Portland Cement MACT May Breathe New Life into Mercury Monitoring

A recently proposed EPA rule may rekindle the mercury CEMS and sorbent trap market...

On April 21, 2009, the USEPA proposed to reduce mercury emissions from Portland cement kilns. There are currently 93 portland cement plants operating 163 kilns and associated clinker coolers in the U.S. that are expected to be affected. These plants represent the fourth-largest source of mercury air emissions in the U.S.

The proposal, known as a Maximum Available Control Technology standard, or MACT standard, would set the nation’s first limits on mercury emissions from existing Portland cement kilns and would strengthen the limits for new kilns. The EPA believes that the proposed new source MACT "floor" is approximately 30 percent lower than the not-to-exceed standard EPA promulgated in December 2006, which was 41 µg/dscm.

The proposed MACT standards also would set limits for emissions of total hydrocarbons (THC), particulate matter (PM), and sulfur dioxide (SO2) from cement kilns of all sizes, and would reduce hydrochloric acid (HCl) emissions from large kilns. Under the proposed amendments, existing kilns would be expected to add one or more control devices to meet MACT, most likely a wet scrubber and activated carbon injection (ACI).

Each kiln would also be required to install continuous emissions monitoring systems (CEMS) to monitor mercury, THC and HCl. Direct PM monitoring would not be required; instead, bag leak detectors (BLDs) would be required to monitor performance of all baghouses.

Existing facilities will have three years after the promulgation to comply with the new emission limits for mercury, THC, PM, and HCl. This is the maximum period allowed by law. Most facilities will need this time, since they will likely have to install new emissions control devices to comply with the emissions limits. For new sources (i.e., those that started construction after December 5, 2005), the compliance date will be the date of publication of the final rule or startup, whichever is later.

Generally, mercury emissions from cement kilns reflect the amounts of mercury in the feedstock and fuel inputs of the kilns. The EPA assumes that nearly all the mercury entering a kiln eventually leaves as a stack emission. Usually, up to 80% of the mercury emissions result from trace contaminants of mercury present in the limestone, which typically comes from a quarry located adjacent to the plant. Mercury is also found as a trace contaminant in the other inputs to the kiln such as silica, alumina, and iron, as well as the coal and petroleum coke used to fuel the kilns.

Since the mercury content of the feed and fuels is believed to vary significantly from day-to-day, cement kilns do not emit mercury uniformly. This variation is compounded in modern preheater and preheater/precalciner kilns that have in-line raw mills. Consequently, short term emission tests would not be expected to provide a good indication of long term mercury emissions from cement kilns.

Therefore, the proposed rule will require plants to install either instrument based or sorbent trap based mercury monitoring systems that meet either the requirements of PS-12A or PS-12B. These performance specifications are being added into 40 CFR 60, Appendix B as part of the proposed rule changes (they had been removed after CAMR was vacated

The EPA states that they “see no technical basis to say that [mercury CEMS or sorbent trap monitoring systems] will not work as well on a cement kiln as they do on a utility boiler. In addition, we are aware that there are 34 cement kilns that have operating continuous mercury monitors in Germany.”

We'll see if the rest of the industry agrees. The proposed rule is currently in the midst of a 60-day comment period that expires on or about June 20th. After that, we’ll have to see what, if any changes the EPA makes before publishing the final rule. Stay tuned…

More information on EPA’s Official Site.

Sorbent Trap Mercury Monitoring…What does the Future Bring?

On May 20th, the U.S. Court of Appeals for the District of Columbia Circuit rejected EPA and utility industry requests to rehear and overturn the ruling that vacated the agency’s Clean Air Mercury Rule (CAMR). Barring an unlikely appeal of this case to the U.S. Supreme Court, it appears certain that the federal CAMR and its associated mercury monitoring requirements are, for the moment anyway, dead before arrival. (read more about this ruling)

So, what does this mean for the future of sorbent trap monitoring systems like the MET-80? It’s probably too early to tell what the long-term prognosis is, but some of the early indications thus far may indicate that sorbent trap sampling is more alive and well than ever. Thanks to the inherent functionality, adaptability and robustness of the sorbent trap approach, it is now being seen by some pragmatists as the most logical and effective means of gathering mercury emissions data. What’s more, sorbent trap systems like the MET-80 are out there collecting good and reliable data NOW, something very few (if any, really) Hg CEMS owners can claim.

The fact is, mercury measurements are perhaps needed now more than ever. With the vacatur of CAMR, the winds are shifting towards the EPA developing a MACT standard for EGUs based on Section 112 of the CAA. It is highly likely that any mercury MACT standard will include continuous monitoring requirements for mercury (and perhaps any one of the other HAPS). The importance of this data with respect to the total potential cost to the utility would only increase. So, although the deadline for monitoring may get pushed back a little, the need still looms.

Then there are the states that developed their own mercury rules. Massachusetts has required monitoring since the beginning of this year, and Dominion’s three MET-80s at Salem Harbor are the first sorbent trap systems in the country that are pumping out certified data for a utility. Michigan and Wisconsin are two more states that have their own mercury monitoring requirements in the pipeline. Although the dates for these are a little further out there, two of our MET-80 clients, Detroit Edison and Wisconsin Public Service Corporation, are pushing forward now with the installation of their systems.

Today, there are MET-80 systems collecting data in five states – Massachusetts, Michigan, Wisconsin, Indiana and Ohio. Although the end use of some of this data may now change with the departure of CAMR, the interest in the data has intensified. This has been even further magnified by the fact that the mercury CEMS are still being plagued by operational problems, and the ones to which our MET-80 systems were planned as back-ups have yet to collect any useful data. As they say, timing is everything.

Salem Harbor Plant Back On-Line

Dominion's Salem Harbor Station in Salem, MA brought its Unit 2 boiler back on-line late last Friday night. The plant had been shut down since November 6, 2007, when a tragic steam explosion killed three plant workers.

The Salem Harbor Station uses CleanAir's MET-80 sorbent trap systems to monitor mercury emissions from three of its four boilers. Under Massachusetts regulations, coal-fired utilities in the state have been required to continuously monitor mercury emissions since the beginning of 2008.

Massachusetts is one of the first states to implement requirements for mercury monitoring at utilities. As such, Salem Harbor's three MET-80 systems are some of the first mercury monitoring systems of any type in the nation to be officially certified for regulatory reporting purposes. All three systems passed their Relative Accuracy Test Audit (RATA) certification tests last October.

MET-80 Hangs Tough Amidst the Winds of Regulatory Change

The U.S. EPA regulatory requirements for mercury (Hg) monitoring effectively went into limbo after the Washington D. C. District Court of Appeals vacated the Clean Air Mercury Rule on February 8, 2008. This ruling was mandated on March 14, 2008, but the mandate was subsequently appealed on March 24, 2008. The outcome of this appeal is still pending.

If the ruling to vacate the CAMR stands, the federal requirements for continuous mercury monitoring under 40 CFR Part 75 may disappear. However, most experts agree that control of mercury emissions from coal-fired power plants will remain a high priority both nationally and world-wide. Any plan to curb power plant Hg emissions will require the accurate monitoring of the Hg content in combustion flue gas streams.

Although future monitoring requirements may change slightly from those currently specified, the overall technological requirements and challenges of mercury monitoring remain essentially the same. Appendix K sorbent trap monitoring remains a viable option for facilities that are required to monitor mercury emissions, regardless of the underlying regulatory driver. CleanAir's MET-80 system, with its design based on maximum versatility over a wide range of installation requirements, will continue to lead this charge.

A Brief Introduction to the MET-80 Sorbent Trap Monitoring Systems

Hg emissions can be monitored via continuous emissions monitoring systems (Hg CEMS) or a less complex sorbent-based monitoring system (STMS) approach. There are several sorbent-based monitoring systems available that are based on similar operating principles. Clean Air Engineering’s monitoring system, the MET-80, consists of the following main components:

• Sampling probe housing the actual sampling media (sorbent traps) in a temperature controlled compartment
• Multiple-zone heated sampling line that is integrated with the probe to reduce the number of connections
• Gas conditioning module with acid mist filters removing flue gas moisture and preventing acid mist breakthrough
• Condensate retrieval module facilitating a direct measurement approach for the sample gas moisture content
• Gas sampling and the control module governing the instrument operation, data handling and network connectivity

The picture above shows two typical layouts of sorbent trap based monitoring equipment installations. Sorbent trap monitoring systems were traditionally installed at stack platforms in close vicinity to the sampling port and housed in a rugged and environmentally controlled cabinet (figure on right). As an alternative, these systems can also be installed at the bottom of the stack with an extended heated sample line connecting the sampling system to the sampling probe at the stack platform (figure on left). Sampling line lengths at these installations can typically extend up to several hundred feet. In case of a remote installation, the system usually includes distributed control equipment at the top of the stack connected to the main control system via fiber-optic cabling. Network connectivity to the plant, in either configuration, is achieved via Ethernet-based Modbus TCP.