INDUSTRY CONVERSATION REGARDING THE SOURCE OF HIGH HEX CHROME EMISSIONS
Carlos E. Davis, Certainteed/St. Gobain                             14 April 2011
This call was made because I was looking over the test report that BTEC did on KCK  furnace...K2 furnace..at Kansas City.  I was wondering if there was a problem with the testing, and if the test was done correctly. Mr. Davis confirmed that the test was done correctly; that the math was checked on the calculations.  I led in that I was interested in where all this chrome was coming from. By the testing, and the industry's test agreement that the tested furnace was representative of the K1 furnace,  I calculated over 900 pounds a year total chrome from the KCK plant.  This was higher than Mr. Davis remembered, but confirmed that they had a lot of chrome.  He was not aware that 92% of the chrome was hexavalent.
In a glass furnace, you feed the fine sands and by that I mean actual sand with other refined and purified raw materials in a mixture into the furnace.  It's a continuous furnace.  The furnace temperature is about 1560⁰ C.  At the surface of the glass, is the metal-line glass contact.
Now the furnace itself extends above the surface of the glass..the metal line glass contact.  We have constructed the K2 furnace from what is called `High Chrome' or sometimes also called `Jade Green' refractories.  They are just that color, a deep beautiful green color.  They are that color because they are made of between 98-96% chrome.  We construct our furnace using noncorrosive materials because you get a much longer life from your furnace. In older designs, you have to rebuild a glass furnace at least every 7-10 years.  You can add on to the outside of the furnace (meaning additional refractories) up to a point (2 layers), but after that, you have to shut it down and tear it out. 
In the K2 furnace you have glass to a depth of a little more than a meter deep.  Then to the top of the straight sides of the furnace, it's about 2 meters, so there's this additional straight wall up to the beginning of what we call the `crown', this kind of domed curved top of the furnace.  That's the `furnace crown'. The surface of the glass is about 70 square meters (70m[2]).
The molten glass is very corrosive.  So there is high corrosivity and errosivity.  And as the molten glass moves through the furnace, it starts to eat away at the refractories, and it doesn't really eat evenly, it eats little holes in the refractory wall inside the furnace.  So these are like little caves, little holes.  And so the surface area of the refractory wall is actually increasing with the life of the furnace.  Because the molten glass continues to move through these caves or holes, it continues to erode even larger areas inside the refractory wall from the inside of the furnace reaching out.  So when that outer wall gets hot, you just add some more refractory onto the outside. Otherwise you'd have a real bad situation, a breakthrough and you can't have that. So your emissions of chromium are actually increasing, not decreasing over the life of the furnace due to this ever-increasing surface area being exposed to the molten glass. So the degradation accelerates with the age of the furnace. We didn't realize that this was almost all hexavalent chromium, we've only tested total chrome. That's a big surprise. 
K2 furnace at Certainteed in KCK has a planned shutdown date of 2013.  This furnace is on it's final end run.  

Update 1 from Region 7:
Certainteed is planning to shift all products to the K2 line, but that line will only hold up for another year.
Update 2 from Region 7:
We found some 2005 testing in the NEI.  We don't know the test method they used, and the test report is not in our files, but the results of that testing were 0.0166 TPY of total chrome, which is about 300 pounds a year, facility-wide.  We don't know what fraction of that total chrome is hexavalent chrome; we don't know which furnace was tested, or if one was tested as representative of the other.  So there are a lot of unknowns with this test, but this shows that even in 2005, these high levels were being emitted.

