580
Discussion
Author's Reply
While it is agreed that Reverse Osmosis economy depends on membrane lifetime, this factor is
more critical for straight water demineralization than for certain industrial applications, in which the
value of the recovered dissolved materials can more than offset the economic losses associated with
reduced membrane lifetime.
The industrial-scale experiments reported in this paper pertain to an intermediate stage of development, covering a 3-month trial period. It is agreed that, from the process data obtained, it is difficult
to predict the ultimate membrane lifetime with accuracy. As a first approximation, an 18-month
lifetime was selected - this being less than half the working lifetime of membranes in current use for
the desalination of brackish water. However, even if, as Dr. BERNARD suggests, a membrane lifetime
of 9 months and a 50% overdimensioning of the Reverse Osmosis unit were to apply, it can be readily
calculated that the nickel recovery process would remain economically attractive at each of the 3
levels of operation considered.
The ambient temperature was consistently high in the plating shop during the field trial (>32 C).
The Reverse Osmosis unit was the standard model marketed by the supplier at the time. High
operating temperatures (over 35 C) can reduce the effective working life of cellulose acetate membranes in two principal ways:
(i) by increasing the rate of hydrolysis, marked by an increase in flux rate and loss of rejection
efficiency,
and (ii) by increasing the rate of membrane compaction, with loss of flux rate.
Another factor, which reduces the membrane flux rate, is fouling by deposition of ferric
hydroxide. Thus it would appear that a number of factors were simultaneously at work in the system,
each tending to affect the flux rate. Since the nickel rejection efficiency remained consistently high
throughout (99.8%), hydrolysis does not appear to have occurred to a serious extent. The acid flush
procedure was found effective in restoring about half of the early loss in flux rate, which seems to
suggest that the unrestored loss of flux was caused by compaction of the membrance at the high
temperature of operation.
To prolong membrane lifetime, it is recommended that an inexpensive, simple heat exchanger be
incorporated into the Reverse Osmosis system, utilizing cooling water which is readily available in
most electroplating installations.
G.E. Eden, UK.
Was the plant still in operation and if not had the author's opinion of the recommended method of
cleaning changed?
Reply by A. Golomb
The reverse osmosis nickel reclamation system is not in operation at the present time. During the
latter stages of the trial period, however, it was found that improved cleaning could be attained by
occasional flushing with a 5% W/W aqueous solution of citric acid or, better still, with a 1% W/W
solution of ethylenediamine-tetra - acetic acid disodium salt, an excellent (if more expensive)
sequestering agent.
Discussion
Author's Reply
While it is agreed that Reverse Osmosis economy depends on membrane lifetime, this factor is
more critical for straight water demineralization than for certain industrial applications, in which the
value of the recovered dissolved materials can more than offset the economic losses associated with
reduced membrane lifetime.
The industrial-scale experiments reported in this paper pertain to an intermediate stage of development, covering a 3-month trial period. It is agreed that, from the process data obtained, it is difficult
to predict the ultimate membrane lifetime with accuracy. As a first approximation, an 18-month
lifetime was selected - this being less than half the working lifetime of membranes in current use for
the desalination of brackish water. However, even if, as Dr. BERNARD suggests, a membrane lifetime
of 9 months and a 50% overdimensioning of the Reverse Osmosis unit were to apply, it can be readily
calculated that the nickel recovery process would remain economically attractive at each of the 3
levels of operation considered.
The ambient temperature was consistently high in the plating shop during the field trial (>32 C).
The Reverse Osmosis unit was the standard model marketed by the supplier at the time. High
operating temperatures (over 35 C) can reduce the effective working life of cellulose acetate membranes in two principal ways:
(i) by increasing the rate of hydrolysis, marked by an increase in flux rate and loss of rejection
efficiency,
and (ii) by increasing the rate of membrane compaction, with loss of flux rate.
Another factor, which reduces the membrane flux rate, is fouling by deposition of ferric
hydroxide. Thus it would appear that a number of factors were simultaneously at work in the system,
each tending to affect the flux rate. Since the nickel rejection efficiency remained consistently high
throughout (99.8%), hydrolysis does not appear to have occurred to a serious extent. The acid flush
procedure was found effective in restoring about half of the early loss in flux rate, which seems to
suggest that the unrestored loss of flux was caused by compaction of the membrance at the high
temperature of operation.
To prolong membrane lifetime, it is recommended that an inexpensive, simple heat exchanger be
incorporated into the Reverse Osmosis system, utilizing cooling water which is readily available in
most electroplating installations.
G.E. Eden, UK.
Was the plant still in operation and if not had the author's opinion of the recommended method of
cleaning changed?
Reply by A. Golomb
The reverse osmosis nickel reclamation system is not in operation at the present time. During the
latter stages of the trial period, however, it was found that improved cleaning could be attained by
occasional flushing with a 5% W/W aqueous solution of citric acid or, better still, with a 1% W/W
solution of ethylenediamine-tetra - acetic acid disodium salt, an excellent (if more expensive)
sequestering agent.