Page 207 - Membranes for Industrial Wastewater Recovery and Re-Use

176 Membranes for lndustrial Wastewater Recovery and Re-use

The system hydraulic resistance has a significant impact on the specific energy
demand (i.e. the energy consumption per unit, permeate product volume). High
retentate flow rates reduce concentration polarisation, and hence may be
expected to increase the flux, but also increase the pressure drop along the length
of the membrane element. A high retentate flow rate is associated with a large
transmembrane pressure (TMP), and hence a higher flux and recovery.
However, this results in an increase in CP. Clearly, there is a balance to be struck
between the pressure and the membrane area requirement, such that the flux is
not unreasonably high at the module inlet and the cross-flow velocity not
reasonably low at the outlet.
It is normal practice to provide a high membrane area by linking several
membrane elements in a single module, and by staging (Section 2.4.1) when the
retentate flow decreases by 33 or, more usually, 50%. Many RO plant comprise
so-called “2-1” arrays, where the flow entering the second stage of the array is
half that of the first stage. Staging reduces the membrane area requirement but
can also lead to substantial retentate pressure losses. An intermediate pump may
then be required to increase the feed pressure to the second stage: this process is
sometimes referred to as a “two stage concentrator” where, in this instance, the
stage may include sub-stages.
Another way of increasing the overall recovery is to recycle a fraction of
the retentate to the feed inlet port of the bank, producing what is normally
referred to as a “single pass with recycle” process configuration. This
increases the retentate flow rate as a result of combining the recycle stream
with the feed stream, such that it is possible to increase the recovery while
still maintaining the same degree of concentration polarisation. However, the
increase in the retentate concentration has a negative impact on both
the scaling propensity and the permeate water quality. It is also possible
to treat the permeate from the first stage, producing a “twin pass” system
(Fig. 2.21).

4.1.3 RO design software
The design of an RO plant involves the adequate specification of parameter
values and an appropriate choice of process design. As a result of the
multitude of parameters, complex interactions and different design variants, it
is not possible to produce RO software able to automatically produce the ideal
RO design for a specific application. The design process essentially proceeds by
trial-and-error through selection of different types of membranes, design
variants and operating parameter values, with the software package
indicating when the selected design is unreasonable and (usually) identifying
the cause. The packages therefore all demand a rudimentary understanding of
the RO process and the effect of the design parameters. Provided this is the
case, the packages are normally quite user-friendly and perhaps even didactic
in construction. Most have an input “wizard” that guides the user through
each of the design elements in sequence. The individual inputs are listed
below.