Batch reactor emissions

[atiaran]

If you were starting from scratch, how would you approach evaluating the amount of emissions occuring from a batch reactor that processed different chemicals?

I have about a dozen reactors to do this for, each of which produce several different chemicals throughout the year. I'm starting small and doing one reactor. I have a production number of total pounds made last year, and the number of days each reactor was used per chemical. I figure if I get the typical lb/batch, I can calculate how many batches there are made. Would then doing a material balance around the reactor be a sufficient estimate?
We don't have source test information on all of the reactors for emission factors.

 

[StoneCold]

Atiaran
We just recently went through this with our plant. We run multiple processes in multiple reators as well. Our approach was to develop an emmision factor based on the major solvents we buy. So for solvent B I know that it is unloaded from the truck, loaded into a reactor, distilled and then drummed, etc. So I developed a factor for all the common processing steps that all my solvent went through. Then I added a product specific factor I know that at my plant solvent A is used in Reaction 1 and Reaction 2. Last year 50% of the solvent was used for reaction #1 so I am going to develop an emmision factor for Rx1 and Rx2 and give them each a 50% weighting in the final emmission factor. Then you just sum all the differnt steps up and get a factor. The final factor is easy to manage because you just track the amount of solvents you buy. You can do the same thing for all your hazardous materials. Just make the factors work on the amount of material you buy. That makes it easy to keep track of your emmisions because almost everyone knows what they buy.

Regards
StoneCold

 

[atiaran]

So in developing your emissions factor, did you use a % yield or % conversion for amt of product drummed compared to amt solvent purchased?

 

[StoneCold]

No. When we made the factors we built them out of operational steps but not conversion. So the final factor was a summation of loading, any venting during the reaction step, distillation, and then drumming or where ever the product went. If we were to have a product that only partially reacted and then was drummed the unreacted material would/could be a contributor to the emmision factor of the drumming operation. Documents and programs that were a big help were US EPA program called Fire 6.25 and the EPA document called AP-42. We also used the list of materials considered VOC's and HAP's for our state.
We have a large number of products, intermediates etc so we grouped our factors in ways that made sense for us. I think in the end we have only ten factors that when multiplied by a bulk solvent or HAP of a large size give us our tons of emmisions. Our total VOC emmisions are only about 25 tons and our HAPs are about 16 tons if that gives you any comparison to your own project.

Goodluck I hope I answered more questions than I raised.

Regrards
StoneCold

 

[ChemE123]

Stonecold's method may be acceptable in the case of a plant-wide emissions cap, but in determining specific emissions for a process/emission unit it may not provide the level of detailed information a regulatory agency would typically accept. Your intial thought of a material balance around the unit would be more appropriate in such a case. What goes in either gets reacted, or it comes out. Be sure to account for components of interest in all waste streams. If it isn't in the product or liquid/solid waste, it can be assumed to be in the air emissions. Once you have this calculated for each reaction scheme, you can scale as appropriate to your annual emissions. Many state regulatory agencies will want to know not only what you emit (types and amounts), but where it is emitted (which process unit), unless the facility is under a plant-wide emissions cap.