United States
            Environmental Protection
            Agency
            Municipal Environmental Research
            Laboratory
            Cincinnati OH 45268
EPA-600/2-78-169
September 1978
            Research and Development
£EPA
Demineralization
of Carbon-Treated
Secondary Effluent
by Spiral-Wound
Reverse Osmosis
Process

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health  Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has  been assigned  to the  ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                           EPA-600/2-78-169
                                           September 1978
DEMORALIZATION OF CARBON-TREATED SECONDARY EFFLUENT

       BY SPIRAL-WOUND REVERSE OSMOSIS PROCESS
                         by

                   Ching-lin Chen
                   Robert P. Miele

  County Sanitation Districts of Los Angeles County
             Whittier, California  90607
               Contract No. 14-12-150
                   Project Officer

                  Irwin J. Kugelman
            Wastewater Research Division
     Municipal  Environmental  Research Laboratory
               Cincinnati, Ohio  45268
                            j 1 01           •'•••-•"•.•.Ion igo-ney

                            . x           .  .:,.,o^ 167Q
     MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
         OFFICE OF RESEARCH AND DEVELOPMENT
        U.S.  ENVIRONMENTAL PROTECTION AGENCY
               CINCINNATI, OHIO  45268

-------
                                 DISCLAIMER
     This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.

-------
                                  FOREWORD


     The Environmental Protection Agency was created because of increasing
 public and government concern about the dangers of pollution to the health
 and welfare of the American people.  Noxious air, foul water, and spoiled
 land are tragic testimony to the deterioration of our natural environment.
 The complexity of that environment and the interplay between its components
 require a concentrated and integrated attack on the problem.

     Research and development is that necessary first step in problem solution
 and it involves defining the problem, measuring its impact, and searching for
 solutions.  The Municipal Environmental Research Laboratory develops new and
 improved technology and systems for the hazardous water pollutant discharges
 from municipal and community sources, for the preservation and treatment of
 public drinking water supplies, and to minimize the adverse economic, social,
 health, and aesthetic effects of pollution.  This publication is one of the
 products of that research; a most vital communications link between the
 researcher and the user community.

     One of the goals of wastewater treatment is renovation of wastewater so
 that it can be reused.  It is expected that partial  demineralization of
 conventionally treated wastewater will  be required if the wastewater is reused
 for any purpose which requires high quality water.   Among the techniques for
 demoralization that which is newest but shows the most potential  is reverse
 osmosis.   In this process water is forced through a membrane which can reject
 salts.   The permeability of these membranes is low so high pressure is required
 to achieve an economical  production rate.   Special  configuration of the mem-
 brane and its support system are required to withstand the high pressure and
maintain  a high ratio of membrane surface to system volume.   In the studies
 reported  in here a reverse osmosis system using a spiral  membrane-support
configuration was tested  for its efficacy in demineralizatton of secondary
effluent.   Included  in the study was .an evaluation  of pretreatment  of the
reverse osmosis feed  with activated  carbon to reduce  membrane fouling
                                       Francis  T.  Mayo,  Director
                                       Municipal Environmental  Research
                                        Laboratory
                                      m

-------
                                 ABSTRACT

     A 56.8 cu m/day (15,000 gallons/day) spiral-wound reverse osmosis pilot
plant, manufactured by the Gulf Environmental Systems Company, San Diego,
California, was operated at the Pomona Advanced Wastewater Treatment
Research Facility on the carbon-treated secondary effluent.   The specific
objectives for this study were (a) to establish the effective membrane life
for wastewater demineralization with carbon adsorption pretreatment; (b) to
determine the reliability of the process performance; and (c) to derive a
realistic process cost estimate.

     The study was first conducted on a constant feed pressure basis, and
then it was run on a constant product water flux rate basis.   During the
first phase of the study, pH adjustment was not practiced for the weekly
enzyme-detergent membrane cleaning procedures.  However, this was practiced
in the second phase of the study.  The results from both phases of studies
substantiated the fact that the membrane effective life was only about one
year in demineralizing the carbon-treated secondary effluent.

     A cost estimate for a 37,850 cu m/day (10 MGD) reverse osmosis plant
indicated that for membranes with only one-year life the process cost was
about 14.9(^/1,000 liters (57.4^/1,000 gallons).  However, the cost could
be substantially reduced to 10.7^/1,000 liters (41.3^/1,000 gallons) for
membranes with two-year life.  Both cost estimates did not include the
costs for carbon adsorption pretreatment and brine disposal.   These cost
estimates were based on August, 1973 material and construction costs.

     This report was submitted by County Sanitation Districts of
Los Angeles County in fulfillment of Contract No. 14--12-150 under the
partial sponsorship of the Municipal Environmental Research Laboratory,
Office of Research and Development, U.S. Environmental Protection Agency.
Work was completed as of January 13, 1972.
                                     TV

-------
                                 CONTENTS


Foreword . . ......  .-..  .  . , -.-.-.  «  .  ,  ,  .  .  ,  ,  .  ,  ,  .  ... -,•-..,' .  ill
Abstract	• •  • •  iv
Figures  . . . . .  .  ....  .  . .  .  .  .  ....  ,  .... .  .  .  .  . -.".  .   vi
Tables	vii
Acknowledgement	•	•  «	V111


   1.   Introduction  	«	    1
   2.   Conclusions   .  ,  	  ,,,.,,,.,,  	    3
   3.   Reconmendattons   .  ,  .  ,.,....,.....,.••••«    5
   4.   Pilot Plant Description .  .  .  .  .  .  .  .  .  .  >  ,	    6
   5.   Pilot Plant Operation	,	   11
             Operating  conditions   ..,,,..«..,,..,•««   H
             Membrane cleaning  .,..,,...,,  ,  ,.,»...,.   16


   6.   Results and Discussions ....  ...  <•••••  .  «  >  .,,.,,,   19
             Constant feed  pressure operation	,  . , ,  ,   19
             Constant product flux  rate operation  ,  ,  .  ,  .  .  , , ,  ,   27
             Membrane module  stability  .,,.,.,,,.«,.,,.   45


   7.   Process Cost  Estimate  . .  .  .  .  .  ...  •  •  ,,  •  .. •  .  ^ • •  '   56


References	•	   59

-------
                                 FIGURES
Number                                                                page
   1   Schematic flow diagram of the reverse osmosis pilot plant  ,  ,    7
   2   Schematic diagram of the activated carbon pretreatment
          system ,,...,.,,.,,,.,	   10
   3   Variation of flux rate and salt rejection during the
          initial 800 hours of operation with constant feed
          pressure of 32.1  Kg/sq cm (465 psi)	 , .  ,  .   20
   4   Total feed COD and product water flux rate vs. operation
          time (constant feed pressure operation)  , , .  , , , ,  .  .   22
   5   Decline rate of product water flux under constant feed
          pressure operation . . .	  .,...,..,,   25
   6   Variation of rejection vs. operation time under constant
          feed pressure operation,  ,,,,,.,.,.,,,. ,..:.'.«   26
   7   Salt rejection vs. operation time in pressure vessel No, 1   ,   28
   8   Salt rejection vs. operation time tn pressure vessel No. 2   .   29
   9   Salt rejection vs. operation time in pressure vessel No. 3   .   30
  10   Salt rejection vs. operation time in pressure vessel No. 4   .   31
  11   Salt rejection vs. operation time in pressure vessel No. 5   .   32
  12   Salt rejection vs. operation time in pressure vessel No, 6   .   33
  13   Salt rejection vs. operation time in pressure vessel No, 7   .   34
  14   Salt rejection vs. operation time in pressure vessel No, 8   .   35
  15   Salt rejection vs. operation time in pressure vessel No, 9   .   36
  16   Salt rejection and feed pressure variation vs. operation
          time under constant flux rate operation .  ,	    40
                                   VI

-------
                                    TABLES
Number                                                                   Page
   1   Physical Characteristics of the Granular Activated
         Carbon fn the Pretreatment System. ....... .  ,  > •  » • •    8
   2  Module Loading Arrangement at Start-Up of Constant Product
         Flux Rate Operation  .,...,,	,   12
   3  History of Membrane Modules Used for Constant Product
         Flux Rate Operation	, .  ,  .  , •, ,  , .  .  , ,  . , .13
   4  Operating Conditions at Start-Up and TOO Hours Later of
         Constant Product Flux Rate Operation  ,.  ,,,,,,,,,«,   17
   5  Individual Module Salt Rejection Tests Conducted at the End
         of Constant Feed Pressure Operation Study, , ,  ,	   37
   6  Summary of Water Quality Analyses for the Period of Zero to
         9,475 Hours of Constant Feed Pressure Operation, ,,,,,..   38
   7  Summary of Water Quality Analyses for the Period of Zero to
         6,700 Hours of Constant Product Flux  Rate Operation,  , ,  , , .   46
   8  Summary of Water Quality Analyses for the Period of 6,700 to
         7,803 Hours of Constant Product Flux  Rate Operation.  . ,  , , ,   47
   9  Performance of Modules in Each Pressure  Vessel from Time Zero
         to 4,800 Hours of Constant Feed Pressure  Operation  ...,,.   48
  10  Results of Module Tests Conducted at the End of Constant
         Feed Pressure Operation Study	 .   52
  11   Results of Dye Checking, Visual Inspection and Membrane
         Sample Testing	   53
  12   Process Cost Estimate for 37,850 cu m/day (10 MGD)
         Spiral-Wound Reverse Osmosis Plant 	  ,  	   58
                                      vi i

-------
                                ACKNOWLEDGEMENTS

     This study was jointly sponsored by the U.S. Environmental Protection
Agency and the County Sanitation Districts of Los Angeles County.

     The authors are deeply grateful to Dr. James E. Cruver of Gulf
Environmental Systems Company, San Diego, California, for his advice and
cooperation in this effort.

     Mr. James Gratteau and Mr. Harold H. Takenaka, former project engineers
at Pomona Advanced Wastewater Treatment Research Facility, were instrumental
in initiating the pilot plant study.

     The efforts of the laboratory and the pilot plant operating personnel
of the Pomona Advanced Wastewater Treatment Research Facility are also
gratefully acknowledged.
                                    vi,n

-------
                                SECTION 1

                               INTRODUCTION

     All uses of water serve to increase the mineral  and organic contents
of the water.  The organic impurities are normally removed by the bio-
logical oxidation and activated-carbon adsorption processes, while the
inorganic minerals are effectively removed by the demineralization pro-
cesses, such as ion exchange, electrodialysis, and reverse osmosis.  There-
fore, the wastewater demineralization is an indispensable part of the total
effort to achieve the following environmental goals:

     A.  To conserve the natural water qualities of the receiving
         water systems; and

     B.  To treat the wastewater to meet the quality requirements
         for various water reuses.

     County Sanitation Districts of Los Angeles County in conjunction with
the U.S. Environmental Protection Agency initiated a series of wastewater
demineralization studies in 1967.  Three demineralization processes--
reverse osmosis, electrodialysis, and ion exchange were extensively studied
at Pomona Advanced Wastewater Treatment Research Facility.  Since reverse
osmosis process was still at the development stage, several operating
parameters had to be established in the beginning of the pilot plant study.
The process was first applied directly to the secondary effluent without
proper membrane cleaning procedures.  This direct application was quickly
proved to be a failure by the rapid decline in the system performance.

     On June 16, 1969, a new experimental  run with a 56.8 cu m/day (15,000
gallons/day) spiral-wound reverse osmosis pilot plant, manufactured by the
Gulf Environmental Systems Company, was initiated with a carbon adsorption
pretreatment on the secondary effluent.  The objectives of this study were:
(a) to evaluate the effect of the carbon adsorption pretreatment on the
system performance; (b) to obtain data on the system reliability; (c) to
establish the effective membrane life; and (d) to derive a realistic pro-
cess cost estimate.

     The study was divided into two phases.  The first study was conducted
with a constant operating pressure, while the second phase was conducted
with a constant product water flux rate.  After the initial 9,475 hours
of on-stream operations in the first phase of the study, the pilot plant
operation was temporarily suspended on August 16, 1970, as a result of the
serious membrane deterioration.  This was revealed by the substantial

-------
reduction in both product water flux rate and salt rejection.  All the mem-
brane modules were subsequently removed from the pilot plant system and
sent to the Gulf Environmental Systems Company for membrane evaluation to
determine the causes of membrane deterioration.

     Based on the membrane evaluation results, the pilot plant operation
was resumed on December 21, 1970 for the second phase of the study.  New
sets of operating conditions and membrane loading arrangement were employed
in this second study.  Only three of the original twenty-seven membrane
modules were kept in the system for this new study, while fifteen of the
other twenty-four modules were replaced with the new production membrane
modules.  The remaining nine modules were replaced with the partially used
modules from a similar system being concurrently operated at Pomona Research
Facility.   All  the used modules were still in good performance condition.
This second part of the study was finally terminated on January 13, 1972,
after a total  of 7,803 hours of on-stream operation.

-------
                                 SECTION 2

                                CONCLUSIONS

     The principal conclusions drawn from this pilot plant study are out-
lined as follows:

     A.  The Gulf Environmental System Company's spiral-wound reverse os-
mosis system was capable of achieving a 95 percent salt rejection, a 566
1/sq m/day (13.9 gal/sq ft/day) product water flux rate, and an 80 percent
water recovery under a constant operating feed pressure of 32.1 Kg/sq cm
(465 psi) in its initial stage of operation.

     B.  A regular membrane cleaning operation, including a weekly enzyme-
detergent (BIZ) or sodium perborate cleaning and a daily air-tap water
flushing, was essential even with a carbon adsorption pretreatment in con-
trolling the product water flux decline, which resulted from membrane
fouling.

     C.  A minimum brine flow at approximately 11.3 1/min (3 gpm) was help-
ful in minimizing the product water flux decline.

     D.  Both modes of operations, constant operating feed pressure, as in
the first phase of the study, and constant product water flux rate, as in
the second phase of study, showed similar performance and product water
quality.

     E.  The water quality data prior to the deterioration of the membrane
modules indicated that on the average the product water had:

     a.  Less than 3 percent of the feed phosphate content;

     b.  Less than 7 percent of the feed total chemical oxygen
         demand (TCOD) content;

     c.  Less than 1 percent of the feed sulfate content;

     d.  Less than 3 percent of the feed calcium content;

     e.  Less than 11 percent of the feed ammonia nitrogen content;

     f.  Less than 8 percent of the feed total dissolved solids
         (TDS) content; and

-------
     g,  Less than 5 percent of the feed turbidity.

     F.  The cause of membrane deterioration was partially attributed to
the hydrolysis of the membrane which was caused by the exposure to the
high pH of the enzyme-detergent cleaning solution during the first phase
of the study.

     G.  The results from both modes of pilot plant operations indicated
that the effective membrane life was only one operation year based on
initial performance parameters.

     H.  The process cost estimate for a 37,850 cu m/day (10 MGD) reverse
osmosis plant is about 14.9<£/1,000 liters (57.4^/1,000 gallons).   However,
if the membrane life could be improved from one year to two years, then
the cost would be reduced to 10.7^/1,000 liters (41.3^/1,000 gallons).
Both cost estimates do not include the costs for carbon adsorption pre-
treatment and brine disposal.

-------
                                 SECTION 3

                              RECOMMENDATIONS
     The short membrane life as concluded from the study on wastewater de-
mineralization is rather discouraging.  An optimum membrane life was shown
to be about three years for a practical  and economical application of the
reverse osmosis process to the wastewater demineralization(l).   There-
fore, it is recommended that further studies be pursued primarily in the
areas of membrane improvement.  Other parameters such as pretreatment
methods, membrane cleaning techniques and frequency, feed pressure, brine
recirculation, membrane module configuration, and brine velocity should
also be thoroughly evaluated and investigated.

-------
                                  SECTION 4

                           PILOT PLANT DESCRIPTION


     The 56.8 cu m/day (15,000 gallons/day) reverse osmosis pilot plant
consisted of 9 steel pressure vessels.  Each vessel measured 3.05 m (10 ft)
in length arid 10 cm (4 in) in diameter.  Three ROGA spiral-wound membrane
modules, manufactured by the Gulf Environmental  Systems Company, were in-
stalled in each of the steel pressure vessels.  Each membrane module;was
approximately 10 cm (4 in) in diameter, 0.91 m (3 ft) long, and contained
4.6 sq m (50 sq ft) of modified cellulose acetate membrane.  The total
membrane area in the pilot plant system was aoout 125 sq m (1,350 sq ft).

     Figure 1 shows a schematic flow diagram of the spiral-wound reverse
osmosis pilot plant.  The carbon-treated secondary effluent was chlori-
nated to a 1 to 2 mg/1 chlorine residual and acidified to a pH close to
5 using sulfuric acid before it was fed to the membrane system.  The pilot
plant system was in a 3-2-2-1-1 array to maintain sufficient brine velo-
cities in the downstream modules.  Some necessary provisions for a daily
air-tap water flushing, a weekly enzyme-detergent cleaning cycle, and a
chlorinated tap water flushing during downtimes were made,  A flexible
metal hose was installed between the main feed pump and the lead modules
to prevent the fatigue failure of the piping in the system, which other-
wise would be caused by the serious vibration of the feed pump.

     Sufficient sample valves were installed on the pilot plant system,
so that samples from the raw feed (carbon treated secondary effluent)
blended feed (mixture of carbon-treated secondary effluent, sulfuric acid
and chlorine solution), brine, and product streams could be taken regu-
larly.  Instrumentation was included to measure the temperature and the
pressure of the blended feed, brine and product streams.  A proportional
chemical feed pump was used to add sulfuric acid to the feed stream for
pH control.  The pump rate was regulated by a pH controller.  Chlorine
was added to feed stream through a gas chlorinator.

     The carbon-treated secondary effluent was obtained from the concur-
rent activated carbon adsorption pilot plant study at Pomona Research
Facility.  The carbon pilot plant was a four-stage downflow pressure sys-
tem.  Each stage contained about 3,020 Kg (6,650 Ib) of Calgon
Filtrasorb-400 granular activated carbon in a 1.83 m (6 ft) diameter steel
column.  The depth of the carbon bed was about 3.04 m (10 ft).  Table 1
shows some of the physical characteristics of the activated carbon used

-------


^-TYPICAL



1
1 h"

2

3


-H

S
i

/!FEED\ /MIXING \ _. , ' , CARBON-TREATE
\PUMPy \ TANK / ,, i SECONDARY EFR
unnin r 	 ^ \ ' 1 BR|NE RECIRCULATIO
'- pi. IN IFCTIOM
	 1 FLOW
, 	 , i 	 1 CQNTROLLER-7
-»• 4 • *" 6 I
* 8 ~1 9 ^
-*. 5 jj ' "" 7 ^

:D
.UENT
N
^ ,
BRINE
PRODUCT

COMPRESSED CHLORINE

                                     AIR
Figure I. Schematic  flow diagram of the reverse osmosis pilot plant.

-------
                            TABLE 1
           PHYSICAL CHARACTERISTICS  OF THE GRANULAR
         ACTIVATED CARBON IN THE PRETREATMENT SYSTEM
Surface Area, m2/g (BET)                :            1000
Apparent Density, g/ml                   :               0.44
Density, backwashed & drained,          :
        Ib/cu ft                        r              25
        Kg/cu m                         :             401
Real Density, g/ml                      ;               2.1
Particle Density, g/ml                   :               1,3
Effective Size, mm                      :               0.55
Uniformity Coefficient                  :               1.9
Pore Volume                             :               Or94
Mean Particle Diameter, mm              :               0.9
Iodine No.                              :          •  1000
Abrasion No. minimum                    :              75
Ash, %                                  :      !         8-5

-------
in the study.  The empty-bed detention time for each stage of treatment was
about 10 minutes.  Therefore, a total of 40 minutes contact time was used
in the study.

     As shown in Figure 2, the carbon pilot plant included the carbon re-
generation system.  The carbon from the lead column was normally regen-
erated whenever the total chemical oxygen demand (TCOD) of the carbon plant
effluent reached a level of approximately 10 mg/1.   The lead carbon column
was backwashed daily with a maximum backwash rate of 6.8 Ips/sq m (10 gpm/sq
ft).  The results of the operation and performance of the four-stage carbon
adsorption pilot plant were presented elsewhere(2).

-------
                                                           TO
                                                       ATMOSPHERE
SECONDARY
 EFFLUENT
           TO
          PRIMARY
        CLARIFIER

   BACKWASHf
     TANK   I
            CARBON COLUMNS




n
LL



«••






m
L
V


J_










T
-

^_
r
1
t


r
SPENT
CARBON I

RE6EN
A ABB Aft
                 CARBON
                DEWATERING
                  TANK
                  FUEL
                                                      AIR
                                                               AFTERBURNER
                                                                         BLOWER
                                           T
                                                  'WATER
               PRODUCT
                 TANK
1
                                  CARBON OUT"f
QUENCH
 TANK
                                                  MULTIPLE
                                                   HEARTH
                                                   FURNACE
                                                     CYCLONE
                                   Y
                                  DUST TO
                                  WASTE
                         TO
                       REVERSE
                       OSMOSIS
                        PILOT
                        PLANT
        TO
      CARBON
      COLUMN
MAKE-UP
WATER
                 MOTIVE WATER
                                                      COMPRESSED AIR
                                                      • FOR PULSED AIR
                                                      CLEANING OF BAGS
                                                                             BAGHOUSE
                                                                     DUST TO
                                                                     WASTE
                EDUCTOR
     Figure 2. Schematic diagram of the activated carbon pretreatment system.

-------
                                  SECTION 5

                            PILOT PLANT OPERATION

OPERATING CONDITIONS

Phase I: Constant Feed Pressure Operation

     The pilot plant was operated under a constant feed pressure of 32.1
Kg/sq cm (465 psi) during the first phase study.   The other initial operat-
ing conditions for the pilot plant are summarized as follows:
     A. Feed Water:


     B. Feed pH:

     C. Feed Flow:
Carbon-treated secondary effluent chlorinated
to a chlorine residual  of 1  to 2 mg/1.

Controlled to 5 using sulfuric acid.

61.3 1pm (16.2 gpm).
     0. Product Flow:  50 1pm (13.2 gpm).

     E. Brine Flow:    11.3 1pm (3 gpm).
     F. Product Water
        Flux Rate:

     G. Water
        Recovery:

     H. Salt
        Rejection:
566 1/sq m/day (13.9 gal/sq ft/day) at 25°C.


81.5 percent.


95 percent.
     A daily air-tap water flushing and a weekly *enzyme-detergent cleaning
cycle were conducted to maintain the product water flux rate during the first
phase study.

Phase II; Constant Product Flux Rate Operation

     During the second phase of the pilot plant study, the twenty-seven mem-
brane modules in the system were made up of 12 used and 15 new production
modules.  The module loading arrangement for the system is shown in Table 2.
A summary of the operating history of the modules  is presented in Table 3.

     The system for the second phase of the study  was also operated on the
carbon-treated secondary effluent with pre-chlorination to provide 1 to 2
                                    11

-------
                           TAB.LE 2
          MODULE LOADING ARRANGEMENT AT START-UP OF
             CONSTANT  PRODUCT  FLUX  RATE  OPERATION
Pressure Vessel                     Module Loading Arrangement

       1                  Used modules from other system
       2                  New production modules
       3                  Used modules from first part of study
       4                  Used modules from other system
       5                  New production modules
       6                  New production modules
       7                  New production modules
       8                  Used modules from other system
       9                  New production modules
Note:   The sequence of the pressure vessels is shown in
       Figure 1 .
                              12

-------
                           TABLE  3
             HISTORY OF MEMBRANE MODULES USED FOR
             CONSTANT  PRODUCT  FLUX RATE OPERATION
Pressure Vessel
	  No. T

   12-21-70
    9-23-71
    1-13-72
Pressure Vessel
     No. 2

   12-21-70

    9-23-71
    1-13-72
Commenced study with 3 used modules which had
1,335 hours of operating time.

At 5,678 hours of unit operations, module #1,
with 7,013 hours total operating time, was
removed and replaced with a used module, which
had 1,933 hours of operating time.

Study terminated at 7,803 hours of unit
operation.
                  Module #1
                  Module #1R
                  Module #2
                  Module #3
                    7,013
                    4,058
                    9,138
                    9,138
hours of operation;
hours of operation;
hours of operation;
hours of operation.
Commenced study with 3 new production modules

At 5,678 hours of unit operations, module #1
was removed and replaced with a used module
which had 1,933 hours of operating time.

Study terminated at 7,803 hours of unit
operation.
                  Module #1
                  Module #1R
                  Module #2
                  Module #3
                  - 5,678 hours of operation;
                  - 4,058 hours of operation;
                  - 7,803 hours of operation;
                  - 7,803 hours of operation.
                           (conti nued)
                               13

-------
                      TABLE  3  (Continued)
Pressure Vessel
     No. 3

   12-21-70
    7-28-71
    9-23-71
    1-13-72
Pressure Veesel
     No. 4

   12-21-70
    1-13-72
Commenced study with 3 used modules  which had
9,475 hours of operating time.

At 4,414 hours of unit operations,  all  3
modules were removed and replaced with  3 used
modules which had 1,933 hours of operating
time.

At 5,678 hours of unit operation, module #1R
was removed and replaced with a used module
which had 1,933 hours of operating  time,

Study terminated at 7,803! hours of  unit
operations.
                  Module #1 ,
                  Module flR
                  Module #1RR
                  Module #2R and
           2, & 3 - 13,899 hours of operation;
                  -  3,197 hours of operation;
                  -  4,058 hours of operation;
               3R -  5,322 hours of operation.
Commenced study with 3 used modules whiclr had
1,335 hours of operating time,

Study terminated at 7,803 hours of unit
operations.
                  Module #1
                  Module #2
                  Module #3
                     9,138 hours of operation;
                     9,138 hours of operation;
                     9,138 hours of operation.
                          (Continued)
                               14

-------
                      TABLE 3 (Continued)
Pressure Vessel
     No. 5
   12-21-70     :  Commenced study with 3 new production modules.

    1-13-72     :  Study terminated at 7,803 hours of unit
                  operation.

                  Module #1         -  7,803 hours of operation;
                  Module #2         -  7,803 hours of operation;
                  Module #3         -  7,803 hours of operation.

Pressure Vessel
     No. 6      '    ••    -    .    ' .   • • •   ' -  .-    ,

     Same as Pressure Vessel No. 5

Pressure Vessel
     No. 7
     Same as Pressure Vessel No. 5

Pressure Vessel
     No. 8
     Same as Pressure Vessel No. 4

Pressure Vessel
     No. 9
     Same as Pressure Vessel No.
Note:   The sequence of the pressure vessels is shown in
       Figure 1.
                               15

-------
mg/1 of total residual  chlorine,   A summary of the operating conditions  at
the start-up and 100 hours later, when the water recovery was increased
from 75 percent to 80 percent by increasing the brine recirculation from
9.5 1pm (2.5 gpm) to 12.5 1pm (3.3 gpm), is shown in Table 4.  An attempt
was made during the second phase of the study to operate the system at a
constant 407 1/sq m/day (10 gal/sq ft/day) apparent product flux rate, and
80 percent water recovery by varying the feed pressure.

     The membrane cleaning procedures adopted for both phases of the study
were very similar; however, the pH of the cleaning solution was adjusted
from 10.0 to 7.5 with sulfuric acid during the second phase of study.  In
addition to the enzyme-detergent (Biz) solution, a 2 percent sodium perbo-
rate solution was also tested in this study.

     The procedures used in the air-tap water flushing and the enzyme-
detergent (or sodium perborate solution) cleaning cycle throughout the en-
tire study are described in the following sections.

MEMBRANE CLEANING

Enzyme-Detergent  (or Sodium Perborate) Cleaning Procedure

     The enzyme-detergent cleaning solution was made up by adding 2,84 Kg
(100 oz) of a commercial enzyme-detergent, BIZ, into 379 1 (TOO gal) of
tap water, while  the sodium perborate cleaning solution was made up of 2
percent sodium perborate and 0.15 percent Triton X-100 non-ionic detergent
with 1 percent  (based on detergent weight) carboxy methyl cellulose  (CMC)
soil suspending  agent.  The enzyme-detergent  (or sodium perborate) cleaning
was conducted once  a week.

     During  the  cleaning cycle,  the  system  (pressure vessels 1 to 9) was
first  filled with either enzyme-detergent or  sodium perborate  cleaning so-
lution using the  main feed pump.  The pressure vessels which were not being
flushed remained soaking  in the  cleaning  solution, while others were being
flushed according to the  following sequence.  Here the term  flush refers  to
cycling the  cleaning solution  through the pressure vessels and membrane
modules.

     A. Pressure vessels  1 to  3  were flushed  for  10 minutes  at a feed pres-
sure of 5.5  Kg/sq cm  (80  psi), and at a  flow  rate  of  about  22.7  to  30.3  1pm
 (6 to  8 gpm)  per pressure  vessel.

      B.  Pressure vessels  4 to  5  were flushed  for  20 minutes  at a feed pres-
 sure  of  5.5  Kg/sq cm  (80  psi),  and at a  flow  rate  of  about  22.7  to  30.3  1pm
 (6 to  8  gpm) per pressure vessel.

      C.  Pressure vessels  6 to  9  were flushed  for 20  minutes  at a feed pres-
 sure  of 5.5 Kg/sq cm (80  psi), and  at a flow  rate of  about  22.7  to  30.3  1pm
 (6 to  8 gpm) per pressure vessel.
                                    16

-------
   TABLE 4.  OPERATING CONDITIONS AT START-UP
    AND 100  HOURS LATER OF CONSTANT PRODUCT
              FLUX RATE OPERATION
Parameter
Start-Up
TOO Hours
Feed Pressure
Kg/sq cm
psi
Raw Feed Flow Rate
1pm
gpm
Product Flow Rate
1 pm
gpm
Waste Brine Flow Rate
1 pm
gpm
Brine Reci rcul ation Rate
1 pm
gpm
Water Recovery, %
Salt Rejection, %

24
360

47
12
35
9

11
3

9
2
75
95

.8

.3
.5
.6
.4

.7
.1

.5
.5
.5

24
360

44
11
35
9

8
2

12
3
80
95

.8

.3
.7
.6
.4

.7
.3

.5
.3
.5
                      17

-------
     During the cleaning cycle, the cleaning solution was recycled for the
specified time period in the first set of pressure vessels, and then the
same cleaning solution was applied to the next set of pressure vessels un-
til the sequence was completed.  After each cleaning solution flushing, an
air-tap water flushing was also conducted to rinse the membrane modules,

Air-Tap Water Flushing Procedure

     The air-tap water flushing was conducted once every day either as a
main cleaning process in non-chemical  solution flushing days or as a rinse
process in chemical solution flushing days.  For the air-tap water flushing,
the same sequence of application to the pressure vessels was used as for the
chemical solution flushing.  This consisted of flushing each pressure vessel
with tap water for two minutes and then with a mixture of air and tap water
for another three minutes.  The air-tap water mixture was, however, not re-
cycled, it went directly to waste.

Acid Flush Procedure

     This particular acid flushing was employed whenever the decline of the
product water flux was due to the loss of pH control  in the system.  The
procedure consisted of depressurizing the system and flushing with an acidi-
fied water (maintaining pH between 2 and 3) for thirty minutes.   The acid
flushing was then followed by a cleaning chemical solution and air-tap
water flushing to provide maximum cleaning of the membrane modules.

-------
                                 SECTION 6

                 ;         RESULTS AND DISCUSSIONS

CONSTANT FEED PRESSURE OPERATION

Product Water Flux Rate

     The variation of the product water flux rate during the initial 800
hours of on-stream operation under a constant feed pressure of 32.1 Kg/sq
cm'^.(465 psi) is shown in Figure 3.  As indicated in Figure 3, the product
water flux rate decreased rapidly from 566 1/sq m/day (13.9 gal/sq ft/day)
to  391 1/sq m/day (9.6 gal/sq ft/day) during the first 200 hours of on- ~
stream operation;  The primary cause for this rapid decrease in flux rate
was possibly due to the high membrane compaction during the initial hours
of  operation.  An indication that the decrease in flux rate for the initial
period of operation was due to membrane compaction and not organic fouling
was the fact that the enzyme-detergent flushing of the unit at 50 hours of
operation failed to restore the product water flux rate.  After 200 hours
of  operation, the weekly enzyme-detergent flushing procedure was found
successful in removing the fouling materials and in controlling the decline
of the product water flux rate.

     At 250 hours of operation, the system operation was temporarily sus-
pended due to the loss of carbon effluent feed which was caused by a power
failure in the carbon pretreatment system.  Chlorinated tap water with
approximately 1 mg/1 chlorine residual was run through the unit for about
65 hours until the carbon effluent feed was restored.  When the system was
placed back onstream, an increase in product water flux rate occurred.  The
increase was attributed to the flushing action resulting from 65 hours of
chlorinated tap water feed.

     As indicated in Figure 3, there were two sharp drops in product water
flux rate at 430 hours and 550 hours of operation.   These drops were caused
by the problems with the acid feed system.  The acid pump air-locked after
430 hours of operation and it resulted in a loss of feed pH control for
approximately 12 hours.   At 550 hours, an electrical failure in the pH moni-
toring system resulted in a partial  loss of pH control over a 3 day weekend.
As soon as each malfunction in the acid feed system was noted, the system
was taken offstream and corrective measures were taken to restore the prod-
uct water flux rate before it was placed back onstream.   In both cases* the
acid flush cleaning procedure as described in previous section was applied
successfully to the system to restore the flux rate.  The incidents fully
                                    19

-------
o
o
10
0>
        SALT REJECTION, %
o
o>
in

o
oo
in
*•  o
   o
   00
                                            o
                                            g
                                                      o
                                                      o
                                                       c
                                                               .S o
                                                               l-'f

                                                              •'*!!{•
                                                               '5T.10
                                                               o o>
                                                               CO ^
                                                               •o co
                                                               c 2
                                                               o «>
                                                               o> o.

                                                               O T>
 

                                          It.
                                20

-------
substantiated the fact that the pH control was very essential in prevent-*
ing the membrane fouling caused by the precipitation of calcium salts.

     Figure 4 summarizes the variations in the product water flux rate and
in the total COD of the feed water from July 1, 1969 to February 28, 1970,
This period corresponded to the operating times from 200 hours to 5,600
hours.  The solid circles on the product water flux curve of Figure 4 in-
dicate the applications of enzyme-detergent cleaning to the system.  The
solid triangles on the total feed COD curve indicate times when methanol
was present in the feed water.  This methanol leakage occurred on several
different occasions during a denitrification study being conducted con-
currently in the carbon adsorption system and resulted in abnormally high
COD values.  Some special notes are shown under each curve of Figure 4 to
explain the deviations from the normal operation.

     The effectiveness of the enzyme-detergent flushing in restoring the
product water flux rate is illustrated by the distance between the two
solid circles.  The two solid circles are, respectively, the product water
flux rates before and after the enzyme-detergent cleaning cycle.

     The decline of the product water flux rate during the entire first
phase of this pilot plant study is shown in Figure 5.  The product water
flux rate decreased from 566 1/sq m/day (13.9 gal/sq ft/day) at time zero
to 350 1/sq m/day (8.6 gal/sq ft/day) at 6,000 hours of operation.  How-
ever, the product water flux rate between the period of 6,000 hours to
9,475 hours (end of the first phase of study) was found to increase from
350 1/sq m/day (8.6 gal/sq ft/day) to 374 1/sq m/day (9.2 gal/sq ft/day).
This increase in flux rate corresponded with a decrease in the overall
salt rejection.

     The flux decline slope was determined several times during the study.
The initial slope, determined after 1,500 hours of operation, was - 0.09.
At 6,000 hours,  the flux decline slope changed to - 0.07.   After 6,000
hours, the product water flux rate began to increase due to the deteriora-
tion of the membrane.  This caused a reversal of the flux decline slope.
Finally, at 9,475 hours, the flux decline slope was about - 0.055.  The
most meaningful  flux decline slope would be that calculated for the first
6,000 hours of operation, that is - 0.07.

Salt Rejection

     The salt rejection variations from July 1, 1969 to April 30, 1970 are
shown in Figure  6.  This period corresponds to the operation times from
200 hours to 6,950 hours.  The salt rejection was found to decrease slight-
ly when the concentration of the nitrate ion in the feed water increased
due to either the nitrification of the Pomona activated sludge plant,
which supplied the secondary effluent to the carbon pretreatment system, or
the addition of  sodium nitrate to the feed of the carbon adsorption system
during the denitrification study.  The reason for this decrease in salt
rejection was that the nitrate ion was not rejected as well  as other ions
                                   21

-------
ro
ro
             UJ

            -15
             cc
             X
                12
                10
             CC-*.
             uj«:
O
I  6
SE
\  30
             Q
             020
                10
                                               I        I

                                                  • BIZ FLUSHING
                    Cli TAPWATER
                       FLUSH
                       PERIOD
                                   .pH CONTROL
                                      LOST
                     FEED PRESSURE = 32.1 kg/sq cm
                                      (465 psi)
                    4 METHANOL LEAKAGE -/'
                    NOTE : VALUES FOR COD
                    OF 3J OR GREATER WERE
                    PLOTTED AT 31 mg/l
                   ^
                          J_
                       I
I
I
I
                                                                        _L
                           10       20
                             JULY-69
                              30
10      20
  AUGUST
       30       10        20
               SEPTEMBER
                  30
            Figure 4. Total feed GOD and product  water flux rate vs. operation  time.
                       (constant feed pressure operation)

-------
TO
OJ
        8
        tr
        a.
        E
        I
        0
           30
20
           10
                                                                                 •PRESSURE METER
                                                                                   ADJUSTED
                                                                                  n
                     J.
               MAIN COLUMN
               CARBON
               REGENERATION

                    I

            30       10      20
                      OCTOBER - 69
                           30
J.
10       20
 NOVEMBER
J_
30       10       20
          DECEMBER
                 30
                                                                                          10
       Figure 4. Gontinued

-------
             12
ro
-ts.
         x
         o

         u.

         ac
         UJx.
           "
         p

         Q
         O
         CC
         QL
             10
              8
30
          E  20

          o
          O
          o
             10
                    'METHANOL
                    ADDITION •
                    TO MAIN
                    COLUMN
                    STOPPED
                                                _L
                                             _L
               10      20      30
                  JANUARY-70
10       20       30
      FEBRUARY
                                                     10
  20
MARCH
                                                                                   30
                    _L
10      20
   APRIL
         Figure 4. Continued

-------
ro
en
          10'
       XI

       Q.

       CD


       I


       LU
oc
UJ


I

I-
o

Q
O
IT
CL-
                                   I I  I
                          	Op

                       SLOPE =-0.055
                FEED PRESSURE :

                MEMBRANE AREA :

                INITIAL PRODUCT WATER FLUX
                                                     I    r
                                                                   .2 = 40.7lpd/m*
                                                             PH CONTROL FAILURE
                                         465 psi (32.1 kg/sq. cm)

                                         1350 ft2 (125m*)

                                         13.9 gpd/ft2 (566 Ipd/m*)
     10                          |QZ                         10s

                                       HOURS ON  STREAM



Figure 5. Decline rate of product water flux under constant feed pressure operation.
                                                                                               10'

-------
en
          100
           90
          -80
        o
        LU
        3 90
        tr
         <
         (O
           80
           90
           80
                              \
           pH CONTROL LOST
                                                                                      • BIZ FLUSHING
1969-JULY
 10        20      30
 I	I	I
                                                  AUGUST
                                                10        20
                                               _J	I
                                             30
                                                    SEPTEMBER
                                                   10       20
                                                   I        I
     |NaN03 ADDITION TO
     'CARBON COLUMN BEGUN
              OCTOBER
20
       30
 NOVEMBER
JO       20
I	|_
                                    30
                                     I
 DECEMBER
10       20
I        I
                      t
  I
                     30
NaN03 ADDITION TO
CARBON COLUMN ENDED

         FEBRUARY
        10      20
	I	1_
                          30
                    MARCH
                  10       20
                 J	U
        30
                                                        TIME
                                                     30
      1970-JANUARY
30        10        20
 I	I
                                                                 APRIL
                                                              10        20
                                                                                                      30
         Figure 6. Variation of rejection vs. operation time under constant feed pressure
                   operation.

-------
 in the feed.   Thus,  the increased concentration  of the nitrate ion in the
 feed caused the overall rejection to  decrease  slightly.

      During March  of 1970,  at approximately  6,000  hours  of operation, the
 overall  salt rejection  started to decrease slowly.   This trend continued
 throughout  the remainder of the study.   On August  16,'1970,  when  the sys-
 tem was  taken offstream, the overall  salt rejection  decreased  to  77 per-
 cent after  a  total of 9,475 hours of  on-stream operation.

      Figures  7 through  15 summarize the  salt rejection variations  from
 June 18,  1969 to August 14,  1970 for  each of the nine  pressure vessels.
 As indicated  in these figures,  the greatest  decline  in salt  rejection
 occurred  in pressure vessels 1,  2, and 3.  The initial  and final  salt re-
 jection  values for th'ese pressure vessels averaged about 93  percent and  45
 percent,  respectively.   The  salt rejection for the pressure  vessels 4
 through  7 decreased  from 93  percent to 80 percent, and for the pressure
 vessels  8 and 9 from 94 percent  to 90 percent.                     :

      Since  each pressure vessel  contained three spiral-wound modules  in
 series,  the salt rejection  calculated for each pressure  vessel  represented
 the  overall performance of  the  three  modules.  In order  to determine  which
 modules  in  each  pressure vessel  were  responsible for the decline  in salt  re-
 jection,  a  conductivity probe was  used to make conductivity measurements  of
 the  entire  system.   The results  of these measurements  are  summarized  in
 Table  5.  All  modules in  pressure  vessels 1, 2, and 3  showed a  deterioration
 in their  ability to  reject salts.  The No. 2 module in the pressure vessel
 4, No. 2  and  3  modules  in the pressure vessel  5, and the No'. 2 module  in
 the  pressure  vessel  6 also showed a decline  in salt rejection.

 Water  Quality

     The  chemical analyses conducted  on the feed water, product water, and
 the  brine waste  during  the first  phase of the pilot plant study are summar-
 ized in Table 6.  The percent rejections for the various ions are calculated
 using  the blended feed  and product values only.  As indicated in the table,
 the  overall  rejection of the inorganic ions,  as measured by the TDS reduc-
 tion, was about  91  percent.   The system demonstrated excellent rejection of
 calcium, magnesium, sulfate, and phosphate ions,  while it seemed very poor
 in the rejection of potassium and nitrate ions.

 CONSTANT PRODUCT FLUX RATE OPERATION

 Feed Pressure

     During  the first phase  of study,  the feed  pressure for the system
operation was maintained constant at  about 32.1 Kg/sq cm (465 psi).  How-
ever, the feed pressure  was  varied during the second phase of the  pilot
 plant study  to maintain  a constant product water  flux rate of 407  1/sq
m/day (10 gal/sq ft/day).
                                    27

-------
ro
CO
          100
          80
          60
                         I    T
I    I
   JUNE
10   20  30
  JULY        AUG.
10  20  30  10 20  30
   SEPT.
10  20  30
            I    I
   OCT.        NOV.
10  20  30   10  20  30
                                         DEC.
                                      10 20  30
                                                                                                   JAN. -70
                                                                                                 10  2p  3

          60
          40
                FEB.        MARCH       APRIL       MAY         JUNE     w  JUi,Y   ^^ *^AUG.        SEPT.       OCT.
                20  30   10  20  30  10  20 30   10  20  30  10 20  30   10  20  30  10  20  30   10  20  30  10
                 I    I    I    I    I   I   I   I    I    I    I   I   I    I	I    I    I    I   I    I    1    I    I   I
                                                           TIME
       Figure  7.  Salt rejection vs. operation time in pressure vessel  no. I.

-------
ro
10
          100
          80
        | 60

        o
        UJ
        uj 40
        QC  U

         100
        CO
          80
          60
          40
                                  T	T
                   T	1	1	1	1	1	1	T
                                                                           1	1	1	T
 ^ JUNE       JULY        AUG.        SEPT.        OCT.        NOV.        DEC.       JAN -70
 10  20  30   10  20  30   10  20  30  10  20  30  10  20  30  10  20  30   10  20   30   10  20  30
-I	1	L__l	1	1	1	1	1	1	J	1	|	i    i    i   i    I    i    i   i   i    i    i
               FEB.
               20  30
                I    I
          MARCH
        10  20  30
        I    I   .1
   APRIL
 10  20 30
J	I	L
AUG.        SEPT.     OCT.
 20  30   10  20  30   10
 I    I    I	'   '   '
                                                       TIME
        Figure 8. Salt rejection vs. operation time in pressure vessel no.  2.

-------
CO
o
          100
           80
        g  60

        o
        UJ

       •3. 40
          100
<
CO
           80
           60
           40
                 i  :  \    i    i   i    i    i   \    i    i   i—i    ii
                                                                                        1	1	T
            JUNE       JULY        AUG.       SEPT.        OCT.        NOV.        DEC.        JAN.-70
         10  20  30   10  20  30  10  20  30  10  20  30  10  20  30  10  20 3O   10  20  30  10  20  30
          I    I   I   'I    '   I'   '	I	1	1	J	1	1	1	1	1	1	1	1	1	1	L_
FER        MARCH
 20  30  10  20  30
  I'll'
                               APRIL
                             10  20  30
                             ' _ I _ I
                                                  MAY
                                               10  20  30
  JUNE
10  20  30
I   I    '	L
                                                                         J	L
AUG.        SEPT.      OCT
 20  30   10  20  30  10
 I'll'	l_
                                                    TIME
       Figure 9. Salt rejection vs. operation time in pressure vessel no. 3.

-------
OJ
        100
         80
 .60
z
o


§  40
UJ
01 100
         80
         60
        40
T — i
                    i — i — i — r~i — i — i — i — i — r~r — T — i
                                                                             T — i — i — i — i — r
                 JUNE         JULY        AUG.        SEPT.        OCT        NOV.        DEC.        JAN.-70
               10  20  30  10  20  30  10  20  30  10  20  30  10 20  30  10  20  30   10  20  30   10  20  30
               II   I   II    I   I    |   I   I    I    I   I   I    I   I   I    I    I    I   I    I    i   i
              FEB.        MARCH       APRIL       MAY        JUNE       JULY        AUG.        SEPT.     OCT.
              20  30  10  20  30  10  20  30  10  20  30   10  20 30   10  20  30   10  20  30  10  20  30  10
               I    I   i   I    I   I   I    I   I   I   I    I    I   I    I    I   I    I    I    I   I    I    I   I
                                                   TIME
      Figure 10. Salt rejection vs. operation time in  pressure vessel no. 4.

-------
                           SALT  REJECTION, %
CD
                  o>
                  o
             oo
             o
                                 o  *
                                 o  o
O)
o
00

O
O
O
CD
o
CO
•

o
T3
CD

Q


5'
CD
T3

CD


CO
C

CD



CD


CO
CD_




O


O1
        H

        2
        m
              CD
•o


•o
              3
                          -o
                                       •s
             toe
            •oz
               m
            — W
           *^o
            -»c
           •^os
            -o
                                       roc
                                       •02
                                         m
                           -,
                          •s?
                            H
                          -SQ
                                       •85

-------
                                   ee
(Q


 CD
                             SALT REJECTION,%
 Of
"CD"
 o

 o'
 3
O
•o
CD
O

-------
                          SALT  REJECTION,  %
I
OJ
      0)
      O
              CD
              O
O  *
00
0)
o
CD
O
O
o
CD
o

o'


CO

o
•o
0>

3
(0
 CD
 O)
 (0

 -t
 CD


 CD
 CO
 CO
 CD
 O

 74'
•—o
  O


ho
             e_
            roc

           •8
           -o
           •8
             o
                                      ro<
                                     'Of


                                     •8
                          -8
                                       !l
                                     •8
                           -8
                           •8
                                     -o
                           _i\»
                           8
                                     •8
                          _ro
                             m
                                     —o
                          _ro£
                          "o =
                                        ->i
                                       |0

-------
co
          100
           80
        o
        UJ

        UJ
           60
 40
                                    1   T
                                                            I   -I   I    IIIIIT
         JUNE        JULY        AUG.        SEPT.        OCT        NOV.         DEC.       JAN.-70
       10  20  30  10  20  30  10  20  30   10  20  30  10  20  30  10  20   30   10  20  30  10   20   30
       1    1   II   I    t   I   lit    I   I   I    I    I   _l   1   I    i   I   1   J	L	l_
lOOb-
   p  •>
        CO
           80
           60
           40
                FEB.        MARCH      APRIL        MAY        JUNE       JULY        AUG.         SEPT.      OCT
                 20  30   10  20  30  10  20  30  10   20 30   10  20  30  10   20  30  10  20  30   10  20  30   10
                 I    I    I    I   II   \    I   I    II   I    II   II   I   I    I   III    I    I
                                                       TIME
       Figure 14, Salt rejection vs. operation time  in pressure vessel  no. 8.

-------
                                 9£
                          SALT  REJECTION, %

o
00
o
o   •*
o   o
0>
o
00
o
o
o
m
             T»m
             OCD
     o
                                        04

                                      *^O
                              -o
                               roc

                               °5
    roc
    •oz
     m
               CO

                                          m
                    -o
                              •83
                              •g
                                        •85
                                        •g
                                        •g
                                       -o

-------
            TABLE 5. INDIVIDUAL MODULE SALT REJECTION
           TESTS CONDUCTED AT THE END OF CONSTANT FEED
                    PRESSURE OPERATION STUDY
               Module #1         Module #2         Module #3
    1        Off Off Off Off   Off Off Off Off   Off Off Off Off

    2        Off Off Off Off   Off Off Off Off   Off Off Off Off

    3        Off Off Off Off   Off Off Off Off   Off Off Off Off

    4        7.8 8.2 7.7 7.6   6.9 Off Off 7.4   6.3 8.6 8.7 10.0

    5        5.9 6.2 5.9 5.7   5.5 5.8 6.9 Off   Off 8.0 9.6  9.0

    6        6.77.98.25.4   7.5 Off. Off Of f ... 9.8 10.0 6.6  7.4

    7        6.4 5.2 6.9 6.8   4.1 3.9 4,.0 4.4   7.5 8.3 7.8 .8.0

    8        4.9 3.8 3.7 4.3   5.2 5.4 5.8 5.9   6.5 6.8 3.6  4.1

    9        6.57.05.95.1   6.26.07.26.7   6.76.86.7  6.4
Notes:   1.  Measurements  taken at one foot (30.5 cm)  intervals
           using an Industrial  Instruments  Model  RA  4-WA-S4-Kf.

        2.  Readings should be multiplied  by 30 (cell  constant)
           to get conductivity,  ymhos/cm.

        3.  Off reading was off scale.
                              37

-------
          TABLE 6. SUMMARY OF WATER QUALITY ANALYSES
           FOR THE. PERIOD OF ZERO TO 9,475 HOURS OF
               CONSTANT FEED PRESSURE OPERATION
Parameter
Sodium, mg/1 Na
Potassium, mg/1 K
Calcium, mg/1 Ca
Magnesium, mg/1 Mg
Chloride, mg/1 Cl
Sulfate, mg/1 SO/+
Phosphate, mg/1 PO^-P
Ammonia, mg/1 NHs-N
Nitrate, mg/1 N03-N
Turbidity, JTU
Total COD, mg/1
TDS, mg/1
Blended
Feed
129
16.5
40.8
24.5
95
318
10.4
13.9
7.7
1.0
10.1
744
Product
15.0
4.7
1.5
0.8
14.1
3.0
0 . 1 5
1.6
3.5
0.1
1.0
67
Brine
452
49 . 1
132
98.5
326
1310
38.8
44.9
16.6
2 . 9
32.7
2800
Rejection
%
88.5
71 .5
96.5
97.0
85.0
99.0
98.5
88.5
54.5
90.0
90.0
91 .0
Notes:  1.  Analyses  were run on once-a-week grab samples  taken
          at 8:00 A.M.

       2.  Blended feed  was a mixture of carbon-treated
          secondary effluent, sulfuric acid and chlorine
          solution.

       3.  Rejection (%) = 100X (Blended feed concentration -
          Product concentration)/(Blended feed concentration)

       4.  COD = Chemical oxygen demand.

       5.  TDS = Total  dissolved solids.

                               38

-------
      A summary  of  the  overall  performance  during the second phase of the
 pilot plant  study  is shown  in  Figure  16,   As  indicated  in  Figure 16, the
 initial  feed pressure  necessary to maintain this constant  flux was about
 24.8  Kg/sq cm (360 psi).  The  system  remained at this feed pressure until
 120 hours of operation when  the water recovery was  increased from 75 to 80
 percent.  After 150 hours of operation, the feed pressure  required "for the
 system to maintain the 407  1/sq m/day (10  gal/sq ft/day) product flux rate
 was found to fluctuate between 26.9 Kg/sq  cm  (390 psi)  and 34,5 Kg/sq dm
 (500  psi).   Further increase of the feed pressure was noted at about 2,400
 hours of operation.  This increase was believed to Be a result of the in-
 sufficient velocity in the  circulation of  the cleaning  solution and the
 water flush  through the pressure vessels during the membrane cleaning  I
 cycle.   At 2,830 hours of operation,  the modules were cleaned twice" a week
 instead  of once a  week.  This  new practice was continued for a three week
 period to thoroughly clean  up  the membrane surface.  The cleaning solution
 flow  rate through  each pressure vessel was increased from  11.4 to 34.1 1pm
 (3  to 9  gpm).   The water flushing flow rate was also increased from 11.4 to
 26.5:;Jpm (3  to  7 gpm).  After  this flow rate adjustment, there was a de-
 crease in the feed pressure.   The pilot plant system was depressurized for
 approximately 84 hours after an enzyme-detergent cleaning  at 3,553 hours of
 operation.   This special depressurization  treatment resulted in a 8.3
 Kg/sq cm (120 psi)  decrease  in the feed pressure to maintain the constant
 407 1/sq m/day  (10 gal/sq ft/day) product water flux rate.  At the,end of
 the pilot plant study, the  rapid decline of the salt rejection was accom-
 panied with'a low  feed pressure, about 20.7 Kg/sq cm (300  psi).  This be-
 havior could be attributed  to'  some membrane breakup developed in the system.

 Salt  Rejection

      During  the second phase of the pilot plant study, the product water
 flux  rate was-kept constant  at 407 1/sq m/day (10 gal/sq ft/day) by vary-
 ing the  feed pressure.  As indicated  in Figure;!6, under this mode of
 operation, the  overall salt  rejection was steadily maintained at 95-per-
 cent  throughout the initial   3,600 hours of opeatiom  After 3,600 hours of
 operation, the  salt rejection started to decline gradually.  This decline
 was primarily attributed to  the poor  salt rejection of the modules in  \
 pressure vessel No. 2.  At.4,414 hours of operation^ these modules were
 replaced with three used modules which had 1,933. hours of operating time
 accumulated  from other similar study.   At the time of the module replace-
 ment,   the salt  rejection was about 60 percent for the original  set of  ;
 modules.  The new  set  of modules substantially improved the salt re-
 jection to 94 percent.  However, the product water flux rate for the new
 set of modules  in  the  pressure vessel  No.  3 was only about 317 1/sq m/day
 (73  gal/sq  ft/day), while the overall flux rate for the entire system
was 407 1/sq  m/day  (10 gal/sq ft/day).  The explanation was that the   :
 membranes might have been affected by the irreversible compaction and
 fouling.  In  addition, the three replacement used modules were operated
 under*38 Kg/sq cm  (550 psi)  feed pressure in a previous study,  while
 they were operated under 27.6 Kg/sq cm (400 psi)  in  this study.
                                    39

-------
               017
FEED PRESSURE, PSI
SALT REJECTION,  %
«o
c
p
a°
o 5
!₯
D :±

5' 3
? Q
3
Q.
-h
CD
Q.
TJ
-t
CD
CO
CA

CD
<
Q
•^
Q"

f
<
O
•o
CD
Q
o'
—+
3
CD
C
. 3
Q.
CD
~* ,
0
O
w
1
— h
E"
X

o
0?


ro
O
o




o
o




O)
o
o

I-
o
c
30
00
z o
§
rn
• 9>
0
0
o


Fo
0
o




—
o
0

35
o
o
a -^ 01
> O O 00
> o o o
	 •- 1 J 1
!
• " * 1 • ' £
• . 1

* * *oi|§
• — * -n
» » M ' i
• O (A 00 Z O
.A* bgF5"Oi
• to c "33 o ^
— • 5 S c ° -n
* . "N m i z >
tl II
' * A ' 2
; . • • *™1 , .
' m
•
- 0 ^n
r~
— ' . • • . • • c ' :.
0 • 0)
• x
*
0
0
0 •
:, *
• - • t * . .
• . ,...'.'
8 f
' • "• ' • • '
• .-.
* • : • •
' • ." 5§|
**. >>^
"" * 3 $
. • . • . ^^ £x , ^n
v ^^ O 71 :

• ^^ID
• - • c - •
• X
• >
s
•
0* • - • ,
• ' • •
• ' ' • I • . . •
1 , 1 1
CO C
0 . G
1 *

!
•^
••
•
•
*
•
• , _
•
y
«
&
_
' ,
. • . • ™"
•
•
c
•
0
• ' •
*
i
. '*•
/•
.«
• . _j
• *
/
•
^ _
•
v
•
• • '. *• .
• ' .
.
»• •
*
?
.*
^
i

-------
                                 Lfr
~n
(O*
ov


o
o
l 4^ en
:> O O 00
3 O O O
I • I "
0 •
*
•" •
*.
o •
.•
•* - p
*  c m
• 3 H w z
0 * • ? 00
• 55
• o o
* mm
0) V)
• mm
o • °°
s *°
• si
* . W W
oo
o •
* O 
-------
                                      Zfr
31

     CO
     r- m
     m o
     (A

-------
«o
CD
o>

O
3;
•wV
3
C
CD
Q.





































0
8




Ol
o
o
o

Ol
10
0
o
Ol
4^
o
o
X •
o
c
CO 01
0§
z o
CO
H
m
J>
2 CO
0
0

en
o
o
0


CO
IM
0
o
0)
4*
0
o
. «..v ^i • »b.ww« **~9 • - *f • W*^fc»l IAL_UI«.\/ 1 IV/ll* /O
'=---'• •-''... : '".'"' ' -"
O O O 00 (0
o o o o o
. «1- ........ f 1 1 •
o • ,

• " .
* •
• •
• : • .
"" • * ' * "•
•
" • ': •
0 • •
• - • •
"" • • ""'
• 0

• •
"" • • ~
* - ':'':•:*
• •
* •
* , •
9 ' A ••

• * " r*

• WSJ •
• 2m ' ••:
•on •
° * CO *
~" r~ PI * *
•mo \
o •
•
9
~ •
*• '"' *
* "-
0 • • ..'
*'.


• 9
0 • •
*. * ..
• • -
• 0
	 _Q 	 L_f._. I I ••
Q
O















































-------
•n
(0*
c
55
O
0
3_
3*
c
(D
a.







































FEED PRESSURE, PSI SALT REJECTION, %
«S ol w -4 CD 
O)
o
o

0>
00
0
o




-g
o
o
o
o
c
3D
W ->l
ro
o o
z o
H
3D
m
c
o




a
o
o


a

c
00
o
o
o
1 • . ' '

9* * ,
• ' *
0 • % •
* ,
r- . ''• • ' '*
T) OI * •
s£ i 1 •
5 *" * ~
" Si • /
o • •
r * •
o • .
30 • * •
Z • - •
5 ° *
m •_ .
o •
*• * •
• •
• •
« • ~
m • ' ' ' —
• •
• * m
e • •
• t * .
*•* *
* • *
o • •
• to •
• H •
• C •
o
• _1
j> • H * —
• 2 •
• i •
* z *
• -t •
• r»i •
• 0 •
.* w . *
"" *



1 1 1 1


-------
     At 6,677 hours of operation, the feed pump failed, which necessitated
a preservation of the membrane modules by feeding the system with chlori~
nated tap water for a period of 365 hours until the pump was repaired.
Upon resumption of the system operation, the salt rejection was found to
have decreased from 91 percent to 81 percent.  The occurrence of the sub-
stantial decrease in salt rejection efficiency at the same time that the
system was inoperative and on chlorinated tap water could have been a co-
incidence.  Within the next 1,100 hours of operation, the salt rejection
rapidly declined to 65 percent at which time the entire study was terminated
at 7,803 hours of operation.

Water Quality

     The average water quality data for the system operations during the
initial 6,700 hours are presented in Table 7.  As shown in both Table 6
and Table 7, similar rejection efficiencies for various ions were achieved
by the system operated with constant feed pressure in the first phase of
the study and the system operated with constant product flux rate in the
second phase of the study.  However, the ion rejection by the system with
constant product flux rate mode of operation greatly reduced after 6,700
hours of on-stream operation.  Table 8 shows the summary of the water
quality analyses from 6,700 to 7,803 hours of operation in the second phase
of the pilot plant study.

MEMBRANE MODULE STABILITY

Under Constant Feed Pressure Operation

     The performance of the membrane modules in each of the nine pressure
vessels (three modules per vessel) from time zero to 4,800 hours of opera-
tion is summarized in Table 9.  The following points of clarification may
be necessary for interpreting the data presented in this table.

     A. The difference between the initial (at time zero) and present (at
4,800 hours of operation) feed pressures in the downstream modules was the
result of decreased pressure drop through the system.  At time zero, the
total feed (equivalent to product plus brine) was about 61.3 1pm (16.2
gpm), while at 4,800 hours the total feed was dropped to 45.4 1pm (12 gpm).
Consequently, the pressure drop through the system also decreased from
5.9 Kg/sq cm (85 psi) at time zero to 3.1 Kg/sq cm (45 psi) at 4,800 hours.
This caused the feed pressure in the downstream vessels to increase.  The
initial and present feed pressure of vessels No. 1, 2, and 3 remained the
same because the feed pressure was a controlled operating parameter.

     B. The small  differences between the "A" (water permeability co-
efficient) values  specified by the Gulf Environmental Systems Company
(GESCO) and those Calculated from the initial operating conditions could
be explained as follows: The values of "A" specified by GESCO were taken
from a test run at 41.4 Kg/sq cm (600 psi) and 2,000 mg/1 sodium chloride
feed solution.   The "A" values calculated at time zero were based on a
                                    45

-------
            TABLE 7.  SUMMARY OF WATER QUALITY ANALYSES
             FOR THE  PERIOD OF ZERO TO 6,700 HOURS OF
               CONSTANT PRODUCT FLUX RATE OPERATION
Parameter
Na
K
Ca
Mg
Cl
SO,,
P04-P
NH3-N
NOs-N
TCOD
DCOD
TTOC
DTOC
TDS
TURBIDITY, JTU
Raw
Feed
mg/1
109
13.
58.
12.
104
69
18.
14.
1.
8.
4.
2.
1.
547
1.
Blended
Feed
mg/1

3
5
6


3
4
30
1
7
4
2

2
168
20
97
56
173
467
15
24
1
12
7
3
2
1084
1

.2
.3
.9


.4
.3
.43
.7
.4
.2
.5

.8
Product
mg/1
23.
2.
2.
0.
31.
4.
0.
2.
0.
0.
0.
0.
0.
77.
0
8
5
4
47
1
3
55
5
85
43
38
47
40
7

Brine
mg/1
390
40.
226
49.
520
1093
34.
52.
2.
26.
18.
6.
5.
1778
3.

2

0


1
1
17
5
7
3
0

1
Rejection
%
86.
87.
97.
99.
82.
99.
96.
89.
40.
96.
94.
85.
83.
92.
100
9
1
5
2
1
1
4
6
9
6
8
1
7
8

Notes:  1.  Raw feed was carbon-treated secondary effluent.

       2.  Analyses were run on once-a-week grab samples taken
          at 8:00 A.M.

       3.  Difference between raw feed and blended feed was
          due- to. H2SQif addition, chlorination and brine
          reci rculation.        .       .        .

       4.  TOC - Total organic carbon.
                               46

-------
             TABLE 8, SUMMARY OF WATER QUALITY ANALYSES
             FOR THE PERIOD OF 6,700 TO 7,803 HOURS OF
                CONSTANT PRODUCT FLUX RATE OPERATION
Parameter
Na
K
Ca
Mg
Cl
SOt
P04-P
NH3-N
N03-N
TCOD
DCOD
TDS
TURBIDITY, JTU
Raw
Feed
mg/1
114
12.
53.
11.
132
58.
10.
17.
0.
5.
2.
572
1.

3
0
4

6
0
0
75
0
6

4
Blended
Feed
mg/1
136
14.9
67.7
14.3
143
324
12.2
18.3
0.55
6.0
3.4
807
2.3
Product
mg/1
57
5.8
11.5
2.89
109
52.8
3.56
7.88
0.47
2.4
0.8
291
0
B r i n e
mg/1
257
27.3
135.6
24.6
193
760
28.2
36.7
0.45
138
6,3
1491
3.6
Rejection
%
' 58.1
61.1
83.0
79.8
23.8
83.7
70.8
56.9
14.5
60.0
76.6
63.9
100
Notes:  1.  Raw feed was carbon-treated secondary effluent-

       2.  Analyses were run on once-a-week grab samples.
          taken at 8:00 A.M.

       3.  Difference between  raw feed and blended feed was
          due to H2SOlf addition, chl ori nation, and brine
          recirculation.
                               47

-------
                             TABLE 9
         PERFORMANCE OF MODULES IN EACH PRESSURE VESSEL
FROM TIME ZERO TO  4,800  HOURS  OF  CONSTANT  FEED  PRESSURE OPERATION
Pressure Vessel
Initial Feed
Pressure, psi
Present Feed
Pressure, psi
Initial Flux
Rate, gpd/ft2
.pa
•°° Present Flux
Rate, gpd/ft2
% Reduction
in Flux Rate
Initial "A" x
1 0 5 Specified
by GESCO
Initial "A" x
105 Calculated
Present "A" x
105 Calculated
% Reduction
in "A" Value
1

465

465

15.4


10.7

30


2.17

2.29

1.61

30
2

465

465

15.8


10.9

31


2.21

2.36

1 .64

31
3

465

465

14.8


10.5

29


2.19

2.20

1 .56

29
4

430

450

13.9


9.2

34


2.18

2.25

1 .43

36
5

430

450

14.2


9.4

34


2.17

2.29

1 .46

36
6

405

440

13.8


8.8

36


2.23

2.36

1 .40

41
7

405

440

12.9


8.0

38


2.03

2.20

1.26

43
8

390

428

12.8


8.5

34


2.18

2.28

1 .38

39
9

380

420

12


8

29


2

2

1

36






.1


.6




.16

.22

.43


                        (Continued)

-------
                                 TABLE 9 (continued)
Pressure Vessel
Initial Influent
Flow, gpm
Initial Effluent
Flow, gpm
Average Initial
Flow, gpm
Present ..Influent
Flow, gpm
Present Effluent
Flow, gpm
Average Present
Flow, gpm
Initial Salt Re-
jection Specified
by GESCO
Initial Salt Re-
jection Calcu-
lated
Present Salt Re-
jection Calcu-
lated
Flux Decline Slope
1

5.36

3.77

4.57

3.94

2.83

3 . 39


93.5


93.0


90.0;
-0.066


5

3

4

3

2

3


94


93


89
-0
2

.36

.77

.57

.94

.83

.39


.5


. 0


.0
.077


5.

3.

4.

3.

2.

3.


93 .


93.


88.
-0.
3

36

77

57

94

83

39


0


0


0
081


5.

4.

4.

4.

3.

3.


93.


93.


93.
-0.
4

65

19

92

24

27

76


5


5


0
077


5

4

4

4

3

3


94


94


95
-0
5

.65

.19

.92

.24

.27

.76


•5


.0


.0
.075


4

2

3

3

2

2


94


93


94
-0
6

.19

.80

.50

.27

.40

.84


.0


.5


.0
.104


4

2

3

3

2

2


94


94


96
-n
7

.19

.80

.50

.27

.40

.84


.0


.5


.0
.089


5

4

4

4

3

4


93


93


96
-n
8

.59

.26

.93

.79

.90

.35


•5


.5


.5
.077
9

4.26

3.00

3.63

3.90

3.00

3.45


92.5


91.5


96.0
rO.052
Notes:  1.  "Initial"  = at time  zero;      "Present"  =  at  4,800  hours  of  operation.
       2.  GESCO = Gulf Environmental  Systems  Company, San  Diego,  California.

-------
feed containing approximately 700 mg/1  TDS and a feed pressure varying from
26.2 Kg/sq cm (380 psi) to 32,1  Kg/sq cm C465 psi).   The GESCO pointed out
that at lower feed pressure, the value of "A" was hfgher than at higher
feed pressure.  This would explain why all "A" values calculated from the
initial conditions were slightly higher than the values specified by the
GESCO.  A second explanation for the discrepancy in  "A" values was that the
osmotic pressure of the feed water was neglected in  the calculations.  This
simplified the calculations and  only introduced an error of 1 or 2 percent,

     C. The influent, effluent,  and average flows were calculated by
assuming equal flow distribution in the parallel pressure vessels.

     The following observations  were made from the data presented in Table
9:                                              '

     A. The modules in the pressure vessels 6 and 7, which experienced the
highest reduction in flux rates, the highest reduction in "A" values, and
the highest in flux decline slopes, showed the lowest average feed flows.
Initially, the feed flow to these modules averaged 13.2 1pm (3.5 gpm); at
4,800 hours of operation, this declined to 10.61pm  (2,8 gpm).  The GESCO
recommended that the minimum flow in each module should be between 11.4
1pm (3 gpm) and 15.1 1pm (4 gpm).

     B. The modules in vessel 9  had the lowest flux  decline slope.  Since
these modules received the poorest quality feed, one would expect the flux
decline slope to be greater than that experienced in the preceding modules.
This apparent discrepancy may be explained by noting the differences be-
tween initial (26.2 Kg/sq cm or  380 psi) and present (29 Kg/sq cm or 420
psi) feed pressures.  The lower  flux decline slope observed for the modules
in the vessel 9 occurred because the operating pressure increased with time.
This indicates a true picture of the effect of fouling on the entire sys-
tem cannot be obtained by looking at the individual  flux decline slopes.
A better measure of fouling would be the decrease in "A" values between
time zero and 4,800 hours.  The  percent reduction in "A" value for the
modules in vessel 9 was greater  than those experienced by the modules in
vessels 1, 2, and 3.  This indicates that the fouling in the downstream
modules was more severe than in  the upstream modules.  The changes in the
feed pressure and average flows  during the study make it difficult to de-
termine the effects of fouling through the entire system.

     C. The salt rejection for the modules in the vessels 1, 2, and 3 de-
creased from the initial values.  While they stayed  the same in vessels
4 and 5, they increased in vessels 6 to 9.  The modules in the vessels 1,
2, and 3 had exhibited an increase in salt rejection up to approximately
3,000 hours after which the salt rejection started to decrease slowly.

     At the end of the first phase of the pilot plant study, all the spiral•
wound modules were removed from the system and sent  to the Gulf
Environmental Systems Company for testing to determine which modules had
lost salt rejection ability and why this had occurred.  The results of;
                                      50

-------
 the GESCQ tests are summarized in Table 10,   The tests  were conducted at
 55,2 Kg/sq cm (800 psi) with 10,000 mg/1  sodium chloride solution,   !
      As indicated in Table 10, the salt rejection of the lead modules de^
 finitely fell  off, while some of the modules  in the pressure vessels  8
 and 9 still  had rejections above 80 percent.   The number of distribution
 of salt rejection range is indicated below:
                                           Number of Modules
            Salt Rejection (%}                in Each Range
               ,0-9                            0
                 10 - 19                            1
                 20 - 29                            7
                 30-39                            1
                 40 - 49                            1
                 50-59                            4
                 60 - 69                            4
                 70-79                            4
                 80-89                            5
                 90-99                          _0_
                                                  27
     Table 10 also shows the water permeation coefficient before and after
the pilot plant  study,  which accumulated a total of 9,475 hours of ori-
stream operation.  In all cases except three, the permeability coefficient
dropped below the  initial value.  No significant location dependence of
the decline was  demonstrated.
     Four modules among the twenty-seven modules were selected for further
dye checking, visual inspection, and membrane sample testing.  The results
of these observations are shown in Table 11.
     The GESCO membrane tests did not show the exact cause of the membrane
deterioration.   However, three possible fouling mechanisms were postulated:
     A.  Hydrolysis of the membrane caused by  the high pH of the enzyme-
detergent cleaning solution.
     B.  Some trace substances in the feed water attacked the membrane,
                                     51

-------
                           TABLE 10.  RESULTS  OF  MODULE  TESTS CONDUCTED AT

                         THE END  OF CONSTANT FEED  PRESSURE  OPERATION  STUDY
en
ro
: MODULE #1
Pressure A R
Vessel
Ti
1 2,27
2 2.15
3 2.05
4 2.23
5 2,30
6 2 , 1 7
7 2.23
8 2,07
9 2.12

Tf
1 .96
2,04
2,06
1,72
.1.82
1,60
1 .87
1.69
1 ,57

Ti
93,1
94,3
93.5
93,3
94.8
93,7
91.8
95.9
93.2

Tf
19.0
24,6
26.0
56,5
61,7
61.1
56,1
88.2
83,1
MODULE #2
A R

Ti
2,29
2,25
2,29
2,26
2.28
2.33
1.94
2.21
2,14..

Tf
2,25
1,96
2,25
1,85
2.03
2.07
1,69
1,44
1.49

Ti
92,7
94,7
93,8
92,6
94,3
93,0
96,3
94,2
90,1

Tf
20,0
21, Q
27,3
27,4
53,5
67,5
81,1
77,5
81,8
MODULE #3
A R

Ti
2,04
2,23
2,23
2,04
1.94
2,18
1,92
2,26
2.21

Tf
1,92
1 V91
2,85
3.13
1,66
1,60
1,34
1,77
1,66

Tt
94,9
94,3
91,0
94,6
94,5
95,3
94,6
91,1
94.1

Tf
56,2
28,9
42.0
37,2
66,4
72.0
74,9
82.3
75,6
  Notes:   1. A = Water permeability  coefficient,  (g/sq cm/sec/atm) X 105,
          2. R = Salt rejection, %,••
          3, Ti= Initial value; Tf > Final value  after 9,475  hours of operation,

-------
              TABLE  11.  RESULTS  OF  DYE  CHECKING,
        VISUAL INSPECTION AND MEMBRANE SAMPLE TESTING
      MODULE
           TEST  RESULTS
Module #2
Pressure Vessel No.
Module #1
Pressure Vessel No.
Module #3
Pressure Vessel No.
Module #2
Pressure Vessel  No.
 Integrity was good except for a
 small product tube leak and a mem-
 brane pinhole caused by a crease
 in the product water channel
 material.  The module appeared to
 be quite clean.

 No leaks were observed and module
 integrity was good.  There was
 visible evidence of fouling and
 membrane rejecting surface attack
 (indicated by dye pickup).  Some
 local areas did not pick up dye.

 No leaks were observed and module
 integrity was good.  Moderate
 fouling and membrane rejecting
 surface attack were evident.
 There seemed to be more membrane
 surface attack near the product
tube.
No 1eaks were
integrity was
some  evidence
brane surface
observed and module
good.   There was
of foul ing and mem-
attack.
                              53

-------
     C. Some substances in the fouling layer attacked the .membrane,

     The hydrolysis was believed to be the most probable cause for the de-
terioration of the membrane surface.

Under Constant Flux Rate Operation

     During the second phase of the pilot plant study, three membrane
modules, one from each of the first three pressure vessels, were removed at
5,678 hours of on-stream operation and sent to the GESCO for membrane
evaluation.  The results of the membrane evaluation are summarized below:

     A. Module from the Pressure Vessel No. 1: .

        Testing Feed        "A", g/sq cm/sec/atm       Rejection. %
       Tap Water
       pH adjustment)            2.62 x 10"5               79.2

       2,000 mg/1 NaCl               —                   57.3

       "A" = Water permeability coefficient

     The module was probed to check for possible leaks, but the results in-
dicated a uniformly poor rejection over the entire module length.  The
module was also dye-checked and opened for visual inspection.  A substan-
tial amount of dirt was present, both in the brine spacer and on the mem-
brane.  A white, flaky deposit (probably calcium sulfate) was noted near
the product water tube.  Some areas of the membrane seemed to adsorb the
dye more than others, indicating poorer rejection in these areas.

     Membrane samples were taken from the two types of areas and tested.
The results are as follows:

           Sample           "A", g/sq cm/sec/atm       Rejection, %

       Heavy Dye area            5.41 x 10"5               40.1

       Light Dye area            4.14 x.10"5               52.4

     B. Module from the Pressure Vessel No. 2:

        Testing Feed        "A", g/sq cm/sec/atm       Rejection, %

       Tap Water (No
       pH adjustment)            1.57 x 10'5               95.7

       2,000 mg/1 NaCl              —                    88.4

       Tap Water (pH ad-         1.51 x 10"5               94.6
       justed to 5.5 to 6)

                                      54

-------
      C, Module from the  Pressure Vessel No, 3:

        Testing Feed         "A", g/sg cm/sec/atm       Rejection. %

       Tap Water  (No
       pH adjustment)            1,76 x 10"5               94.1

       2,000 mg/1 NaCl               —.                   82.4

       Tap Water  (pH ad-
       justed to  5.5 to
       6)                        1.69 x 10"5             ;  93,3

     The membrane tests indicated;that the membranes were hydrolyzed
The degree of hydrolysis seemed quite severe.   Visually, it was difficult
to determine the extent of membrane degradation.  Exposure to high pH
feeds of cleaning solutions could cause hydrolysis.   However, this did
not occur in this instance.  Bacterial  action  was  another factor, but
continuous chlorine addition with 1.5 to 2.0 mg/1  chlorine residual
should be an adequate preventative.   Therefore,  the  real  cause for the loss
in the rejection  ability of the membrane could not be clearly determined.
                                    55

-------
                                    SECTION 7

                              PROCESS COST ESTIMATE

     Based on the pilot plant studies conducted at Pomona Advanced Wastewater
Treatment Research Facility, a cost estimate has been prepared for a 37,850
cu m/day (10 MGD) spiral-wound reverse osmosis plant in demineralizing a
carbon-treated secondary effluent.   The major assumptions made for this cost
estimate are listed in the following:

     A. The TDS of the blended feed for the reverse osmosis plant is about
1,200 mg/1;

     B. The water recovery for the process is about 80 percent;

     C. The product water flux rate is approximately 407 1/sq m/day (10
gal/sq ft/day) at 25°C;

     D. The process is capable of reducing or rejecting 90 percent of
the blended feed TDS;

     E. The sodium perborate at 2 percent concentration is used as the
membrane cleaning solution, with pH of the solution adjusted to 7.5 with
sulfuric acid;

     F. The membrane cleaning is performed once a week, or at an interval
of 3,054 liters of product water per square meter of membrane area (75
gallons per square foot of membrane area);

     6. The effective membrane life is only one year;

     H. The capital cost  is amortized for 20 years at 5 percent interest
rate;  and

     I. The reference date of the cost estimate is August, 1973.

     The  initial  capital  cost including the feed  pumps, membranes, pH
controllers,  chlorinators, chemical  feed  systems, booster  pumps, brine
recirculation pumps, and  a post treatment  system  for final pH adjustment
is about  3.66 million dollars for a  37,850 cu m/day  (10 MGD) spiral-
wound  reverse osmosis plant.  The total membrane  cost  is about 1.15
million  dollars.   Since the membrane  has  to be  replaced every year, the
cost for  membrane replacement is approximately  8.2
-------
 to two  years,  then the membrane replacement cost .will  be substantially re-
 duced to  4.'Q4/1 ,QQO liters 05,4^/1,000 gallons),

      The  annual  maintenance material  cost is based on  5 percent of the capi-
 tal  cost,  excluding the cost of membranes.   The  labor  requirements include:

      A. One  man-hour per clearring  schedule  for a 378.5 cu m/day (0.1  MGD)
 section of the plant;  and

      B. Three  man-years for operating the 37,850 cu m/day (10  MGD) plant.

      The  total power cost (U/kwh)  for the  37,850  cu m/day (10 MGD) plant
 operation  is estimated to be about  2.0^/1,000  liters (7.8(^/1,000  gallons).
 The  unit  costs for the various  chemfcals  used  in the reverse osmosis  pro-
 cess  are  estimated as  follows:

      A. Sodium perborate  =  $0.37/Kg  ($0,17/1b);

      B. Triton X-100 non-ionic  detergent  =  $0.84/Kg ($0.38/15);

      C. Carboxy methyl  cellulose = $0.97/Kg  ($0.44/lb);

      D. Sulfuric acid  = $Q.04/Kg ,($0.02/lb>; and

      E.  Chlorine =  $Q,Q9/Kg  ($0.04/lb).

     According to  the  above  chemical  unit costs, the total expenses for
process chemicals will  amount to approximately 1  -U/l,000  liters  (4.3<£/
1,000 gallons).

     Table 12 summarizes the various  parts of the total process cost esti-
mate.  As  indicated  in  the table, the total  process  cost is approximately
14.9(^/1,000 liters  (57.4<£/l ,000 gallons) for one year membrane life.  The
cost can be reduced to about 10.7<£/1,000 liters  (41,3<£/l ,000 gallons) by
improving  the membrane  life to two years.  Both  cost estimates do not in-
clude the  costs for the carbon adsorption pretreatment and the brine
disposal.
                                    57

-------
                           TABLE 12
PROCESS COST ESTIMATE FOR 37,850 cu m/day (10 MGD)  SPIRAL-WOUND
                     REVERSE OSMOSIS PLANT
Amortization of Capital
   $3.66 x 106; 20 years @ 5%
Operation and Maintenance
   Chemicals (H2SOit, C12
   and clean ing agent)
   Membrane Replacement
      One-year membrane life
      Two-year membrane life
   Maintenance Materials
   Power
   Labor
Total Process Cost:
   One-year membrane life
   Two-year membrane life
                                   .000 gallons   <£/! ,000 1 iters
                                      8.8               2.3
 4.3

31.5
15.4
 3.4
 7.8
 1.6

57.4
41.3
                                                        -1.1

                                                        8.2
                                                        4.0
                                                        Q.9
                                                        2.0
                                                        0.4
                                                       14.9
                                                       10.7
                               58

-------
                                REFERENCES

1. Dryden, Franklin D.,. "Mineral Removal by Ion Exchange, Reverse Osmosis,
   and Electrodialysis."  Presented at workshop on Wastewater Reclamation
   and Resuse, South Lake Tahoe, California (June, 1970).

2. English, John N., Masse, Arthur N., Carry,  Charles W., Pitkin, Jay B.,
   and Haskins, James E., "Removal of Organics from Wastewater by
   Activated Carbon."  Chemical  Engineering Progress, Vol.  67, No.  107
   (1970).
                                    59

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
1. REPORT NO.
 EPA-600/2-78-169
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
 Demineralization  of Carbon-Treated Secondary
 Effluent by Spiral-Wound Reverse Osmosis Process
             REPORT DATE
             September 1978(Issuing  Date)
            6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

 Ching-lin Chen  and  Robert P.  Miele
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 County Sanitation  Districts of Los Angeles County
 Whittier, California   90607         ,    •
            10. PROGRAM ELEMENT NO.
              1BC611
            11. CONTRACT/GRANT NO.

              14-12-150
12. SPONSORING AGENCY NAME AND ADDRESS
 Municipal Environmental  Research Laboratory--Cinn.,  OH
 Office of Research  and. Development
 U.S. Environmental  Protection Agency
 Cincinnati, Ohio  45268
            13. TYPE OF REPORT AND PERIOD COVERED
              Final  Report 7/71 - 6/73
            14. SPONSORING AGENCY CODE
              EPA/600/14
 15. SUPPLEMENTARY NOTES
 Project Officer:   Irwin J. Kugelman  513-684-7631
 ^.ABSTRACT A  56.8 cu m/day (15,000 gallons/day)  spiral-wound reverse osmosis  pilot plant
 was operated  at the Pomona Advanced Wastewater  Treatment Research Facility on  the
 carbon-treated  secondary effluent.  The specific  objectives for this study were  (a)
 to establish  the effective membrane life for wastewater demineralization with  carbon
 adsorption pretreatment; (b) to determine  the reliability of the process performance;
 and (c) to derive a realistic process cost estimate.   The study was first conducted
 on a constant feed pressure basis, and then it  was run on a constant product water flux
 rate basis.   During'the first phase of the study, pH  adjustment was not practiced for
 the weekly enzyme-detergent membrane cleaning procedures.  However, this was practiced
 in the  second phase of the study.  The results  from both phases of studies substantiate*:
 the fact that the membrane effective life  was only about one year in demoralizing  the
 carbon-treated  secondary effluent.  A cost estimate for a 37,850 cu m/day (10  i>1GD)
 reverse osmosis olant indicated that for membranes with only one-year life the process
 cost was about  14.94/1,000 liters (57.44/1,000  gallons).  However, the cost  could be
 substantially reduced to 10.74/1,000 liters (41.34/1,000 gallons) for membranes  with
 two-year life.   Both cost estimates did not include the costs for carbon adsorption
 oretreatment  and brine disposal.
 17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                   DESCRIPTORS
                                               b.lDENTIFIERS/OPEN ENDED TERMS
                                                                            COSATI Field/Group
  Demineralizing
  Desalting
  Water Reclamation
  Membrane Fouling
  Reverse Osmosis
  Carbon Adsorption
 13B
 18. DISTRIBUTION STATEMENT

  Release to Public
19. SECURITY CLASS (ThisReport)
 Unclassified
21. NO. OF PAGES
  68
                                               20. SECURITY CLASS (This page)

                                                Unclassified
                                                                          22. PRICE
 EPA Form 2220-1 (Rev. 4-77)
                                              60
                 *USGPO: 1978 — 657-060/1474 Region 5-11

-------