Page 467 -
P. 467

ACTIVATED CARBON PROCESSES              14.15

          lOO


                                 2o×3o
           80                                         12 × 16


                                        16 ~< 20            10x12
         ~  so



           40



           20




                         10            20            30           4o
                                      BACKWASH RATE (gprn/f~
        F|GURE  14.5   Backwash characteristics  of various mesh sizes at  25 ° C.


           GAC particle size distribution and wetted density vary among different carbon brands
         and even among different deliveries of the same carbon.  Appropriate backwash rates can
        be obtained from the manufacturer for each type of carbon (Figure  14.5). Backwash rates
        must be  adjusted  to account for specific media characteristics  and  for changes  in back-
        wash water temperature (Figure 14.6). Installing a surface wash or air scour system to as-
         sist  with filter cleaning may be necessary to control mudball formation.
           A  good conservative design should allow for 75%  to  100%  expansion of light GAC
         media,  but  50%  is  generally considered to  be  adequate.  The  design  should  provide for
         sufficient freeboard to reduce media losses during the backwashing cycle.

         Carbon  Usage Rate.  The carbon usage rate (CUR) determines the rate at which carbon
         will be exhausted  and how often the carbon must be replaced.  The CUR essentially de-
         termines  the  size of the  entire regeneration  system.  The CUR for GAC  systems remov-
         ing  organic  compounds  may  be  determined  by  using  physical  models  or  adsorption
         isotherm models.  A  pilot-scale test is often used to evaluate the complexity of multiple
         chemical interactions.  A  quicker and  more economical method for evaluating GAC col-
         umn  performance is  the  rapid small-scale  column  test (RSSCT).  Small systems  can  be
         designed to best simulate the performance of a full-scale system using dimensional anal-
         ysis. Dimensionless parameters for the full-scale system and the small system are designed
         to be equal.
           Carbon treatment effectiveness improves as contact time increases. The percentage of
         total carbon in a bed that is exhausted at breakthrough  is greater in a deep bed than in a
         shallow bed.  At a point beyond the optimum bed depth,  the additional  adsorber volume
         provided acts primarily as a storage capacity for spent carbon. The actual selection of bed
   462   463   464   465   466   467   468   469   470   471   472