Good post.
But, I have to disagree with you on this point. (no pun)
The melting point will tell you the purity of a compound in two ways: the relative temperature of the melting point, and also the range of beginning to melt through completely melted.
A pure compound will have a range between 1-2 degrees of the actual melting point. While a mixture of compounds will drag that range out to several degrees based upon the melting point and the amount of the impurity.
Even if the impurity has a similar meling point, it's presence will disrupt the lattice of the target compound causing a slightly lower beginning point, as well as a larger range until the entire sample is melted.
Hope this helps.
I know what you are saying, but it's hard to tell for the untrained eye. It happens fairly quickly and if you don't know what to look for, its not exactly obvious. I've done a fair bit of melting point testing, and I still have trouble with it. Practice makes perfect, I guess. When you say a slightly lower beginning, I`ve seen that happen many times. Would you need a melting point apparatus which is accurate at 0.1C or is 1C enough to capture the change? You really think a minor trace impurity like a heavy metal is enough to disrupt the crystal structure of the sample enough for it to be captured by a typical melting point apparatus? I`m not debating with you, btw, I`m honestly interested in your answer.
COA from China means ziltch...nothing at all. I've received a COA indicating 98.5% and it would come back at 92%. I've received a COA once, and the product was inactive.