570
A. Golomb
Fig. 1 illustrates the effect upon flux rate* of increasing concentration at operating
pressures of 600 psi and 1000 psi respectively, for laboratory prepared rinse waters and
for a diluted industrial Watts bath. It is necessary to determine the practical upper limit
of concentration that can be treated with the particular membrane employed, since this
factor is important in designing a recycle system in which chemicals are returned to the
plating bath as a concentrated solution. Apparently the practical upper limit for
concentration by reverse osmosis lies in the region of about one-third of the Watts bath
concentration. Consistently high rejection efficiencies for nickel ions, generally 99+ per
cent, were observed throughout these runs.
The different flux rates noted for laboratory and plant solutions are attributed to
relatively lower concentrations (and osmotic pressures) of the particular Watts bath
samples obtained from industrial sources.
ECONOMIC EVALUATION FOR A TYPICAL PROCESS DESIGN
Golomb (1970) outlined the rationale for design of a process, based on RO, for
recovery of nickel values from Watts rinse waters by reverse osmosis. As representative of
a typical industrial plating operation, a three stage counter-current rinse system was
chosen in which deionized water is used for rinsing, the concentration of total dissolved
solids (TDS) in the final rinse is 0.002 oz/gal, and part of the solution from the first tank
is returned to the plating bath to balance evaporative losses, the remainder going to drain.
Two levels of operation were considered, one in which the dragout is 500 gal/day and
evaporation from the plating tank is 5000 gal/day and the second at one-tenth of these
values.
For the higher level of operation (dragout rate 500 gal/day) the chemical loss as nickel
is 240 lb/day and as plating salts, including water of hydration, is 1200 lb/day. These are
equivalent to dollar losses of $310/day and $550/day respectively, based on chemical
costs of Ni - $1.30/lb; NiS0 4 .6H 2 0 -$0.46; NiCl 2 .6H 2 0 - $0.62 and H 3 B0 3 -
$0.10/lb. The losses are one-tenth of these values for the smaller scale operation (dragout
rate 50 gal/day).
A reverse osmosis treatment may now be applied to the nickel rinse waters such that
the volume of concentrate returned to the plating tank again equals the evaporative
losses. With the entire permeate being discharged as a waste stream, the total daily
amounts and concentrations of the various chemical components in the process streams
can be calculated. More than 99 percent of the nickel salts is returned to the plating bath,
but some deficiency of boric acid is anticipated, which can be corrected by make-up. The
waste effluent contains only 32 ppm of nickel, the major component being boric acid. As
reported by McKee and Wolf (1963) most water authorities have indicated no strong
objection to the presence of boric acid in effluents being discharged to water courses,
(except where the water is to be used for irrigation of certain crops), but nickel is not
generally permitted at levels above 5 ppm in effluents entering sewage lines. A certain
degree of dilution would therefore be required, as appropriate.
Some saving in water and a reduction in the chemicals discharged to waste can be
achieved by recycle of part or all of the permeate to the rinse line. This will result in a
higher TDS in the final rinse tank which would need to be considered in relation to its
* gfd = flux rate, U.S. gal/ft
2 of membrane/24 hr day.
A. Golomb
Fig. 1 illustrates the effect upon flux rate* of increasing concentration at operating
pressures of 600 psi and 1000 psi respectively, for laboratory prepared rinse waters and
for a diluted industrial Watts bath. It is necessary to determine the practical upper limit
of concentration that can be treated with the particular membrane employed, since this
factor is important in designing a recycle system in which chemicals are returned to the
plating bath as a concentrated solution. Apparently the practical upper limit for
concentration by reverse osmosis lies in the region of about one-third of the Watts bath
concentration. Consistently high rejection efficiencies for nickel ions, generally 99+ per
cent, were observed throughout these runs.
The different flux rates noted for laboratory and plant solutions are attributed to
relatively lower concentrations (and osmotic pressures) of the particular Watts bath
samples obtained from industrial sources.
ECONOMIC EVALUATION FOR A TYPICAL PROCESS DESIGN
Golomb (1970) outlined the rationale for design of a process, based on RO, for
recovery of nickel values from Watts rinse waters by reverse osmosis. As representative of
a typical industrial plating operation, a three stage counter-current rinse system was
chosen in which deionized water is used for rinsing, the concentration of total dissolved
solids (TDS) in the final rinse is 0.002 oz/gal, and part of the solution from the first tank
is returned to the plating bath to balance evaporative losses, the remainder going to drain.
Two levels of operation were considered, one in which the dragout is 500 gal/day and
evaporation from the plating tank is 5000 gal/day and the second at one-tenth of these
values.
For the higher level of operation (dragout rate 500 gal/day) the chemical loss as nickel
is 240 lb/day and as plating salts, including water of hydration, is 1200 lb/day. These are
equivalent to dollar losses of $310/day and $550/day respectively, based on chemical
costs of Ni - $1.30/lb; NiS0 4 .6H 2 0 -$0.46; NiCl 2 .6H 2 0 - $0.62 and H 3 B0 3 -
$0.10/lb. The losses are one-tenth of these values for the smaller scale operation (dragout
rate 50 gal/day).
A reverse osmosis treatment may now be applied to the nickel rinse waters such that
the volume of concentrate returned to the plating tank again equals the evaporative
losses. With the entire permeate being discharged as a waste stream, the total daily
amounts and concentrations of the various chemical components in the process streams
can be calculated. More than 99 percent of the nickel salts is returned to the plating bath,
but some deficiency of boric acid is anticipated, which can be corrected by make-up. The
waste effluent contains only 32 ppm of nickel, the major component being boric acid. As
reported by McKee and Wolf (1963) most water authorities have indicated no strong
objection to the presence of boric acid in effluents being discharged to water courses,
(except where the water is to be used for irrigation of certain crops), but nickel is not
generally permitted at levels above 5 ppm in effluents entering sewage lines. A certain
degree of dilution would therefore be required, as appropriate.
Some saving in water and a reduction in the chemicals discharged to waste can be
achieved by recycle of part or all of the permeate to the rinse line. This will result in a
higher TDS in the final rinse tank which would need to be considered in relation to its
* gfd = flux rate, U.S. gal/ft
2 of membrane/24 hr day.