As stated in my recent CSAPR post, the New Year has been a tumultuous period for power generators, with respect to both emissions regulations and other industry developments. In this post, I want to focus on the recently finalized US EPA Utility Maximum Achievable Control Technology (MACT) rule, as well as a few additional developments that have the effect of upping the ante for optimized boiler operations.
The Utility MACT was issued December 21, 2011. It was required by a court-ordered Consent Decree which arose from the litigation resulting from its predecessor, the Clean Air Mercury Rule (CAMR), which was vacated in 2008. The Utility MACT rule sets maximum achievable control technology (MACT) standards for coal and oil generating stations for removing mercury, acid gasses, and a variety of toxic metals from the flue gas for coal and oil-fired utility boilers. EPA projects that the Rule’s annual incremental compliances cost will be $9.6 billion.
More than two decades in the making, the Utility MACT rule has been a high-profile regulatory development, accompanied by the typical controversies over the costs, benefits and reliability impacts. In announcing the rule, EPA argued that it made a number of changes to provide more flexibility to utilities. As an indication of the interests at stake, the rule from the time it was first proposed to its final issuance drew more than 900,000 comments.
For coal-fired units, the Utility MACT limits emissions for hydrogen chloride (as a surrogate for acid gas hazardous air pollutants (“HAP”), filterable particulate matter (“PM”) (as a surrogate for non-mercury HAP metals), mercury, and organic HAP. The limits vary based on type of coal burned and whether the units are new or were already in operation when the final rule was published. Total non-mercury HAP metals and individual non-mercury HAP metals limits can be used as an alternate to the filterable PM limits. Coal-fired EGUs with flue gas desulfurization (FGD) units can use SO2 limits as an alternative to the hydrogen chloride limits.
The new rule creates a variety of new work practices standards. For example, units must burn natural gas or distillate oil during periods of start-up. Relevant emissions control systems must be operated during periods of start-up and shutdown. And, as discussed below, optimization of NOx and CO, combined with "best practices" efforts to maximize fuel efficiency must be demonstrated and documented for the EPA.
All sources are given three years to comply with the Utility MACT. Sources can also apply to the EPA or appropriate state authority for a one-year compliance extension if the extension is needed to install controls—bringing the total compliance time to four years. The one-year extension is granted on a case-by-case basis.
To address concerns regarding reliability, the EPA has stated -- via administrative orders -- that it may grant some sources an additional year to comply. EPA will only grant these administrative orders to sources that are critical to electric reliability and, due to forces beyond their control, be forced to deactivate temporarily to comply with the Rule or are unable to install controls in time.
Final Rule Modifications
There were several changes made during the public review process embodied in the final rule, particularly allowing hydrogen chloride as a surrogate for acid HAPs and filterable particulate matter as a surrogate for non-mercury HAP metals. The net effect is that substantially fewer coal-fired units will need to install costly scrubbers.
There was also language added to the final rule explicitly referencing "neural network optimization” systems as qualifying for "optimizing NOx and CO." Similarly, the final bill has language rewarding boilers benefitting from "neural network optimization" by delaying the initial EPA "best practices” evaluation by a year and reducing the required frequency for subsequent evaluations from every 36 months to every 48 months. The motivation for mandating these best practice operating protocols is to ensure that HAPs included in the rule are not measurable through current CEM technology (such as dioxin and fiurin) and are being minimized through the best achievable operating efficiencies. More on this below.
Boiler audits to ensure "best practices" efficiency
All publicly released drafts of the rule state that the non-measurable HAPs, such as those described above, would be regulated through requirements in which owners and operators of affected boilers demonstrate they are employing all available efforts to minimize these pollutants through maximizing the efficiency with which the fuel containing them is converted into electricity. As a testament to how much the industry objected this provision, the required frequency has gone from 12 to 18 to 32 months and now 48 months with an optimizer in three successive versions based on the public comment process.
This explicit recognition of the MACT compliance value of optimization not only reduces a painful regulatory burden by 25%; but it also signals that the EPA views closed-loop optimization as part and parcel of the best practices boiler operations as they pertain to minimizing non-measurable HAPS (i.e. fiurin and dioxin). Here are some excerpts from the final rule that were not included in earlier versions:
The initial compliance demonstration for the work practice standard of conducting a tune-up may occur prior to the compliance date of the rule, but must occur no later than 42 months (36 months plus 180 days) from the compliance date of the rule or, in the case of units employing neural network combustion controls, 54 months (48 months plus 180 days).
Subsequent demonstrations of these standards must be conducted at each planned major outage and in no event less frequently than every 36 calendar months, with an exception that if the unit employs a neural-network system for combustion optimization during hours of normal unit operation, the required frequency is a minimum of once every 4 years (48 calendar months).
Each work practice shall also entail optimizing combustion to minimize generation of CO and NOx. NOx optimization includes burners, over-fire air controls, concentric firing system improvements, neural network or combustion efficiency software, control systems calibrations, adjusting combustion zone temperature profiles, and add-on controls such as SCR and SNCR; CO optimization includes burners, over-fire air controls, concentric firing system improvements, neural network or combustion efficiency software, control systems calibrations, and adjusting combustion zone temperature profiles.
One of Several New Efficiency Drivers
The Utility MACT rule is just one of the many developments making fuel efficiency (i.e. heat rate) a bigger concern which has historically been the case. In addition to the above-described need to demonstrate efficient boiler operations out of concern for non-measurable HAPS; the formulae for HG and other measurable HAPS are focused on a year-by-year variability. Since the new MACT limits are not substantially lower than current emissions rates at many units, anything that increases fuel efficiency, reduces year-to-year variability, allows better planning for physical sorbent injection modifications and reduces the sorbent needed to make the various limits will be economically beneficial for MACT compliance.
Another part of the incentive created to maximizing fuel efficiency stems from the "hair trigger" phenomenon for activated carbon, where meeting HG limits with even a few percentage points lower sorbent injection rates can mean the difference between being able to sell fly ash or having to truck it away to a landfill. For many units, the ability to implement sorbent injection while still remaining under the maximum carbon content threshold required to continue selling fly ash will mean millions of dollars of otherwise lost revenue and increased disposal costs.
Another factor stems from recent unprecedentedly low natural gas prices. As I’ve previously blogged about at some length, my view is that these recent prices have resulted from the combination of things. First, the lingering effects of the "Great Recession." The ability to store natural gas in large quantities underground makes for something on the order of a year-long delay in the effects of fundamental supply and demand on price behavior. Second, I believe that natural gas markets have not yet internalized the anticipated double-whammy effects of environmental concerns (groundwater contamination, methane emissions, etc.) and limitations in the availability of the large amounts of water required by this process in the first place on the extent to which "fracking" can make sustained long-term contributions to increasing domestic natural gas supply.
The biggest effect of these low natural gas prices on coal-fired generating units is that the latter are now competing with gas-fired combined cycle plants (CCCTs) not only in organized locational price markets, but also in traditionally regulated markets where there is a centralized dispatch using "merit-order" algorithms to determine which units are run to meet demand at any particular time of the day, season or week. Back when natural gas prices on heat content (i.e. $/mmBtu) were a multiple of coal prices, even the most inefficient smaller coal units would be "in the money."
Currently however, even large, relatively new and efficient coal units are finding themselves shut down or at minimum load being displaced by combined cycles with slightly lower operating costs at today's low gas prices. The difference of even 50-100 BTU/KWh can now make the difference between generating profitable revenue or losing money sitting idle.
Tying these somewhat disparate threads together and attempting to sum things up, a confluence of regulatory developments and market conditions have combined to make squeezing every last BTU from your fuel more important than ever. And with the provisions for optimization written directly into the final Utility MACT rule, the 1200 coal-fired units affected by the rule, generators can simultaneously increase efficiency , be in a better positions to comply with the Clean Air Act and its Amendments, and minimize the frequency with which the EPA is inspecting your operational practices. Makes sense to me. Does it to you?