Chemical Engineering May 2012 - 58

Cover Story
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0.03
0.02
0.01
Oct.
2003
Feb.
2005
July
2006
Nov.
2007
March
2009
10
15
20
25
5
Oct.
2003
Feb.
2005
July
2006
Nov.
2007
March
2009
FIGURE 8. The project team monitored corrosion rate (left; mm/year) and adhesion rate (right; mg/cm2 per month) when reusing
treated WWTP as makeup water to the cooling tower of a large reinery
tists from the membrane vendors, as
well as the Environment Institute of
the Sinopec Beijing Research Institute
of Chemical Industries (BRICI).
The goals of the testing program
were threefold: (1) to investigate the
effects of wastewater qualities on the
membrane performance and selection;
(2) to quantify the performance
of both feedwater pretreatment and
membrane filtration, and identify
the optimum membrane design parameters
and operational conditions;
and (3) to characterize the membrane
fouling and cleaning cycle, including
the cleaning liquid formulation,
cleaning procedure, and control indicator
and method. Figure 9 illustrates
the membrane testing procedure.
Our testing program considered
water qualities from four representative
regions of Sinopec's refining and
chemical subsidiaries, and chose the
appropriate feedwater pretreatment
and membrane filtration scheme for
each region.
Tables 2 and 3 list the water qualities
after UF and RO treatment of the
WWTP effluent together with the desired
water quality for cooling-tower
makeup water. Two observations are
evident. First, not all of the WWTP
effluents require RO treatment to
reach acceptable qualities for reuse
as cooling-tower makeup water. Second,
the water qualities after both UF
and RO treatment are typically of sufficient
quality for use as cooling-tower
makeup water or boiler makeup water
(desalted water).
The key results from this feedwater
pretreatment and membrane experiment
program were as follows: 1)
sand filtration and aerated biological
filtration are effective pretreatment
schemes; 2) water recoveries for UF is
above 84% and for RO is above 70%,
salt removal by RO is above 97%, and
typical membrane fluxes for UF are
40 to 75 L/m2·h, and for RO are 20
to 25 L/m2·h; 3) chemical washing for
both UF and RO membranes takes
place once every two to three months;
4) UF membranes can last for five
years, and RO membranes need replacement
after two years; and 5) the
operating cost for feedwater pretreatment
and UF-RO membrane filtration
is about $0.50/ton of clean water
recovered, and the operating cost for
feed pretreatment and UF membrane
filtration is about $0.15/ton of clean
water recovered.
With these onsite testing results
and experiences, our project team
was able to apply UF and RO for deep
treatment of WWTP effluents for
reuse as cooling-tower makeup water,
or as boiler makeup water in over 20
subsidiaries. With an increasing number
of UF and RO applications, their
unit investment cost has continued to
drop in recent years.
Our project has also applied the
technology of membrane biological
reactors (MBRs) that combine membrane
filtration for salt removal with
biological wastewater treatment in
several petroleum refining and chemical
plants.
To further minimize the wastewater
discharge, our project team continues
to investigate the use of multi-effect
membrane distillation and other ap40
CHEMICAL ENGINEERING WWW.CHE.COM MAY 2012
proaches for treating and reusing the
brine water rejected by RO.
In order to minimize water consumption
and to approach zero-liquid
discharge in chemical and biofuel
plants, it is essential to apply the
same water-saving practices as those
used in the Sinopec project, particularly
reusing the treated WWTP effluent
as cooling-tower makeup water, or
as boiler makeup water.
Steam condensate recovery
Steam condensate recovered from
chemical processes, after some purification,
can serve as makeup water to
utility boilers. This water reuse technique
generates cost savings in water
resources and desalting operations, as
well as energy recovery from steam
condenstate.
In July 2010, one of Sinopec's refining
and chemical subsidiaries started
up its 1 million ton/yr ethylene plant.
One project team member was in
charge of the design of utility and
wastewater treatment systems of this
plant. Throughout her design, the project
team member made a serious effort
to incorporate all proven technologies
that the project team had implemented
throughout Sinopec's refining and
chemical subsidiaries. This resulted in
a final design that reuses 98.5% of the
plant's water resources. Additionally,
this plant has developed an efficient
process to recover, purify and reuse
1,120 ton/h of steam condensate, which
represents the largest condensate recovery
system ever built in China.
Recovered condensate should meet
the following key quality requireCorrosion
rate, mm/yr
Adhesion rate, mg/cm2-month

http://WWW.CHE.COM

Chemical Engineering May 2012

Table of Contents for the Digital Edition of Chemical Engineering May 2012