James Belcher
Sept 6 1999
NHBB Emissions Inventory
Project/Chemical Information:
I have completed an evaluation of the emissions produced in the composite department of New Hampshire Ball Bearing’s Astro plant. The source chemicals that were focused on were n-methyl 2-pyrrolidinine (NMP), methanol and isopropanol. The NMP compound is found in the polyamide resin, MVK-19. Methanol is also found in the resin, produced in the polyamide reaction during B-stage curing, and found in the mold and mandrel release spray. Isopropanol is only found in the mold and mandrel release spray. MVK-19 is used for almost all of the production parts in the composite department. NMP is the largest constituent of MVK-19 resin, the rest is water, methanol and solids. NMP is a volatile organic compound and evaporates 1.3 times as fast as BuAc and has a molecular weight of 99 g/mol. The solids from the resin do not become airborne during any part of the composite manufacturing process. MeOH, NMP and isopropanol are listed under title 3, section 313 and are subject to reporting in the Toxic Release Inventory. In addition to these reporting requirements, methanol has a CERCLA RQ of 5,000 lbs. MeOH and NMP are New Hampshire Regulated Toxic Air Pollutants (Env_A 1300 and Env_A 1400). The purpose of the inventory is to update emission data gathered in the 1997 emissions inventory.
Method:
The First Stage of the composite manufacturing process where there is a chemical release is the mandrel coating hood. The mandrels require a graphite release coating so that the parts do not adhere to them during B-stage and heat press. The graphite is suspended in 93.25 wt% isopropanol/methanol mix in a 945 ml SureShot spray canister. The graphite spray can was measured at the start of a couple shifts and at the end to determine the amount of spray used and the amount of the alcohol evacuated up the stack.
Another source of emissions is the preheat step. In order for the resin to be impregnated properly to the carbon fiber braid, the resin must be preheated to a certain temperature. This temperature is high enough to evaporate most of the methanol in the resin. The assumption was made that only trace amounts of methanol remained in the resin after preheat. With this assumption made, the daily release of methanol from the preheat oven could be calculated.
If parts are underweight after the cutting stage, they go back to the impregnation stage to have more resin added. If the parts are overweight, the parts are sanded down.
The largest source of emissions in the composites department is the B-stage oven. The wet rods are place in the oven. Since they are wet, there is some dripping of the resin off them that is caught in the bottom of the oven with a drip pan. The most commonly produced part in the composite department is the 4-4021 bushing. These are cured in the oven in 55 rod batches. The mass of the drip pan before one of these batches and after was recorded to estimate the amount of resin that dripped off the rods. The weight of the rods after B-stage was recorded on the Impregnation Tracking and Raw Material Withdrawal Form. A thermal analysis (TGA) test done on the heat press revealed how much NMP was left in the part. An overall process mass balance using the rod wt. In, drip pan deposits, and rod mass allowed me to calculate the total mass of resin up the stack. In order to calculate the mass percentages of the stack flow, another TGA test needed to be done. The test recorded mass changes at certain temperatures. One of these temperature holds revealed how much methanol was in the stack stream. With the mass of the methanol and the chemical equation of the polyamide reaction known, the exact percentages of the emissions could be calculated. All of the mass balances for each component verified the calculations of the output streams.
The second and last step in the composites department that releases NMP is the heat press. The preform is ct and weighed then sent for molding to the heat press. The part loses some mass during the heat press stage. The composite dept. engineer, Allen McIntire, performed TGA tests on a representative sample of parts. He observed a sharp but insignificant weight loss just past 100C, this was attributed to any moisture left in the part cooking off. After 250C there is a larger, more gradual weight loss of about 3%. This is caused by the remaining NMP in the part vaporizing. All of these calculations were done on the representative part, the 4-4021 bushing.