Page 139 - Pressure Swing Adsorption
P. 139

114                                   PRESSURE SWING ADSORPTION                EQUILIBRIUM THEORY                                           115

              when lhe adsorht~nt  is pressurized with feed, the adsorbent near the feed  end
              contacts  the  feed  at essentially  the  lowest  pressure  m  the  cycle.  This allows
              the  more  strongly  adsorbed  comoonent  to  penetrate  farther  into  the  bed,
              because  less  1s  adsorbed  by  that  adsorbent  than  if  the  gas  were  fully
              pressurized.  As  a  result,  that  adsorbent  is  less  than  fully  utilized.  Thus,
              desmte  the  fact  that  oressunzat1on  does  not  contribute  directly  towards
              production.  1t  can diminish  the  useful  capacity of the  adsorbent.

                4.4.3  Four-Step PSA Cycle:  Incomplete Purging
                                                                                                                  Purge      Pressur1zotlon   Feed
              The two cycles considered so far m this section are stmole, and relatively easy                      S1ep         Step          Step
              to  anaiyze.  Unfortunately,  however,  they  are  not  oarticularly  efficient  m   Figure 4.7  Paths of characteristics m the  purge, prcssunzat1on, and feed  steps. for  a
              terms  of  their  performance  relative  to  power  requirements.  That  1s,  they   fractional extent of purge of X. Note:  z' is  measured from  the bottom toward  the top.
              generally  require  high  pressure  ratios  to  attam  high  recovenes.  For  this   and  z  1s  measured  in  the reverse direction.  7
              section,  we  consider  a  simple  modification  of  the  four-step  PSA  cycle
              described  in  Section 4.4.J. It will  be seen that this modification, which simply
              involves varymg the extent of purgmg, can ieact  to remarkably higher recovery   and  parameter  vaiues.  Generally,  reducmg  the  amount  of  purge  always
              at rclat1vely  low  pressure  ratios.                                           results  tn  mcreased  recovery,  but beyond  some  limit (described  below)  there
                 Incomplete purging has been common  m  industrial practice, as mentioned     1s  no assurance that pure product can  be obtained.
               by  Wagner  14   and Wankat.  15   Quantitative studies of the extent of purge have   It  may  help  to  visualize  the  action  of  the  steps  in  terms  of wave  move~
               focused  mainly  on  flow  rate  ratios,  especially  the  purge-to-feed  ratio.  For   ments,  as  m  the discussion  of Figure  4.1.  A  schematic diagram  of the  wave
               example,  the effects  of the  purge-to-feed  ratio  on  light-product  purity were   movements m the relevant steps 1s  given m  Figure 4.7,  aithough m  this figure
                                                                   1
               studied by  Yang and Doong,  16   Doong and Yang,  17   and YangY Their results   the  shading  that  represents  the  mflux  of  the  heavy  component  has  been
               implied  that  it  was  not  feasible  to  increase  recovery  by  decreasing  the   omitted.  The  ourge  step  ts  still  characterized  by  a  simple  wave.  This  wave
               purge-to-feed  ratio  and  still  mamtam  high  oroctuct  purity.  Kirkby  and   soreacts  as  1t  propagates  over  the  iength  of  the  coiumn,  as  shown  on  the
                     19
               Kenney  showed theoretically  and  exoenmentally that there is  an  ootimum    left-hand side of the diagram.
               extent  of  purgmg,  for  which  product  ounty  and  recove1y  are  maximized.   The key  feature  of the  extent of ourgmg 1s  the fraction  of the  column  X
               Their  cell  model  suggested  that  the  optimum  corresponded  to  complete   that 1s  completely purged. That 1s,  X  is  the fraction  of L  over which  y  =  0 at
               purge, but  their expenments revealed  tlrnt  a  lesser amount was  oot1mal.  An   the  end  of  the  purge  step,  t L·  Because  they  are  linearly  reiated,  this  1s
                                                                   7
               equilibrium-based  model was develooed  by  Matz and Knaebel to assess  the    identical  to  the  fraction -of  the  amount  of  gas  requJred  to  purge  the  bed
               effect  of  ourge  on  PSA  performance,  based  on  systems  having  linear   completeiyt via  Eos. 4.21  and 4.23, as
               isotherms.  ahct  that  work  is  the  basis  of the  following  discussion.  Recently,   X  =  f3Avintlpu  =  Qtlru
                            9
               Rousar und  Ditl solved  the same basic equilibrium-based eauat,ons analyt1-  I                                                          ( 4.31)
               cally to determine the optmmm ourge amount, and they examined ooerat1on   i                L       Qtlru
               in a  regime that yields impure light product.                                 where ,BAvintlru  is  the distance mto the bed that 1s  fully  purged and Qtliu  1s
                 By  Its  nature. mcomoiete purging ieaves a  comoositton tail, or heei.  at the   J   the number of moles  required  for complete  purge.
               feed  end  of  the  column,  containmg  some  of  the  more  strongly  adsorbed   An arbitrary characteristic  m  the  s1mpk:  wave  is  denoted  by  its compos1-
               compom::nt.  Subsequent  pressurization  with  product  (also  countcrcurrent  to   tlon,  y •  The  one  that  Just  reaches  the  effluent  end  of  the  coiumn  as  the
                                                                                                    0
               the  feed)  oushes that  residual  material  toward  the  feed  end.  More subtly,  1t   ourge steo ends is socciai.  Its mole  fraction  1s  called  Yoi:-i.·  An  operat1onal
               also  reduces  the  gas-phase  mole  fraction  of  the  more  strongly  adsorbed   constraint is that this should be less  than the "expanded" feed  mole fraction,
               component  m  that  region, since  the  heavy component is  preferentially taken   or,  if  it  is  not,  when  1t  1s  reoressunzed  1t  will  s1mpiy  revert  to  the  feed
               up  as  pressurization  proceeds.  The  presence of the  compressed  t.1il  reduces   composition.  In other words, if this constraint is not  met, rcgencratmn will  he
               the  quantities  of  both  the  feed  admitted  antl  lhc  gross  product  obrnmed   mcomplete, and the effective length of the column will  he  reduced, lctiding to
               durmg  the  feed  step.  The  relative  amounts  consumed  and/or  produced    premature  breakthrough.  Malhcmat,cally,  this  amounts  to  the  following
               during each step depend on  the extent of ourge, as well  as on  the conditions   meauality  Ysl,-L  < Ye·  Generally,  the  value  of  Ys  can  be  determmed  from
   134   135   136   137   138   139   140   141   142   143   144