Don’;t choke when estimating plant emissions

Know your model and follow these guidelines to save money by getting it right the first time.

By Don Dickerson, Lockwood Greene

Share Print Related RSS

They can be opaque, translucent or invisible. They parade through the sky in Technicolor green, yellow, orange, brown and red, or basic shades of black and white. Varying in fragrance from odorless to offensive, they hug the ground, disperse rapidly into the atmosphere, or just smother the sky as an oppressive blanket of smog.

What are they? They are plant emissions. Like a many-headed dragon, they come back to life with each plant expansion. Your job is to perform a desktop emissions estimate and to do it well enough so you don’;t have to independently confirm its accuracy with an even more expensive stack test. Intended as a tool to help you estimate plant emissions, this article opens with a set of holistic principles and rounds out with an explanation of the fundamental process activities used to model emissions.

Emissions are best estimated by following these five holistic principles:
1. The process engineer knows his process; the modeler knows his model. Yet the two must collaborate in an open relationship. There are no stupid questions.
2. The goal is to estimate emissions, not assess blame.
3. Iterations to improve the accuracy of the model are unavoidable.
4. Emission modeling and permitting should be assigned to different people.
5. Ensure paper and electronic trails exist so the estimate can be reproduced.

Collaboration between process and modeling. Perhaps it began in the Garden of Eden, but the fact is, we humans can be fearful of peer opinion. This seems especially true for scientific types who tend to seclude themselves with models, textbooks and equations. We can pretend to know something and get along famously for awhile -- until we are called upon to demonstrate our supposed knowledge. At that point, our pretension breaks down and we must confess our ignorance. But if we ask the tough questions in the first place, we can avoid constructing a house of cards that will likely collapse around us.

The modeler needs to establish at the outset a relationship with key personnel who understand the process, can communicate its nuances and are willing to ask the modeler any question they like to make sure the estimate is realistic and representative. Absent this relationship, pretension builds up, questions are avoided and simulation outputs will be contrary to what all parties desire.

Estimate emissions; don’;t assess blame.Early in my career, I was involved in the development of a complex, large-scale emission inventory with a tight deadline and budget in an environment of know-it-alls. I soon found myself too fearful of peer criticism to admit that my estimating procedure was wrong. Staffing for the project was low and problems snowballed, unobserved by others for weeks, until the avalanche was obvious to everyone. Because I had failed to communicate my need for help, the plant was unwilling to purchase several weeks of unfruitful service, resulting in a massive rework at no charge.

Had I been working in an environment that was receptive to cries for help, I would have spoken up before the circumstances spoke for me. The lesson of this anecdote is that you must quickly establish an environment of trust on any emissions inventory team. The goal is to estimate emissions using practical, reproducible procedures, not to chase down errors so those responsible can be ostracized and embarrassed.

Iterations are unavoidable.Regardless of how well the modeler documents the simulation, the inputs will be subject to changes based on professional opinion. After all, levels of education and experience will vary, sometimes vastly, among the professionals involved. Further, there may be honest but significant errors in communicating inputs, and changes in standard operating procedures may take effect mid-project.

Despite these potentially stressful iterative revisions the meticulous modeler can take comfort in one irrefutable fact: He or she has been asked to quantify the emissions from a complex set of processes, the dynamics of which no two professionals are likely to agree upon entirely. Therefore, due to the high likelihood of necessary changes, allocate resources for iterations and clarify in your scope that iterations will be limited to two, three or some other single-digit value.

Emissions estimating and permitting -- separate but equal.Let’;s face it. Emissions inventories are abstract, mind-bending challenges for the most seasoned professionals. Granted, once a solidly representative emissions inventory has been established for a plant, making modifications can be a relatively easy exercise. However, getting a plant’;s first inventory under its belt is difficult and time consuming. Because of this, endeavor where possible to assign the inventory and permitting functions to separate personnel. Each should be competent in his or her area and should work well in collaboration.

Create paper and electronic trails.The breadcrumbs, that is, the documentation that you lay down while preparing your emissions inventory should make Hansel and Gretel proud. A good question to ask throughout the inventory is this: Could I put this down and in three months come back and reproduce every step by following the documentation? If you cannot answer yes to this question, you should not expect anyone else to be able to reproduce what you have created. Sadly, as vital as reproducibility is to a good emissions inventory, it is also one of the most neglected aspects. And it can haunt a plant well into its future, perhaps requiring estimates to be recreated that were good to begin with but were either poorly documented or simply lost through carelessness.

Master emissions estimates
There are several software programs on the market to help you perform an emissions inventory. One program I have used is Emission Master, which estimates emissions based on vapor liquid equilibrium. It produces sound estimates of emissions that conform to EPA-approved calculation methodologies. In fact, EPA agents requested information from the maker of Emission Master when developing its methodologies.

In simplest terms, using software to estimate process emissions is similar to a game you might have played as a child. Remember how you made a flip book by rippling a set of index cards with a figure on each that is drawn slightly different at the top? The rippling appears to set the figure in motion. It is the same with emissions estimation software; only the cards in our estimate are activity models. Prepare enough activities in the proper order and you will have accurately simulated process emissions. Below are fundamental emission activity models found in Emission Master that when properly used, can represent virtually any process imaginable. You will find that other software programs have similar models.

Charge or vessel filling.When a liquid is pumped from one vessel to another, emissions are measured as the volume of headspace vapors displaced in the receiving vessel by the incoming liquid from the sending vessel. The receiving tank may be empty, partly filled with liquid, full of liquid, or even appear to be flooded with liquid. To provide a conservative estimate of the displaced emissions, the program uses the highest temperature or concentration of the compound, whether that value occurs in the sending or receiving vessel.

Depressurization.When the pressure in a vessel is isothermally reduced, volatile components in the vessel headspace discharge as emissions. If this occurs in a matter of minutes, the amount of noncondensables that leak into the vessel during evacuation can be discounted and volatile organic compound (VOC) losses can be accurately estimated. The software uses a depressurization model that considers the noncondensables leaving the vessel to be saturated with its headspace volatiles. Vapor pressures for these volatiles are calculated using the midpoint pressure between the initial and final pressures of the activity. Should the user select a final system pressure that is lower than the saturated vapor pressure of the compound or mixture, the software displays a warning. This warning enables the user to adjust the final pressure so that emissions represent a final state of liquid vapor equilibria rather than a boiling of the entire contents.

Purge/Sweep.Liquid storage vessels are frequently charged with noncondensable gas to prevent an explosive atmosphere, eliminate oxygen to avoid rancidity, or to move the liquid from one vessel to another. The noncondensable is normally saturated with the volatile components of the liquid. Given the purge or sweep rate, the program provides vessel emission rates based on Raoult’;s Law.

Heating.In the heating activity, the modeler must account for two interrelated phenomena: the thermal expansion of the vessel gas space and the increase in vapor pressure of the stored liquid. The software accomplishes this by using thermal displacement and average vapor concentrations at the initial and final temperatures.

Storage tank.Both breathing losses and total volumetric displacement (i.e., working) losses are calculated from the simple input of tank dimensions, vent relief settings, annual turnovers and local meteorology.

Beyond these fundamentals, most emissions estimating programs can simulate special-case emissions from vacuum operations, gas evolution, solids drying and empty vessel purges. With any of these activities, the modeler can set up process condensers and aftercondensers, along with scrubbers and other control devices, to more accurately estimate process emissions.

Estimate with conviction
Combining powerful emissions estimating software with the holistic principles can result in an emissions estimate of convincing accuracy. Unfortunately, a modeler who shorts himself on the application of software, principles or both will find his estimate second-guessed and perhaps rejected, or subject to nerve-wracking rework. With this article and the proper software, you are forewarned and forearmed, so don’;t let it happen to you.

Don Dickerson is a registered chemical engineer and senior environmental engineer for Lockwood Greene, a division of CH2M HILL. He works in Augusta, Ga. E-mail him at ddickerson@lg.com.

 

 

Share Print Reprints Permissions

What are your comments?

You cannot post comments until you have logged in. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments