Hey fireguy, sorry it took me so long to get back to you. The process I am following goes like this. I am taking karat gold (evenly melted scrap) in sample weights of .25 to .5 grams and wrapping it up in a one inch by one inch lead foil with more than enough silver to inquart it. This is placed in a small bone ash cuppel and placed in the oven. I let the temperature get from 1700 to 1800 degrees Fahrenheit. I have noticed It drops to about 1600 degrees when I open the door to check the cuppels. After it has melted into a bead I pull it out and let it cool. Then I place it in a porcelain crucible and begin the nitric dissolve. I start with a three parts water one part nitric (The nitric acid is 70% pure) then my second run I go to three parts nitric and one part water. I have no problem getting it to dissolve and I am very careful in not losing gold during the filtering. I wash my gold and dry it then I weight it. I am careful not to leave any in the crucible. It is always coming out about two to four percent low, sometimes more. I have been refining for years and have the process down very well. I have refined many batches that the samples came from and my refining yields is always higher than the fire assay shows. My XRF also shows there to be more purity there, and the other party's assay is always coming out higher. So I have no doubt that it is something I am doing wrong in this. I have noticed a problem of getting the metals to mix together in the bone ash crucible while in the oven. Maybe I am leaving some in the crucible? I am just stumped on what is causing me to be off on my readings. I really appreciate you helping me, I am lost on this and ignorant to the topic of fire assays. Thank you very much.
OK. While the process you describe would be OK to obtain a rough assay, it is missing some important steps.
1st, it sounds like you are cold loading the cupel (putting the lead on a cold cupel and putting into the oven). This is not good assay technique, especially for this type of bullion analysis. Preheat the cupel up to full temperature and let it soak at temperature for about 15 minutes. Then place the lead button onto the cupel in the furnace.
2nd. It sounds like you are cupelling too hot. I would not exceed 1700F.
3rd. Your first analysis would be a "preliminary", which is done to determine the approximate gold and silver composition of the sample.
4th. Your final analysis should have the proper amount of silver added (not an excess) as determined by the preliminary. This amount of silver is typically 2.5-3 times the amount of gold (don't forget to include the silver in the sample). You will also use the result from the preliminary analysis to determine the composition of your proof sample, which will be made up of high purity gold, silver, and copper. The results of the proof will be used to correct for any gains or losses in the analysis.
The SMA Methods Manual II reference I gave earlier is the procedure that was ultimately adopted by ASTM.
A few other important factors:
-Parting acid is typically performed in two steps. The first uses 33% HNO3, for 45 minutes. The second, 66% for 45 minutes.
-Don't boil the parting acid.
-Wash at least 3 times with DI water
-After drying, anneal the gold (1100-1300F)
-Allow to cool before weighting on microbalance
Hopefully this gets you on the right track to obtain the good results you are looking for.