Page 439 -
P. 439

13.36                    CHAPTER THIRTEEN


         coveries exceeding 98%  can be attained  with backwash  recycling.  The maximum recov-
         ery can be identified by pilot testing.
         Solute and Particle Rejection.   In MF and UF systems,  solute or particle rejection is pri-
         marily  a  function  of membrane  type  and  pore  size rating.  However,  previously rejected
         substances  remaining  on  the  membrane  surface  also  can  act  as  a  "dynamic  membrane"
         and  affect  solute  removal,  especially  for  UF  systems.  Typical  MF  membrane  pore  size
         ratings range from 0.1 to 0.5/xm. Commonly used UF membranes range from about 2,000-
         to  150,000-dalton  molecular-weight-cutoff (MWCO).
         Flux.  The  permeate  or  filtrate  flux  through  UF  and  MF  membranes  depends  signifi-
         cantly on transmembrane  pressure  and water temperature.  Design flux rate used for a sys-
         tem is also strongly defined by membrane  feedwater quality because  the rate of plugging
         and  frequency  of backwashing  and  cleaning  are  affected by  the  operating  flux rate.  De-
         sign flux rate for a specific application is often determined by bench  or pilot testing. Typ-
         ical flux rates  for UF and  MF membranes  vary widely, depending  on membrane  product
         and  specific application,  and range  from  about  20 to  100 gpd/ft 2 (0.034  to 0.170  m/h).
         Temperature.   Temperature  affects the required driving pressure for a UF or MF system
         because  of changes  in  feedwater  viscosity. Figure  13.22  shows  the  viscosity of water  as
         a function of temperature.  At 20 ° C, the absolute  viscosity is approximately  1.00 cP. Be-
         cause  the  filtrate/permeate flux through  clean membrane  is inversely proportional to vis-
         cosity,  about  50%  more  TMP  is  needed  to  maintain  a  constant  flux  at  5 °  C than  at  20 °
         C. Assuming  a constant TMP and membrane  area,  the product  flow at 5 ° C would be ap-
         proximately  35%  less than  at 20 ° C.
           For UF and  MF  systems,  an  approximation  of permeate  flow at  any  temperature  rel-
         ative to  flow at  20 ° C is  as follows:
                                         Qp(20 ° C)
                                 QpT-  e_0.0239(T_20 )

         where
                 apT  =  permeate  flow at temperature  T
              Qp(20 ° c)  =  permeate  flow at  20 ° C
                   T =  water temperature,  °C
                   e  =  2.71828
           In the  design  process,  it is  critical  to  define  the  desired  production  rate  at  a  specific
         water temperature.  In many  cases,  less  production  is needed  in the  colder winter months
         because  of decreased  water  demands  from  the  utility  customers.  The  MF  or UF  system
         could  then  be  designed  to  produce  the  rated  capacity  at  a  warmer  temperature,  thus  re-
         ducing  the  required membrane  area and  saving capital  costs.

         .~   2.0
         ~'~  1.5
         ~..
              1.o
         N~  o.5
         <   0.0
                0     10     20     30     40     50   FIGURE 13.22  Absolute vis-
                                                      cosity of water versus tempera-
                           Temperature  (°C)
                                                      ture.
   434   435   436   437   438   439   440   441   442   443   444