# Dispersion in Chromatography Columns - The Golay Equation > Page 72

Taking a
value of 2.5 x10^{-5} for D_{m} (the diffusivity of benzyl acetate in *n*-heptane)
equation (52) can be employed to calculate the curve relating (H) and (u) for
an uncoated capillary tube. The results are shown in figure 17. It
is seen that the Golay equation produces a curve identical to the Van Deemter
equation but with no contribution from a multipath term. It is also seen that
the value of (H) is solely dependent on the diffusivity of the solute in the
mobile phase and the linear mobile phase velocity. It is clear that the
capillary column can, therefore, provide a simple means of determining the
diffusivity of a solute in any given liquid. The Golay equation (equation (52))
can be put in a simplified form in a similar manner to the equations for packed
columns:-

_{} (54)

Where, _{} and _{}

The form of the HETP curve for a capillary column is the same as that for a packed column and exhibits a minimum value for (H) at an optimum velocity.

Differentiating
equation (54) with respect to (u), _{}

Thus,
when H = H_{min},
then, _{}

and
thus,
_{} (55)

Substituting for (B) and (C) in equation (55)

_{}

or, (56)

It is seen
that, in a similar manner to the packed column, the optimum mobile phase
velocity is directly proportional to the diffusivity of the solute in the
mobile phase. However, in the capillary column the radius (r) replaces the
particle diameter (d_{p}) of the
packed column and consequently, (u_{opt})
is inversely proportional to the column radius.