PVT ANALYSIS FOR OIL 61
consists mainly of methane and ethane, the resulting oil volumes from either
experiment are practically the same. For higher volatility oils, containing a relatively
high proportion of the intermediate hydrocarbons such as butane and pentane, the
volumes can be significantly different. Generally, in this latter case, more gas escapes
from solution in the flash expansion than in the differential liberation, resulting in a
smaller final oil volume after the flash process. This may be explained by the fact that
in the flash expansion the intermediate hydrocarbon molecules find it somewhat easier
to escape into the large gas volume in contact with the oil than in the case of the
differential liberation, in which the volume of liberated gas in equilibrium with the oil, at
any stage in the depletion, is significantly smaller.
The above description is a considerable simplification of the complex processes
involved in the separation of oil and gas; also, it is not always true that the flash
separation yields smaller oil volumes. What must be appreciated, however, is that the
flash and differential processes will yield different oil volumes and this difference can
be physically measured by experiment. The problem is, of course, which type of
experiment will provide the most realistic values of B
o
, R
s
and B
g
, required for relating
measured surface volumes to volumes withdrawn from the reservoir at the current
reservoir pressure and fixed temperature.
The answer is that a combination of both flash and differential liberation is required for
an adequate description of the overall volume changes. It is considered that the
differential liberation experiment provides the better description of how the oil and gas
separate in the reservoir since, because of their different flow velocities, they will not
remain together in equilibrium once gas is liberated from the oil, thus corresponding to
the process shown in fig. 2.9(b). The one exception to this is during the brief period
after the bubble point has been reached, when the liberated gas is fairly uniformly
distributed throughout the reservoir and remains immobile until the critical gas
saturation is exceeded.
The nature of the volume change occurring between the reservoir and stock tank is
more difficult to categorise but generally, the overall effect is usually likened to a non-
isothermal flash expansion. One aspect in this expansion during production is worth
considering in more detail and that is, what occurs during the passage of the reservoir
fluids through the surface separator or separators.
Within any single separator the liberation of gas from the oil may be considered as a
flash expansion in which, for a time, the gas stays in equilibrium with the oil. If two or
more separators are used then the gas is physically removed from the oil leaving the
first separator and the oil is again flashed in the second separator. This physical
isolation of the fluids after each stage of separation corresponds to differential
liberation. In fact, the overall effect of multi-stage separation corresponds to the
process shown in fig. 2.9(b), which is differential liberation, only in this case it is not
conducted at constant temperature. It is for this reason that multi-stage separation is
commonly used in the field because, as already mentioned, differential liberation will
normally yield a larger final volume of equilibrium oil than the corresponding flash
expansion.