You should know more about this than I, but Vf is an extremely variable number. It is indeed possible that the manufacterer has varied the reciepe somehow. What color, what is the average Vf, and what is the range?
Also, how are you bonding to them? Wirebonding is a subject I know something about.
This is a bit like saying "Is it possible that car fuel consumption problems could be related to the engine?" Of course, as Bill Marsden points out there are other possibilities, but in the absence of other information we can't rule out issues with the dies.
Bad/high resistance metal used for the LED leads (I have plenty of LEDs that have leads made of steel).
Poor/improper bonding of the leads to the die
Defective die.
The Vf of LEDs can vary widely anyway, even in the same batch. I've seen differences exceeding 10% on LEDs that came from the same reel of tape.
Well, the Vf specs <1.8V in 100 mA.
we use aluminum wirebonding, but the Vf failures came from the same wafer batch that's why I think it is related to die/wafer.
The LEDs should have been tested and disposed at source, if they are proper LEDs from a reputable source. They should have very good Vf and no failures.
If you buy bulk LEDs from ebay you should be aware that many are rejects that have all failed testing or are a reject batch where the failures are > than the official allowed rate so the whole batch was dumped. And since good LEDs are cheap I hope you got the reject LEDs VERY cheap...