Page 251 - Membranes for Industrial Wastewater Recovery and Re-Use
P. 251

220  Membranes for  lndustrial  Wastewater Recovery and Re-use


            The full details of the design may be viewed at any time from the File menu
          (Preview All) or by the short-cut key command Ctrl V.
            For the problem specified it must be assumed that the hardness is all as calcium
          and the alkalinity as bicarbonate, so that these can be directly entered into the
          Feed Analysis screen. Composition values must be entered in the appropriate
          units (mg/l or mg/l as CaC03 in this case). An estimate of the sodium or chloride
          concentration is needed to complete the analysis and achieve the target TDS of
          1200  mg/l.  The  package  automatically  adjusts  the  specification  with  the
          counter-ion (i.e. chloride or sodium ions) to obtain electroneutrality.
            On the following pH Adjustment screen the required pH to which the water
          feeding into the array must be entered. It is simpler to do this once the scaling
          propensity  (via the Langelier Scaling Index, LSI) has been  calculated. On the
          following Flow Rates and Recovery menu the product flow rate (3024 m3/day
          and the recovery (75%) must be entered.
            Next  screen  is  the  Membrane Array  Configuration, which includes  an
          Array Wizard that allows an estimate of the flux to be entered. RO membranes
          will  normally  operate at a  flux  of  between  20 and  50 LMH,  depending  on
          the degree of  fouling. Membrane elements can be selected  from the left-hand
          drop-down menu and modules from the right hand one. For this large flow rate,
          a  large  membrane  module  is  appropriate  (prefix  “8”  for  8-inch  diameter
          modules  and MAG  for the largest membrane element area). Having entered  a
          flux  value,  the  preliminary  design  is  complete  and  the  GO  button  can  be
          pressed.
            If  a fairly conservative flux of  2 5 LMH is chosen for the array design, the only
          error  arising  will  be  from  the  scaling  propensity  of  the  water.  This  can  be
          corrected by acid addition to being the LSI below zero. Again, trial and error is
          required for this, and through iteration it can be established that adjusting the
          pH to around 6.1-6.3  (depending on the membrane selected) by pre-dosing with
          either sulphuric  or hydrochloric acid is sufficient.
            The product water quality must then be checked to establish if  98% rejection
          has been achieved (i.e. the permeate product TDS is no more than 24 mg/l. 2% of
          the feed TDS of  1200 mg/l) and that the pH is around 7. The permeate product
          pH is always acidic due to the unrestricted permeation of carbon dioxide, which
           can be removed  by  stripping and/or neutralised  by  lime or caustic dosing. In
          reality, the controlling the product  pH  to  7 following  C02 stripping would  be
           extremely  difficult for such a low TDS  water. The permeate  TDS  can also, of
           course, be altered by  membrane selection. Selection, once again, proceeds by
           trial and error  - and more  than  one option  is  possible.  A  possible  design  is
           detailed below.

             Pretreatment: Dosing to pH 6.2 with 93.2% sulphuric acid at 376.3 kg/day
             Membrane: TFC 8832HR-575MAG
             Array: 10:5 2-stage array, 4 elements/module (60 elements total)
             Post-treatment: Degassing to  5 mg/l dissolved carbon dioxide (dosing to pH  7
           with sodium hydroxide)
             Spec. energy dem.: 215 kW for 126 m3/h flow, 1.71 kWh  m-3.
   246   247   248   249   250   251   252   253   254   255   256