Iam looking to develop a method for blood thiamine levels and need to decide on whether to use post column or precolumn derivatisation. I have all the necessary hardware except a purpose make post T piece mixing/reaction coil. How important is the configuration of this reaction coil? - I have considered the reaction time in the lenght of the coil. The maximum fluorescence for the thiochrome is around pH 8.5.

I'm not sure exactly what your question is. The specifics of the reaction coil depend upon the reaction conditions. Generally the smaller the reaction coil the better from a chromatographic efficiency point of view but larger reaction coil volumes are often needed for less favorable reactions. If the reaction kinetics necessitate an excessively large coil the problem can be remedied in one of two ways: increase the reaction temperature (speeds up the reaction)or make use of a knitted reaction coil (reduces band broadening). You can buy reaction coil heaters and knitted reaction coils from a number of sources (e.g. http://www.dionex.com/). In general, best performance is seen with small ID reaction coils, better still if they are knitted.

If you are asking about specialty mixing T's for this purpose, I wouldn't waste your money on any of these products. I have never been able to find any benefit from any of the products I have tested.

I guess my question was can I make my own reaction coil (does it have to be a coil?) the length of which is related to the flow rate and reaction time. Does all the mixing occur at the T connector or does the orientation of the following tubing contribute to the mixing and hence the completeness of the reaction.

The trick of using a specialized reaction coil is used to suppress the bandspreading in the reaction coil. If you have a very fast reaction (e.g. 15 seconds), any ol' piece of tubing will do. If you have longer reaction times, you should consider a specialized geometrically deformed tubing as a reaction coil. The trick has nothing to do with mixing, but geometrically deformed tubes are used to reduce the bandspreading. With respect to this, they can be quite powerful (often a factor of 10 reduction in bandspreading compared to a straight tube).