston Unit 31 (19.5 MW) and Unit 32 (48.1 MW).
Factor driving this project is imminent air settlement with EPA
Paul Spicer, employed by WPS as Vice President – Energy Supply, said in companion Sept. 17 testimony that he primary driver behind the ReACT project is the Notice of Violation (NOV) issued to WPS by the U.S. Environmental Protection Agency alleging that modifications were made to Weston 3 and other generating facilities without first obtaining certain air permits. WPS expects to resolve this NOV through a settlement with EPA that will be formalized in a consent decree filed in federal court. WPS expects that the resolution will include reductions in SO2 and NOx from Weston Unit 3 in the near term. Those reductions can be achieved by emission controls or shutting the unit down. Apart from the NOV resolution, WPS expects environmental regulations applicable to coal-fired facilities to become increasingly stringent in the future.
WPS evaluated the technical feasibility of a number of emission control technologies to achieve significant emission reductions of SO2 and NOx from Weston Unit 3. Through this evaluation, WPS selected two emission control approaches for economic analysis: Selective Catalytic Reduction/Flue Gas Desulfurization (SCR/FGD) and ReACT. The other available method of meeting emission reduction requirements is of course to retire the unit.
In general, in all three of the “futures” analyzed – a base or most likely case, a moderate “coal unfriendly” case and a “very coal unfriendly” case – ReACT would be a significantly less costly multi-pollutant control technology than SCR/FGD. This is primarily due to the higher capital cost associated with SCR/FGD. Also, in two of the three futures, ReACT is significantly less costly to WPS’s customers than retiring Weston 3. In the very coal unfriendly case, ReACT and retirement are about equal in cost.
The economic analysis demonstrates that, at worst, ReACT would be no more costly to customers than unit retirement. But, if the unit were retired, customers would be exposed to significantly higher costs if the future is less unfriendly to coal than what was assumed in the very coal unfriendly case, Spicer noted.
Utility looks at impact of court throwing out CSAPR
Leonard Rentmeester, employed by WPS as the General Manager-Pulliam in the Energy Supply Operations Department, supplied heavily-redacted testimony about what the possible settlement with EPA would involve. Not redacted was a section about the impacts of an Aug. 21 federal appeals court decision to send the Cross-State Air Pollution Rule (CSAPR) back to EPA for further work, leaving the Clean Air Interstate Rule (CAIR) in force in the meantime.
CSAPR created a regional allowance trading market, which was aimed at significantly reducing power plant emissions of NOx and SO2. Under CSAPR, each affected state (including Wisconsin) had a state allowance budget, and allowances were then divvied out to utilities in the state based on a formula, Rentmeester said. CSAPR was supposed to begin in 2012 – but the D.C. Circuit Court stayed implementation of the rule pending resolution of the various challenges to the rule (including challenges from WPS). Based on historic levels of operation, WPS expected to have a significant shortage of SO2 allowances under CSAPR, particularly after 2014. WPS has a significant existing bank of CAIR allowances, so it should not need additional emission reductions at Weston Unit 3 in order to meet CAIR. However, the project would help WPS comply with the replacement rule, or CSAPR itself if EPA challenges the D.C. Circuit’s decision and gets it overturned.
“While it is hard to speculate about what a future transport rule might look like, WPS believes that the Court’s decision, if implemented by EPA, will likely lead to less stringent SO2 and NOx reduction requirements for Wisconsin sources than were contained in CSAPR,” Rentmeester added. “This is because the Court’s decision requires EPA to consider the relative upwind contributions to air quality problems in the downwind states, and Wisconsin has fairly low relative upwind contributions. Of course, EPA’s future transport rule is unknown at this point and EPA could use other regulatory levers to try to make the next transport rule more onerous. In any event, CSAPR is just one of the reasons that WPS believes the project is needed now.”
The ReACT project will also help WPS comply with the impending state and federal mercury rules. Those rules generally require a 90% reduction in mercury emissions by 2015.
WPS’s investigation of the ReACT technology dates back to 2007 with WPS participation in the EPRI testing at Valmy, Rentmeester noted. Following the results of the Valmy testing, WPS initiated a formal evaluation of the ReACT processes and technology beginning in early 2010. Based on the results of the studies, formal evaluations of the ReACT technology, activated coke sources, and sulfuric acid production and marketing were conducted by teams from WPS. These evaluation included trips to both Japan and China.
Initially the activated coke (AC) for the project will be supplied by Hamon Research Cottrell, the U.S. ReACT licensee. “It is our understanding that Hamon will obtain the initial AC from one or more or sources in China,” Rentmeester said. “WPS has also identified a number of potential international suppliers outside of China. In addition, WPS is currently evaluating interest in the domestic manufacturing of activated coke with several potential suppliers.”
The ReACT project has an estimated capital cost of $250m with an estimated $38m in Allowance for Funds Used During Construction (AFUDC). The project estimate represents the total installed cost inclusive of the ReACT system, sulfuric acid plant, spare equipment, contingency, and owner’s costs. The engineering, procurement, construction, and commissioning of the ReACT system is expected to take approximately 42 months from regulatory approval to in-service.