Miniature scrubber can predict both particle removal
and absorption EFFICIENCY
With new ambient fine particulate rules, plants will be forced to make reductions. In many cases, there are other regulations which will also require reduction of mercury, HCl, and SO2. It is, therefore, critical to determine the multi-pollutant capabilities of proposed scrubbers. Due to the concentration of acid mist deposits due to scrubbing, it is essential to address the condensibles problem prior to scrubber installation.
The science of mercury speciation and prediction of mercury capture in SO2 scrubbers is very poor. Yet it is critical to determine the mercury capture which will be achieved with installation of scrubbers. SO2 scrubbers are already designated to play the most important role in mercury reduction.
The typical response from system suppliers when asked about particulate and mercury guarantees is the following:
(1) We will provide a particulate guarantee based on a particle size distribution supplied to us. But, if the emission target is not met, and we do our own particle size analysis later and find that the distribution has changed, then our guarantee is void. The guarantee only applies to discrete particulate and not condensibles. We may be able to give some weak guarantee on SO3 removal.
(2) We will provide a qualified guarantee on mercury removal based only on the oxidized mercury and not elemental mercury.
The reality is that the purchaser who proceeds based on the above type guarantees is no better off than without any guarantee. It is disappointing that 30 years after McIlvaine alerted the industry to the risks, it is still relying on particle size distributions.
Particle size distributions are highly inaccurate (not reproducible), and even if they were accurate, they do not directly indicate the particulate removal ability of a scrubber. All they indicate is the ability of impingement plates to capture particles at varying velocities. The density, particle shape, and electrical characteristics are as important as the mythical particle diameter of a non-spherical particle.
The only sure way to determine the weight reduction of particulate by the scrubber is to install it and measure the results. But, if you need the information now, then the next best way is to install a miniature scrubber which approximates the performance of a full scale scrubber.
The mercury removal ability of the scrubber depends on a number of variables. This art (not science) is in its infancy. Again, the only way to determine how much mercury will be removed by an SO2 scrubber is to install one. So, the best way to determine performance in advance is to install a miniature scrubber and test its performance.
SO3 and other condensibles captured in the scrubber are very complex. The operation of a miniature scrubber and measurement of upstream vs. downstream condensibles would add greatly to the decision-making process.
Is a 1 cfm scrubber nearly as good as a 100 or 1000 cfm?
One approach is for vendors to operate 1000 cfm pilot unit at a specific site prior to bidding. Guarantees can be made based on the performance of the pilot unit. After installation of the full scale unit, the 1000 cfm unit can be run in parallel to the full-scale unit and determination made as to whether conditions changed between the bid date and the date of operation. This approach will work for discrete particulate, condensibles (including SO3) and mercury.
In the case of mercury, the question of meeting guarantees is simplified. If the emissions of mercury from the 1000 cfm pilot are greater after the full scale installation than at the time of the bid, then the supplier should only have to duplicate the results of the pilot unit with the full scale unit.
There are disadvantages of running a 1000 cfm pilot plant. Cost and time are two obvious drawbacks. But there are others. The results are only meaningful for one scrubber design on one plant. It is difficult to extrapolate this data to other plants. More importantly, it is difficult to extrapolate the data to determine the impact of fuel changes or other significant operation changes on scrubber performance.
It is, therefore, desirable to develop a standard scrubber which would function as does the cascade impactor but to directly determine the dust and/or mercury rather than some indirect parameter. Fortunately, there is considerable experience both with 100 cfm and even 1 cfm units. This experience can be used to develop industry standards.
Experience with Miniature Scrubbers
In the late 1960s and early 1970s, Combustion Engineering in partnership with Environeering was offering SO2 removal systems which replaced existing precipitators and provided both particulate and SO2 removal. In order to make accurate guarantees, Environeering developed the 100 cfm Dust Difficulty Determinator. This is just a simple orifice scrubber. Pressure drop can be varied to change particulate and SO2 removal efficiency. This device was utilized to test many boilers. It proved to be as accurate as a 1000 cfm scrubber. Environeering was later acquired by what is now Babcock Power. So, assumably, details of this device are still available somewhere.
McIlvaine took the concept one step further and reduced the orifice scrubber to just 1 cfm so it could be used in standard sampling trains. Dupont, AERE Harwell, Air Pol, Nalco, Martin Marrietta, and other organizations have successfully used this device to predict scrubber performance. It was given the commercial name of the McIlvaine Mini Scrubber.
McIlvaine offered the drawings and concept free-of-charge to the Institute of Clean Air Companies if the organization would work toward adopting some standards. Several members began using the device, but the organization never banded together to make it a standard bid tool.
Now is the time to develop 1 cfm orifice scrubber to predict performance on discrete particles, condensibles, and mercury.
It is unacceptable that expensive scrubber projects should be undertaken with reliance on particle size distributions and other tools which are highly unreliable. The interest in not just discrete particle removal but also condensibles and mercury means that the one device now has multiple uses.
Not only should this device be used by individual vendors to make guarantees, but it should be used by EPA and their contractors for research. It will be much less costly to determine total mercury removal based on certain combustion conditions than it is to try to analyze all the mercury species and make predictions on removal.
The concept should be expanded to include adsorbers, fabric filters, and other APC devices.
The general principle of building a miniature production device and measuring the results rather than measuring parameters and predicting results should apply to other air pollution technologies. For example, a 1 cfm activated carbon impregnated filter disc could be used in a Method 5 train to predict mercury removal for systems that inject activated carbon ahead of baghouses.
Think how valuable such a device would be on a retrofit mercury reduction project. Without this device the purchaser has no way to prove that conditions are equally favorable for activated carbon mercury capture after installation as at the time of the bid. The vendor has no way to prove that it was a change in conditions which caused the full- scale installation to fail the guarantee test.
Ideally this whole concept of basing guarantees on the performance of miniature devices at bid time compared to the time of full-scale guarantee validation makes a great deal of sense and should support a whole new industry segment.
Who should carry this effort forward?
First of all McIlvaine is offering design drawings of the McIlvaine Mini scrubber free-of-charge to anyone who wants to pursue this subject. Beyond this there are a number of organizations which could move this concept along. Thermo Electron is in the business of supplying Method 5 equipment and is a logical entity to offer the components. Clean Air Engineering not only offers Method 5 trains but was the early subcontractor to manufacture the McIlvaine Mini Scrubber.
EPA ETV is a logical entity to use the concept to test the performance on commercial devices. Investors have lost huge amounts of money on scrubber designs which were claimed to be efficient based on particle size analyses. Comparing performance of novel devices to that of a standard miniature scrubber would avoid this problem.
EPA BACT and MACT formulators can use this concept to write much better rules. BACT or MACT can be defined in terms of the performance of the miniature device. Since with the mini scrubber the performance increases with pressure drop, BACT or MACT can be defined in terms of a selected pressure drop.
More information on this subject and on all subjects affecting scrubber/adsorber decisions are contained in the Scrubber/Adsorber Knowledge Network published by the McIlvaine Company. For more information on this service, click on: http://www.mcilvainecompany.com/air.html#online .
Bob McIlvaine
847-784-0012