The occurrence of liquid dropout in natural gas pipelines may cause op
erational problems during storage, transport, and processing. Therefor
e, the availability of a model that accurately predicts the amount of
liquid formed is of great importance for the natural gas industry. The
objective of this study is to develop a thermodynamic model for the a
ccurate prediction of the amount of liquid formed in natural gas pipel
ines at transportation conditions. As input, the model requires an acc
urate gas analysis. A modified Peng-Robinson equation of state was sel
ected for the phase equilibrium calculations. Interaction parameters w
ere optimized from experimental data at conditions of practical intere
st, i.e., at pressures 10 < p < 70 bar and at temperatures 250 < T < 2
90 K. For a number of ''keysystems,'' the interaction parameters were
calculated from new accurate solubility data of heavy hydrocarbons in
some of the main constituents of natural gas like methane and nitrogen
. Also, an extensive experimental program was carried out to study the
influence of minute amounts of nitrogen, ethane and carbon dioxide in
methane on the solubility behavior of decane in these gas mixtures. F
rom a sensitivity analysis, it could be concluded that the liquid drop
out is influenced mainly by the concentration and characterization of
C-7-C-13 fractions. In this work, two characterization procedures to r
epresent these fractions are compared. For two types of lean natural g
as, the model predictions are compared with field measurement data, re
cently supplied by the Dutch natural gas industry.