A study investigating RO membrane integrity testing
showed that online conductivity monitoring was less sensitive
than TOC monitoring (Adham et al., 1998). A further
study showed that conductivity was not a good predictor of
viral rejection (Kitis et al., 2003). These studies indicate that
targeting the organic component, as opposed to small ions
via conductivity, is more indicative of membrane
performance. The limitations associated with TOC for crossconnection
detection highlight the need for additional technologies
that offer greater sensitivity to changes in the water
quality and character. It is noted, however, that the implementation
of multivariate techniques for analysing combinations
of these commonly used parameters could
significantly improve the sensitivity of these parameters
despite large variations, as described by Rosen and Lennox
(2001).