the method base on extraction with HCL 0.2M and heat for 30minute at 90degc,then adjust to ph 5 ,clean the sample and inject on c-18 column .
all other part of validation results are perfectexcept of recovery that range from 80-93%.
i will be glad to get some advice

For this type of 'trace' analysis, why are you not happy with the results? The recovery range is fairly narrow, and reasonably close to 100%, considering all the steps involved, heat, and a clean-up of some kind. Is there someone else who uses this method that gets better results? Is it an 'official' method that has data supporting better results? Or is it failing the validation because of arbitrary standards? Lacking any other info, maybe this method is working as well as can be expected.

A method validation shows what a method will do in an individual analyst's (or team's) hands. If it does not meet some standard, then it must be dissected until the problem step(s) is isolated and corrected. There are many, many ways of doing this, depending on the method complexity, likely problems, etc.

This is where a chemist earns their money, and why this field can be so interesting, entertaining, and frustrating.

Two possible problem areas:
1. What is the stability of riboflavin at 90 deg C in 0.2 M HCl?
2. What does it mean to "clean up the sample"?

1.riboflavin is very stable to heat I see procedures use autoclave and 120deg c.
it is very sensitive to light so i use dark bottle and try not to exposed it as possible.
2.clean up mean percipitation of fat and proteins ;cetrifugation and filtration.
To the first one that answer me ,thank you for encoragment.

Hmm...
1. 120 deg C at pH 7 is not the same as at pH 1.
2. it is not impossible that the riboflavin got stuck in the precipitate.
Still two things to look at. However, the other colleague is quite right, 80 to 93% recovery is not bad.