A ground-state approximation (GSA) is employed to model the structure
of an adsorbed layer of homodisperse and polydisperse polymer. The mod
el uses the basic assumption that the volume fraction at a distance z
from the surface of a component with chain length N can be written as
the product of the square of an eigenfunction g(z) and the N-th power
of an eigenvalue e(epsilon). This approximation implies the neglect of
end effects (tails): only loops are considered. For a homodisperse po
lymer, the eigenvalue is defined through epsilon N = In(1/phi(b)), whe
re phi(b) is the bulk solution concentration. The eigenfunction can be
written in terms of two parameters: a ''proximal'' length D which thr
ough the boundary condition may be related to the adsorption energy, a
nd a ''distal'' length which is inversely proportional to root epsilon
. For a polydisperse polymer, D is the same as for a homodisperse poly
mer, but e has to be computed from an implicit equation which involves
a summation over all chain lengths present. The contribution of each
chain length N in a mixed adsorbed layer is obtained by weighting with
e(epsilon N). This approximate analytical model gives results which a
re in good agreement with numerical self-consistent-field calculations
. Examples are given to illustrate the applicability of the model to p
olydisperse systems. These include adsorption preference of long chain
s in polymer mixtures and the difference between adsorption and desorp
tion isotherms in polydisperse systems. Simple expressions are obtaine
d for the chain length characterising the transition between (long) ad
sorbed and (short) non-adsorbed chains and for the width of the transi
tion zone.