Determine the Efficiency of Air-Pollution Control Devices

Chemical Engineering Progress, May 2007 by Buchanan, Lewis S

For batch processes, use a surrogate feed stream to simulate the inlet conditions necessary for stack testing to demonstrate compliance.

The Miscellaneous Organic NESHAP (National Emission Standards for Hazardous Air Pollutants), or MON rule (1), requires companies to demonstrate the control efficiency of air-pollution control devices used to comply with the regulation (40 CFR §63.2450(g)). The demonstration must be made at the miscellaneous organic chemical process unit's (MCPU) maximum representative operating conditions (40 CFR §63.997(b)).

This evaluation is straightforward for units that operate in a continuous mode. But for batch or cyclical units, the demonstration is much more difficult. Because of the variability in the process venting rates during the batch cycles and because multiple MCPUs can be tied into a common control device, it can be nearly impossible to maintain the necessary operating conditions in all of the units.

This article presents an alternative approach for demonstrating compliance with air-pollution control-device efficiency requirements for batch and cyclical operations. This method, known as "spiking," uses a synthetic vapor stream to simulate the necessary control-device inlet conditions for the period of time needed to conduct the stack testing for the compliance demonstration. It involves: (1) estimating the peak inlet conditions to the control device from multiple vents and multiple MCPUs; (2) consolidating the numerous vented air pollutants down to two or three surrogate compounds; and (3) delivering a synthetic feed of the surrogate compounds to the control device to simulate the maximum representative operating conditions.

When using this approach for MON compliance demonstration, it should be submitted for approval with the pre-compliance notification required tor the MPCU. It should also be included with the subsequent submission of a site-specific test plan (1).

Why use a synthetic feed instead of actual operation?

For air-pollution control devices receiving hazardous air pollutant (HAP) emissions from continuous and single-hatch processes, there may be little reason to use the synthetic-feed approach. But for devices tied to multiple-batch and/or continuous processes, the advantages tire more obvious. The synthetic-feed approach avoids several process challenges:

* timing of worst-case emission events from multiple processes: With varying cycle times and venting patterns, timing multiple batches to get just one valid l-h to K-h test run can be difficult enough. But putting together three valid test runs could take days, even weeks. Eliminating the extra mobilisation or standby time required of the stack testing contractor may more than offset the cost of the synthetic-feed approach.

* difficulty in scheduling the product mix for the worstcase scenario: With multiple units, the worst-case product would need to be running in each unit at the same time to complete the testing. It may not make business sense to schedule special campaigns only for the testing. It may he easier to schedule a single downtime period for all of the affected process units and use the synthetic-feed approach,

* planned future introduction of new products and processes: The initial compliance demonstration must be redone within 180 days of any change in the worst-case conditions. If product or process changes are routine for an MCPU, then a more-robust initial compliance demonstration may be appropriate. Such a demonstration would use a synthetic feed to simulate a broader worst-case than is required for the existing MCPU, thereby limiting or avoiding the need for future retesting. (This approach is only as good as your crystal ball.)

Despite these advantages, the synthetic-feed approach does have certain disadvantages:

* production unil downtime; Typically one or more of the affected process units will be shut down during the spiking event. There may also be some downtime associated with installing the feed ports and preparing the site.

* chemical consumption: Virgin, pure or high-quality chemicals are required for the synthetic feed. Since waste chemicals may not meet the test requirements, the cost of the chemicals consumed to make the synthetic feed should be factored into the decision-making process.

* an extra contractor: A separate "spiking" contractor will be needed to supply the synthetic feed. Stack test firms usually subcontract this work.

* mechanical and electrical modifications: Additional site preparation is likely to be needed to supply the inlet feed port(s), steam, and electrical power required for the synthetic-feed system.

The remainder of this article details the steps necessary to prepare for and complete a compliance demonstration using a synthetic-feed approach. Steps 1 through 5 must be completed prior to the pre-compliance notification deadline (Nov. 10, 2007). because the compliance demonstration approach must be included with the emissions profile in this notice. It could take months to complete these steps, so start early.