United States
Environmental Protection
Agency
Robert S Ker- f
Laboratory
Ada CX 74620
Resaarcti and Development
Total  Recycle
Systems for
Petrochemical
Waste Brines
Containing
Refractory
Contaminants

-------
                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-600/2-79-021
                                           January 1979
TOTAL RECYCLE SYSTEMS FOR PETROCHEMICAL WASTE BRINES
         CONTAINING REFRACTORY CONTAMINANTS
                         by
                    M. A. Zeitoun
                    C. A. Roorda
                    G. R. Powers
                Dow Chemical Company
               Freeport, Texas  77541
                Grant No. S803085-01-0
                   Project Officer

                Thomas E. Short, Jr.
              Source Management Branch
  Robert S. Kerr Environmental Research Laboratory
                Ada, Oklahoma  74820
  ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
         OFFICE OF RESEARCH AND DEVELOPMENT
        U.S. ENVIRONMENTAL PROTECTION AGENCY
                ADA, OKLAHOMA  74820

-------
                                DISCLAIMER

     This report has been reviewed by the Robert S.  Kerr 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.
                                    ii

-------
                                  FOREWORD

     The Environmental Protection Agency was established to coordinate admin-
istration of the major Federal programs designed to protect the quality of our
environment.

     An important part of the Agency's effort involves the search for informa-
tion about environmental problems, management techniques,  and new technologies
through which optimum use of the Nation's land and water resources can be
assured and the threat pollution poses to the welfare of the American people
can be minimized.

     EPA's Office of Research and Development conducts this research through a
nationwide network of research facilities.

     As one of these facilities, the Robert S. Kerr Environmental Research
Laboratory is responsible for the management of programs to: (a)  investigate
the nature, transport, fate and management of pollutants in ground water;
(b) develop and demonstrate methods for treating wastewaters with soil and
other natural systems; (c) develop and demonstrate pollution control technol-
ogies for animal production wastes; (e) develop and demonstrate technologies
to prevent, control, or abate pollution from the petroleum refining and petro-
chemical industries, and  (f) develop and demonstrate technologies to manage
pollution resulting from combinations of industrial wastewaters or industrial/
municipal wastewaters.

     This report contributes to the knowledge essential if the EPA is to meet
the requirements of environmental laws that it establish and enforce pollution
control standards which are reasonable, cost effective, and provide adequate
protection for the American public.
                                     CJ  C. Ji
                                        W. C. Galegar
                                          Director
                      Robert S. Kerr Environmental Research Laboratory
                                      iii

-------
                           ABSTRACT
Petrochemical wastewaters containing relatively high concentra-
tions of salt and refractory organics were selected to study
their feasibility for total recycle.  A combination of reverse
osmosis and electrodialysis was operated as a hybrid system
using the pretreated wastes, to produce reusable water and a
concentrated brine.

Stream 1, characterized by a salinity of 1.4 percent NaCl when
neutralized, contained an average of 0.5 percent organic amines
and ammonia.  It was found to contain enough silica to foul the
membranes.  Stream 2, contained 4.5 percent dissolved solids and
precipitated aluminum salts when neutralized with acid.  Its
organic content of aromatics averages 90 mg/1 of TOC.

A total recycle system for Stream 1 is composed of ammonia
stripping, carbon adsorption and recovery of the organic amines,
neutralization and silica removal before the ED-RO hybrid.  For
a 1 MGD Stream 1 waste treatment facility the estimated capital
cost is $6,949,550 (1975 dollars) and the estimated operating
cost is $5,291,700 per year.  The carbon adsorption step has an
estimated operating cost of $2,358,000 recovering organics suit-
able for recycle at an estimated value of $3,000,000 per year.

A total recycle system for Stream 2 is composed of neutralization,
settling and centrifugation to recover Al(OH)3, and carbon adsorp-
tion with thermal regeneration before the ED-RO hybrid.  For a 1
MGD Stream 2 treatment facility, the estimated capital cost is
$9,170,000 and an operating cost is $2,293,700 per year.  No
valuable organics are recovered from this stream.

The combined electrodialysis-reverse osmosis system is not con-
sidered economically feasible when applied to industrial waste-
waters containing relatively high concentrations of salt.
                                IV

-------
                           CONTENTS

Foreword	   iii
Abstract	**....    iv
Figures	    vi
Tables	 viii

     1.   Introduction	   1
     2.   Conclusions	   4
     3.   Recommendations 	   5
     4.   Waste Waters Selection and Characterization 	   6
     5.   Stream 1.  Total Recycle System	 10
               Pretreatment Studies 	 10
               Reverse Osmosis and Electrodialysis -
                 Stream 1	 18
               Total Recycle System Design and Cost
                 Estimate	 37
     6 .   Stream 2.  Total Recycle System	 41
               Pretreatment Studies 	 41
               Electrodialysis-Reverse Osmosis
                 Studies - Stream 2 . „	 49
               Total Recycle System Design and
                 Cost Estimate	 49
     7.   Economic Evaluation - Brine and Potable
            Water Reuse	 56

Bibliography 	 61
Appendices

     A.   Metric Conversions 	 63
     B.   Analytical Methods 	 64
     C.   Ammonia Removal Tests 	 67
     D.   Total Recycle System Design and
            Cost Estimates - Stream 1	 75
     E.   Total Recycle System Design and
            Cost Estimates - Stream 2	 89
                                v

-------
                             FIGURES

Figure                                                      Page

  1    General Systems Engineering - Total
       Recycle System	  3

  2    Adsorption Isotherms - Stream 1	 12

  3    Carbon Column Test - Cycle 4	 15

  4    Regeneration - Activated Carbon - Stream 1	 17

  5    Silica Adsorption on Activated Alumina	 19

  6    SV-5 Electrodialysis Unit	 20

  7    Reverse Osmosis Equipment	 22

  8    Electrodialysis of Neutralized Untreated
       Stream 1	 23

  9    Electrodialysis of Pretreated Stream 1.	 25

 10    Electrodialysis of Pretreated Stream 1 -
       Limiting Current Density	 27

 11    Continuous Electrodialysis of Pretreated
       Stream 1 Feed	 28

 12    Continuous Electrodialysis of Pretreated
       Stream 1 plus RO Reject Mixture	 29

 13    Limiting Current Density of Combined
       Pretreated Stream 1 and RO Reject Stream	 30

 14    ED-RO Integrated System Treatment of Stream 1	 33

 15    ED-RO Combined Operation - Critical Parameters -
       Stream 1,  without Silica Removal	 36

 16    ED-RO Combined Operation - Critical Parameters -
       Stream 1,  with Silica Removal	 38

 17    Total Recycle System - Stream 1,  Flow Diagram
       and Material Balance	 39
                               VI

-------
Figures (continued)
                                                            Page

 18    Gravity Settling - Aluminum Hydroxide Sludge -
       Stream 2	 43

 19    Gravity Settling - Aluminum Solids Flux - Stream 2... 44

 20    Carbon Adsorption Isotherm - Stream 2	 47

 21    Carbon Column Test - Stream 2	 48

 22    Continuous ED-RO Pretreatment Stream 2	 51

 23    Total Recycle Stream - Stream 2	 53

 24    Cost of ED-RO of Brine Solutions - Water Treatment... 57

 25    Cost of ED-RO of Brine Solutions - Production
       of Process Water	 58

 26    Cost of ED of Brine Solutions Salt Production	 59

 27    Cost of ED-RO Treatment of Stream 1	 60

 C-l   Schematic Diagram of Conceptual LMWT Process	 69

 C-2   Synthetic Stream 1 - Waste Water Liquid
       Membrane Treatment	 72

 C-3   Exchange of NH4+ on Clinoptilolite from
       Stream 1 Waste Water.	 74

 D-l   Activated Carbon Treatment of Stream 1	 76

 D-2   Total Recycle System Stream 1 - Carbon
       Treatment Process Design and Material Balance	 78

 D-3   Total Recycle System Stream 1 - Silica Removal,
       Design Basis and Material Balance	 84

 D-4   Electrodialysis-Reverse Osmosis of Pretreated
       Stream 1, Flow Diagram and Material Balance	 87

 E-l   Neutralization and Solids Removal - Stream 2,
       Flow Diagram and Material Balance	 90

 E-2   Flow Diagram and Material Balance for Carbon
       Adsorption Treatment - Stream 2	 93

 E-3   Electrodialysis-Reverse Osmosis Treatment -
       Stream 2, Flow Diagram and Material Balance	 95


                              vii

-------
                             Tables

Table                                                       Page

  1    Inorganic and Organic Constituents - Stream 1	  8

  2    Inorganic and Organic Constituents - Stream 2	  9

  3    Comparison of Commercial Activated Carbons for
       Amines Adsorption - Stream 1	 13

  4    Carbon Bed Test Conditions - Stream 1	 14

  5    Carbon Bed Test Results - Stream 1	 16

  6    Pretreated and Neutralized Stream 1 Waste
       Water Composition	 24

  7    Reverse Osmosis Evaluation	 32

  8    Electrodialysis-Reverse Osmosis Combined Test -
       Stream 1	 35

  9    Cost for Total Recycle System - Stream 1	 40
                                           r.
 10    Gravity Settling Design Data	 45

 11    Comparison of Activated Carbons Removal of
       Organics for Stream 2	 46

 12    Continuous Electrodialysis-Reverse Osmosis Run
       Pretreated - Stream 2	 50

 13    Continuous Electrodialysis-Reverse Osmosis Run
       with Pretreated Stream 2	 52

 14    Cost for Total Recycle System - Stream 2	 55

 B-l   Gas Chromatographic Analysis Conditions -
       Stream 2	 66

 C^l   Liquid Membrane Treatment of Synthetic Stream 1
       Waste Water	 71

 D-l   1-MGD Stream 1 - Activated Carbon Treatment,
       Estimate of Direct Capital Cost	 80
                             viix

-------
Tables (continued)
  D-2  1-MGD Stream 1 - Activated Carbon Treatment,
       Estimate of Operating Cost	 81

  D-3  1-MGD Stream 1 - Neutralization, Estimate of
       Capital and Operating Costs	,	 82

  D-4  Silica Removal - Stream 1	 83

  D-5  Activated Alumina Treatment of Stream 1 -
       Estimate of Capital and Operating Costs	 85

  D-6  1-MGD Stream 1 - Electrodialysis-Reverse
       Osmosis Design	 86

  D-7  Cost Estimate Summary - Electrodialysis-
       Reverse Osmosis Treatment of Stream 1	 88
                                3.x

-------
                            SECTION 1

                          INTRODUCTION
Petrochemical process waste streams have varied chemical con-
stituents.  Some components are easily degradable and are amenable
to biological treatment processes, some are not, while others are
toxic to living matter.  Consequently, an optimal balance between
physical, chemical, and biological treatment must be considered
in the development of total recycle and reuse master plan.

Most of the highly contaminated waste streams from petrochemical
plants originate from process areas, including condensed steam
from stripping operations, wash waters from drum cleaning opera-
tions, water formed or eliminated during various reactions and
other similar in-process sources.  A variety of contaminants are
found in these waste waters, including unreacted feed stock
chemicals, products, by-products, side-products and spent cata-
lytic materials.

Caustic washes and acidic washes are frequently used to extract
or neutralize acidic and basic components from process streams
resulting in a waste stream containing inorganic salts in addition
to the organic components.

Refractory (non-biodegradable) pollutants are of special interest
because of their possible adverse effects and their persistence
in the environment.  The establishment of more stringent effluent
quality standards will necessitate greater utilization of advanced
treatment processes.  At the present time, however, application
of advanced treatment methods to petrochemical wastes has been
limited.  In addition, there are many unanswered questions regard-
ing the treatment and/or recovery of chemicals resistant to
biodegradation.  Although the salvage of unreacted chemicals and
recovery of usable products and by-products from waste waters is
desirable, economic considerations have usually dictated treat-
ment rather than recovery.

The evaluation and development of advanced chemical and physical
treatment methods, and the determination of the proper integra-
tion of selected unit operations to constitute an optimum waste
treatment system is necessary.  A system of total recycle involves
both component recovery and water reuse.

-------
The types of physical and  chemical  treatment processes commonly
investigated for the treatment  of petrochemical wastes include
stripping, adsorption,  extraction,  ion exchange and oxidation.
Because these wastes almost  invariably contain dissolved salts,
desalination methods such  as  reverse osmosis and electrodialysis
are required to concentrate  the inorganic components of the waste
streams  (total recycle  system).

The Dow Chemical Company has  investigated electrodialysis as a
process  for the concentration of desalination plant brines for
the Office of Saline Water.   Under  OSW Contract No. 14-30-2676,
Dow designed and built  an  electrodialysis brine concentrator
which operated at  Roswell, New  Mexico from February 1971 to
August 1972.  The  Dow testing concluded under Contract No. 14-30-
3031  with the combined  operation of a hybrid reverse osmosis-
electrodialysis pilot plant  from August 1972 to May 1973.

Results of  the operation of  the Roswell facility showed that
reverse osmosis and electrodialysis can be utilized together to
produce desalted water  and concentrated brine.  For a variety of
 feed  water  salinities,  the reverse  osmosis-electrodialysis system
 can economically  achieve a high degree of separation between the
water and the  salt.

 The capabilities  of reverse  osmosis and electrodialysis com-
 plement each other well.   Reverse osmosis produces a low salt
 product quite well from brackish water but is a poor brine con-
 centrator.   Electrodialysis  easily  produces a ten to fifteen-fold
 concentration of  salt in  a single  stage but often requires
 several stages operated in series to produce a satisfactory de-
 salted water quality.   The operation at the Roswell plant de-
 monstrated that the combination of  electrodialysis and reverse
 osmosis can be used to simultaneously concentrate the reject
 wastes from the reverse osmosis unit to more than 20 weight
 percent sodium chloride and  produce reusable product water.

 A general schematic of the total recycle system studied is shown
 in Figure 1.  The main objective of this investigation was to
 apply the necessary combination of  these unit operations to each
 of the simulated wastes so as to approach a total recycle system.

-------



Feed

ontaining
HN3
d
•H
ft
ft
-H
H
-P
CO
U)
     Feed
Containing
 Organic
Contaminants
            Regenerant
                                   NH3.to recovery system-

                          -X-
  d
  o
o -P
XI ft
t_.t y j
rd o
u w
 Feed
                            Chemical
                            Oxidation
Regen-
     -ป
                               erant
                                     
-------
                           SECTION 2

                         .CONCLUSIONS


Petrochemical waste waters containing relatively high concentra-
tions of salt and refractory (non-biodegradable)  organic con-
taminants can be treated to remove the organics by such methods
as carbon adsorption.  The remaining salt solution can be concen-
trated by an electrodialysis-reverse osmosis desalination process
to produce fresh water and concentrated brine.

For the type of petrochemical waste waters studied,  the following
conclusions could be made:

     1.   The carbon adsorption of relatively high concentrations
          of organics from petrochemical waste waters is tech-
          nically achievable, but economically feasible only if
          the organics can be recovered and recycled to the
          production facility.

     2.   Low concentrations of organics that are highly adsorb-
          able on activated carbon cannot be removed from petro-
          chemical waste waters, at a reasonable cost,  because of
          the cost of regenerating the carbon thermally.

     3.   The combined electrodialysis-reverse osmosis  system is
          technically a feasible desalination process that pro-
          duces both concentrated brine and fresh water.
          Pretreatment of the waste water is required to remove
          any contaminants that can foul the membranes  of the ED
          or RO units.

     4.    For the range of salt concentrations (1.5-4.5 percent
          NaCl)  studied,  the combined electrodialysis-reverse
          osmosis system capital and operating costs were found
          to be high and the return from the value of the brine
          and fresh water produced would not economically justify
          the investment in the process equipment.

-------
                            SECTION 3

                         RECOMMENDATIONS
It is recommended that:

1.   In-place regeneration of activated carbon be investigated
     for each particular waste water in order to reduce the cost
     of the carbon adsorption treatment.

2.   Whenever possible, waste waters containing high salt concen-
     trations should be recycled without desalination and/or
     concentration.

3.   The application of a combined electrodialysis-reverse
     osmosis system would be limited to salt concentrations and
     situations where it is economically feasible.

-------
                            SECTION 4

           WASTE WATERS SELECTION AND CHARACTERIZATION


Five waste waters were initially considered for the demonstration
of the total recycle concept.  All were brines resulting from the
production of petrochemicals and contained various organic con-
taminants known to be refractory to biological oxidation.  An in-
depth characterization survey was done on the two waste waters
selected for demonstration.  The results of the study of the
first two waste waters and a cost analysis of electrodialysis-
reverse osmosis treatment of these brines led to the decision to
minimize characterization of the last three waste waters because
these contained much higher salt concentrations and were much
less adaptable to the total recycle system concept.

Procedures used for analysis of the various parameters in the
waste waters were taken, for the most part, from standard ana-
lytical reference texts (EPA, 1971; APHA et al, 1971).  Chromato-
graphic methods were developed as required to measure specific
organic constituents of the waste streams.

In all cases, waste water was sampled during normal operation of
the units producing the waste water.  Sampling was designed to
obtain daily representative samples.  Where required, multiple
daily samples were taken and composited.

The principal objective of the characterization was to determine
if any parameters or contaminants were present that would affect
the operation of the electrodialysis-reverse osmosis (ED-RO)
recycle system.  Such contaminants as calcium, magnesium and
sulfate ions, silica, suspended solids, organic acids and bases,
and some nonpolar hydrocarbons could be expected to foul the
membranes used in the ED-RO recycle system.  If these types of
contaminants are present in the waste waters, it is expected that
pretreatment of the brine would be required.
"Methods for Chemical Analysis of Water and Wastes", Environmental
Protection Agency, Water Quality Office, Analytical Quality
Control Laboratory, Cincinnati Office, 1971.

"Standard Methods for the Examination of Water and Wastewater",
Thirteenth Ed., American Public Health Association, Washington,
D.C. APHA, AWWA and WPCF, 1971.

-------
STREAM 1 - CHARACTERIZATION

Stream 1 is an organic production waste water stream containing
nitrogenous organic compounds.  The summary of the inorganic
constituents, organic constituents, and trace metal concentra-
tions of Stream 1 are given in Table 1.

The results indicated that the waste water might need treatment
for removal of silica, amines, and possibly calcium, magnesium,
and sulfate ions before being processed by the ED-RO recycle
system.  Approximately 70 percent of the total organic carbon
(TOC) and 83 percent of the organic nitrogen were accounted for
by gas chromatographic (GC) analysis.  It is known, however, that
higher molecular weight diamines that cannot be determined by GC
analysis are present in the waste water.

STREAM 2 - CHARACTERIZATION

Stream 2 is a petrochemical production waste water.  The summary
of the inorganic constituents, organic constituents, and trace
metals concentrations of Stream 2 are given in Table 2.

Approximately 95 percent of the total organic carbon (TOC)  was
accounted for by the benzene and ethylbenzene found by the gas
chromatographic analysis.  The results indicated that the waste
water will probably have to be treated for removal of the
aluminate ion, the aromatic hydrocarbons and possibly the silica.

-------
   Table 1.   INORGANIC AND ORGANIC CONSTITUENTS - STREAM  1

                             (Concentrations in mg/1)
   NaOH

 NaOH

 Total Dissolved Solids
   (TDS)

 Suspended Solids

 Ammonia-Nitrogen

 Inorganic Carbon (1C)

 Silica

 Chloride  ion

 Sulfate ion

 Total Organic Carbon
   (TOC)

 Organic Nitrogen

 Calcium

 Magnesium

 Iron

 Cu, Zn, Mn, Cr

 Ni

Al
Minimum
12.3
2,426
1,653
2,380
18
30
8
10
74
27
903
1,110






Maximum
13.2
21,646
12,160
15,770
84
300
64
80
1,015
40
3,843
3,460
•





Average
12.6
10,550
5,190
6,990
37
157
24
45
433
35
1,824
2,046
20
8
1.1
0.2
0.4
0.8

-------
Table 2.  INORGANIC AND ORGANIC CONSTITUENTS - STREAM 2



                               (Concentration in mg/1)
Parameter
PH
Total Alkalinity
as NaOH
Total Dissolved Solids
(TDS)
Suspended Solids
Inorganic Carbon
Silica
Chloride Ion
Sulfate Ion
Color-APHA units
Aluminum Ion
Total Organic Carbon
(TOC)
Total Oxygen Demand
(TOD)
Calcium
Magnesium
Iron
Cu, Zn, Mr, Cr, Ni
Minimum
11.3
1,600
20,800
2
2
nil
10,075
nil
20
325
21
74




Maximum
12.8
9,680
62,500
58
968
8.4
29,500
nil
90
920
197
329




Average
12.0
4,947
42,320
31
166
4.7
21,400
nil
45
680
86
126
7.6
1.0
0.7
1.6

-------
                            SECTION 5

                 STREAM 1.   TOTAL RECYCLE SYSTEM


 Stream 1 was selected to be studied in. detail for  the applic-
 ability for recycle as shown in Figure 1.   The objective was to
 investigate the required combination of unit operations  that
 would approach a total recycle system.  The reverse  osmosis and
 electrodialysis operating as a hybrid to produce reusable water
 and a concentrated brine suitable for recycle is the principal
 ionic management of the advanced treatment system.   Unit opera-
 tions preceeding the RO-ED system are necessary for  removal of
 any constituent that interferes with the operation of the RO or
 ED units, and if possible, for the recovery of valuable  constit-
 uents to improve tne economics of the recycle system.

 A.  PRETREATMENT STUDIES

 The characterization of Stream 1 indicated that at least four
 steps would be necessary before the waste water could be treated
 by the reverse osmosis-electrodialysis process.  These steps are
 ammonia removal and recovery,  amine removal and recovery, neutral-
 ization and possibly silica removal.   Ammonia removal is necessary
 to prevent the ammonia from appearing in the product water or
 the brine,  and recovery is desirable.  The amines  must be removed
 because they may cause fouling of the membranes and  recovery is
 highly desirable because of their high unit price  value.  Neutral-
 ization is required because the pH requirement of  the reverse
 osmosis membrane.   Silica may  have to be removed to  prevent
 membrane fouling.

 Ammon i a Remo va1

 Methods for removal and recovery of ammonia from waste waters are
 reviewed in Appendix C.   The liquid membrane process and ion ex-
 change  on clinoptilolite were  tested on Stream 1 and the results
 are  included in Appendix C.  Neither of these two  methods were
 found  attractive.   Conventional stream stripping of  the  ammonia
 was  used for the purpose of cost estimating the total process.

Amines  Removal  and  Recovery: Carbon Adsorption

 Nine carbons were evaluated to select the  best carbon for the
 adsorption  of the amines.   A standard isotherm was run with each
 carbon  using waste  water,  as received,  with no pH adjustment.
                                10

-------
Increasing weights of each carbon were added to 100 mis of
waste water in bottles that were shaken in a temperature
regulated water bath for 24 hours at 25ฐC.  The samples were
filtered and analyzed for total organic carbon and an adsorption
isotherm was plotted for each carbon.

Figure 2 shows a comparison of the results for three carbons and
the results for all the carbons are summarized in Table 3.  The
carbon with the largest capacity at 1000 mg/1 was Westvaco WV-G,
12 x 40 mesh.  Subsequently this one was used in the carbon bed
tests.

The carbon bed tests were run using four carbon columns operated
in series.  A variable speed peristaltic pump was used to feed
the beds and a three-way solenoid was placed in the line after
each column.  An electrical timer controlled the solenoids such
that after a 19-minute period, the solenoid would open for 15
seconds and divert a 20 ml. sample to a test tube in the SMI
model 1205 fraction collector.  Each column was sampled sequen-
tially from first to last and the sampling cycle started again.
The fraction collector advanced every 18 minutes so that an empty
test tube was always ready for the sample.  The collected samples
were analyzed for total organic carbon.  A few samples were also
analyzed for the amines by gas chromatography.

Table 4 summarizes the conditions for the bed tests.  The first
cycle was run at 22 ml/min.  (1.06 gpm/sq.ft.) during loading and
44 ml/min.  (2.12 gpm/sq.ft.) during regeneration of the carbon.
The balance of the cycles were run with these rates reversed.
Figure 3 gives the results from the last cycle.  The plots for
the other three cycles were similar.  Table 5 presents a summary
of the results.

After the first cycle of loading and regeneration, no further
significant change in capacity or TOG recovered was noted.

Further bed tests were run using one large carbon bed.  The
objective of this test was to provide a large amount of pre-
treated waste water for the electrodialysis-reverse osmosis
tests.  The column was 7' x 4" O.D.  The carbon bed was 6.5'x
3.5" I.D.  The bed volume was 11 liters and the weight of the
Westvaco WV-G carbon was 4400 grams.

The regeneration step was followed closely on the third cycle.
Figure 4 is a plot of the results.  The regeneration stream was
divided into three cuts; the first cut included the void volume
of the bed and neutralization of the caustic by the acid.  This
cut was basic and the TOC was that of the original feed.  This
cut could be recycled to the feed.  The second cut was the pro-
duct cut containing 90 percent of the TOC and was acidic.  The
TOC concentration was approximately 1.5 percent  or approximately
3.0 percent amines.  The last cut was low in TOC and high in acid

                               11

-------
     100

      80
  -   60
  m
mg TOG
g Carbon
      40
      20
      10
               At 25ฐC
               Westvaco Carbons
                  *200         400   600      1000

                          TOC Residual, mg/1
1500
        Figure 2.  ADSORPTION ISOTHERMS - STREAM 1

-------
Table  3.   COMPARISON OF COMMERCIAL ACTIVATED CARBONS
           FOR AMINES ADSORPTION - STREAM 1
         Initial  Concentration 1600 mg/1  TOC
                           Exhaustion Capacity
                                at 25ฐC
                           mg TOC/gram carbon
  Activated Carbon

Westvaco WV-G

Barneby Chaney SC

Westvaco WV-L

Hydrodarco 3000

Westvaco WV-W

Barneby Chaney PC

Witco 517
       •''f'?'
Filtrasorb 400

Filtrasorb 300
                                  90

                                  84

                                  72

                                  68

                                  62

                                  60

                                  56

                                  52

                                  42
                         13

-------
    Table 4.   CARBON BED TEST CONDITIONS - STREAM 1
Feed



Carbon



Columns



Bed Volume



Total Bed Height



Weight of Carbon



Cross Sectional Area



Temperature



Pressure



Regenerant



Regenerant Volume
Unneutralized Stream 1



Westvaco WV-G, 12 x 40 mesh



(4) 4-1/2 foot x 1 inch i. d,



(4) x 690 = 2760 mis



18 feet



1140 grams



5.1 sq  cm
25ฐC
4.5 inch Hg



2N HC1



2 bed volumes
                          14

-------
  1.0
   .8
 o
u
CJ
   .6
   .4
   .2
                                         1525 TOG
                                       I
I
                           10         15         20


                              Service Time, hours
          25
30
                 Figure  3.   CARBON COLUMN  TEST  -  CYCLE 4

-------
        Table 5.   CARBON BED TEST RESULTS - STREAM 1

              Feed, TOC  = 1525 mg/1
              Regenerant = 2N HC1 - 2 Bed Volumes

                                          Cycle	
Loading Rate, gpm/sq. ft    1.06      2.12     2.12      2.12

TOC Adsorbed, g            52.9      53.1     49.3      52.7

Capacity, mg TOC/g Carbon  47        47       43        46

Regeneration Rate,          2.12      1.06     1.06      1.06
  gpm/sq. ft.

TOC Recovered, g           39.1      49.4     48.3      48.5

TOC Recovered, %           74        93       98         93
                              16

-------
                   Large Column, 3rd Cycle

                               Product Cut
                             0.8 Bed Volumes
tn
U
O
28,000


24,000


20,000


16,000


12,000


 8,000


 4,000
             Recycle to
               feed
           0.7 Bed Volumes
I Recycle
I to acid
  Tank
                         I
                                                      I     I
              0.2   0.4   0.6   0.8   1.0   1.2   1.4   1.6   1.8   2.0

                         Bed Volumes  2N  HC1 Passes
      Figure 4.   REGENERATION - ACTIVATED CARBON - STREAM  1

-------
and could be recycled to the acid tank.  The column was finally
washed with a half bed volume of water to flush out the acid
before the feed was restarted.  The water wash also could be
recycled to the acid tank.

Silica Removal

The first electrodialysis-reverse osmosis test using pretreated
Stream 1 was terminated shortly after starting because of fouling
of the electrodialysis and reverse osmosis membranes.  Analysis
indicated the fouling substance consisted largely of amorphous
silica.

A literature survey indicated the best method for removing
silica from water was by adsorption on activated alumina.  This
was described in a report by Bell, et al, (1968) .

An adsorption isotherm was run to determine the equilibrium
capacity of the alumina for the silica.  Two different mesh
ranges of alumina were used.  Figure 5 shows the isotherm results.
The capacity of the 80-200 mesh alumina was 20 mg. silica/g.
alumina, significantly higher than the capacity for the 8-14
mesh.

A column bed test was then run.  A small plexiglass column was
filled with 129 grams of 30-40 mesh Alcoa F-l activated alumina.

About 24 liters  (190 bed volumes) of synthetic Stream 1 con-
taining 102 mg/1 silica was passed through the bed when break-
through occurred.  The breakthrough capacity was 20 mg of silica
per gram of alumina.  This same alumina was used in a three
column set-up to remove the silica from the feed for the electro-
dialysis-reverse osmosis experimentation.

The alumina was regenerated with 2 bed volumes of 1 N sodium
hydroxide.  It was indicated qualitatively that the alumina also
removed some of the trace amines and that they could be collected
in a concentrated fraction during regeneration.  The alumina
pretreatment will be used as required to remove the silica.

B.  REVERSE OSMOSIS AND ELECTRODIALYSIS STUDIES - STREAM 1

The electrodialysis unit used for these experiments was an Asahi
Chemical Industries Model SV-5 laboratory electrodialysis unit.
Figure 6 describes the SV-5 unit and operating parameters.  In
the SV-5 unit, the feed is pumped into the dilution recycle tank.
The dilution recycle is pumped through the dilution half cells
within the electrodialysis stack.  There the anions and cations
Bell, G. R. at al, Office of Saline Water Research and Develop-
ment.  Progress Report No. 286, 1968.


                               18

-------
x e
            O - 80-200 Mesh
            D - 8-14 Mesh
      .1
C0 = 75 mg/1 Silica
Temperature = 25ฐC
                     _L
      I   I  I
I
I
                     4   6  8  10     20    40
                     Residual Silica, mg/1
I  'I  I
                           60 80 100
   Figure  5.   SILICA ADSORPTION ON ACTIVATED  ALUMINA
                            19

-------
   Feed
                                1	ฃ

+




Dilution
Half
Cell
Brine
Half
Cell




—

    Product
                                            I
                      r
                                                  Cathode
                                                  Rinse
       Brine
     Production
Operating Parameters:

  Membranes	
  Active membrane area
  Linear, velocity-dilution  .
  Stack size 	
  Brine recycle rate  .  .  .  .
  Anode rinse flow rate.  .  .
  Cathode rinse flow  rate.  .
Standard reinforced anion &
0.75 dm2
15 cm/sec
5 cell pairs
0.3 gpm          2
100 .mVmin/amp-dit^
100 ml/min/amp-dm
cation -Asahi Chem.Ind.
                    Figure  6.    SV-5 ELECTRODIALYSIS UNIT

-------
are transported to the adjacent brine half cell compartments
under the influence of the direct current potential generated by
the anode and cathode.  The depleted dilution recycle then re-
turns to the tank.  The brine flow is recycled and the increase
in volume caused by the transport of water and brine from the
dilution compartment is bled off as the brine production.  The
depleted product water is bled off the dilution recycle tank.
The cathode is rinsed with HC1 to remove the hydrogen gas and
caustic that is produced at the cathode.  This stream is recycled
and replaced when depleted.  The anode is rinsed with NaCl to
sweep out the chlorine and acid produced at the anode.  This
stream is not recycled to prevent deterioration of the end mem-
brane from chlorine exposure.

A Universal Water Corporation laboratory model reverse osmosis
unit designed for  [small flat] membranes was modified for the use
of Dow hollow fine fiber reverse osmosis cartridges.  The car-
tridge used was a low pressure Dowex XFS-4167.07 Permeator with
a nominal capacity of 250 GPD.  Figure 7 describes the equipment.

Electrodialysis Preliminary Evaluation

Stream 1, neutralized to pH 7, but untreated with carbon adsorp-
tion was used as feed to determine the extent of the problems
encountered with the electrodialysis equipment.  The potential
across the stack was held at 3 volts and the current density was
allowed to vary.

The experiment was run for 12 hours with the brine reaching a con-
centration of 10.4 percent NaCl.  The dilution stream was replaced
with more feed and the run continued for 12 more hours.  The final
concentration was 12.8 percent NaCl  (Figure 8).  The organics were
concentrated in the brine stream.  The TOC was concentrated from
1235 mg/1 in the feed to 5,144 mg/1 in the brine and only 185 mg/1
in the dilution stream.  The organic nitrogen of 1180 mg/1 in the
feed was concentrated to 5,100 mg/1 in the brine and only 112 mg/1
in the dilution stream.  The stack was dismantled and the cation
membranes were found to be fouled.  Analysis of the dilution and
brine streams indicated that the amines were being transported.
However, the heavier amines must have been trapped in the mem-
branes.  Attempts to regenerate the membranes by soaking in
caustic failed.

The electrodialysis apparatus was then tested on a pretreated
Stream 1 feed with a composition shown in Table 6.  The dilution
and brine recycle tanks were filled with 2 liters of feed and
then the unit was operated until equilibrium was reached, in
10 hours.  No additional feed was added.  Figure 9 shows the
results of the test.  A concentration of 15.5 percent NaCl was
considered to be satisfactory level.
                               21

-------
                                                                 200 psi N2
NJ
                                   Hollow Fine Fiber
                                Reversis Osmosis Module
                   -*.
                                                                   I
                   G-
              High Pressure
                Pump
                                   Feed
                                   Jug
Product
  or
Permeate
          Brine or
          R.O.  Reject
Back
Pressure
Valve
                            Figure 7.    REVERSE OSMOSIS EQUIPMENT

-------
  140,000
  130,000
  120,000
  110,000
  100,000
   90,000
^Hi
op  80,000
K
 -  70,000
^^
g  60,000
   50,000
   40,000
   30,000
   20,000
   10,000
       0
                D - Brine  Recycle
                • - Dilution Recycle
J	I
I	I
I   I    I   I   I   I   I
         0  246   8  10  12 14  16 18 20  22 24  26  28
                              Time, hours
       Figure  8.   ELECTRODIALYSIS OF NEUTRALIZED
                    UNTREATED STREAM 1
                             23

-------
Table 6.   PRETREATED AND NEUTRALIZED STREAM 1
           WASTE WATER COMPOSITION

     pH                                  7

     NaCl, mg/1                     14,750

     CaCl2, mg/1                        59

     MgCl2, mg/1                        33

     NH3-Nitrogen, mg/1                134

     Total Carbon, mg/1                170

     Inorganic Carbon, mg/1             13

     Total Organic Carbon, mg/1        159

     Organic Nitrogen, mg/1            107
                      24

-------
J-
o
O - Current
A - Feed
D - Brine
                     4567
                      Time, hours
                            9   10
  Figure  9.   ELECTRODIALYSIS OF PRETEXTED STREAM 1
                           25

-------
As the current density, measured in amperes per square decimeter
 (amp/sq.dm.), is  increased through the system of membranes and
solutions,  the rate of electrical transport increases.  A current
density will be reached at which the concentrations of the electro-
lytes at the membrane interfaces on the depleting sides will
approach zero.  At this density, usually called the limiting
current density,  H  and OH~ ions from the ionization of water
will begin  to be  transferred through the membranes.  The presence
of layers of almost pure water at the membrane surfaces causes
the resistance of the membrane cells to be high, and is accompa-
nied by a higher  increase of voltage.  Figure 10 is the limiting
current density plot for the run.  The plot indicates that the
current density could have been operated a little higher (2.75
amp/sq.dm.) than  the 2.4 amp/sq.dm. at which the run was made.

To simulate the operation of the combined ED-RO unit, a run was
made in which the dilution stream from ED was processed in the RO
unit, then  the reject from RO was combined with the feed to the
ED unit.

The operation was the same as the previous run, except that the
feed was added continuously to the dilution recycle tank and the
dilution product  was bled off.  A small amount of brine was also
removed continuously from the brine recycle.

The dilution recycle flow was set at 0.5 GPM and the brine recycle
was '0.3 GPM.  The stack voltage was 3.0 volts and the initial
current was 1.8 amp/dm2.  At the start, 2500 mis. of feed was
placed in the dilution recycle tank and 1300 mis. of feed was
placed in the brine recycle tank.  Initial feed NaCl concentra-
tion was 1.41 percent.  The feed was added continuously to the
dilution recycle  tank when the dilution recycle concentration
reached 4000 mg/1 NaCl.  The feed rate was 18 mls/min.  The run
lasted 9i hours,  during which time the dilution recycle stabil-
ized at 2400 mg/1 NaCl and the brine recycle increased to 8.6
percent NaCl.

The unit was shut down and the dilution recycle was transferred
to the reverse osmosis unit and processed.  The reverse osmosis
reject concentration was 6760 mg/1 NaCl, which was mixed with
more electrodialysis feed and placed back in the electrodialysis
unit.  The  dilution recycle was now 1.067 percent NaCl.  The unit
was restarted with the same brine recycle.  When the dilution
recycle reached the 4000 mg/1 NaCl concentration, the feed was
started at  27 ml/min.  The results of the runs are plotted in
Figures 11  and 12.  The second run continued for 5 hours, during
which the dilution recycle stabilized at approximately 4000 mg/1
NaCl.  The  brine  recycle reached 12.3 percent NaCl.  The maximum
current density allowable was 2.18 amp/dm2 as indicated in the
limiting current  density plot in Figure 13.  No operational
problems were encountered during the experiment.  Analyses of
grab samples during the run indicated the total organic carbon

                               26

-------
   5.0_
•X  4.5
o
(0

n  4.0
3.5



3.0


2.5
         1.0
                              I
                          2.0


                     Stack Current, amps
                                                 3.0
                 2.0      2.5      3.0      3.5


                   Current Density,  amps/dm2
                                              4.0
4.5
   Figure 10.  ELECTRODIALYSIS OF PRETREATED STREAM 1 -

               LIMITING CURRENT DENSITY
                             27

-------
u
(0
2
                   - Brine Recycle

                   - Dilution Recycle
                         4567


                          Time, hours
9  10
 Figure 11.  CONTINUOUS ELECTRODIALYSIS OF PRETREATED

             STREAM 1 FEED
                          28

-------
a   70,000
                             • - Brine Recycle
                             • - Dilution Recycle
                          2       3   '
                           Time, hours
 Figure 12.   CONTINUOUS ELECTRODIALYSIS OF PRETREATED
             STREAM 1 PLUS RO REJECT MIXTURE
                          29

-------
   l.Or-
•H
nJ
O
    0.8
    0.7
ol   0.6
a,
 CO
    0.5
    0.4
•2   0.3
-P
c

I   0.2
    0.1


     0
0
                   2.18 amp/dm'
1.0
                                      2.0
3.0
                   Current Density, amp/ dm
     Figure 13.  LIMITING CURRENT DENSITY OF COMBINED

                 PRETREATED STREAM 1 AND RO REJECT STREAM
                             30

-------
and the ammonia nitrogen was  transported  and concentrated in the
brine.  The reverse osmosis membrane  also rejected the TOG and
the ammonia nitrogen.

Reverse Osmosis Tests

The electrodialysis dilution  recycle  is feed for the reverse
osmosis unit in the total recycle  system  concept.  Several ex-
periments were made in connection  with the electrodialysis
evaluation experiments.  The  objectives of these experiments
were to determine operating conditions for the reverse osmosis
unit to produce less than 500 mg/1 NaCl product at 200 psi, and
test for the possibility of membrane  fouling.  The results of
the runs are summarized in Table 7.   The  results indicate that:
Satisfactory quality product  water can be produced from the
reverse osmosis unit from 4000 mg/1 NaCl  feed at 200 psi and a
50-50 product-reject split.   The ammonia-nitrogen, TOC and
organic-nitrogen are all concentrated in  the reverse osmosis
reject, and no fouling of the membrane was indicated.

Evaluation of Combined Electrodialysis-Reverse Osmosis Treatment
of Stream 1             "                                ~~~~

The objective of this experiment was  to combine the previously
evaluated electrodialysis and reverse osmosis units and to op-
erate the combined units continuously on  pretreated neutralized
Stream 1 to determine the necessary design parameters.

A schematic of the combined laboratory units is given in Figure
14.  Neutralized pretreated Stream 1  [1]  is mixed with the reject
 [6] from the reverse osmosis  unit. This  stream  [2] is fed into
the dilution recycle loop.  A portion of  this stream  [4] is fed
to the reverse osmosis unit.  It is expected that this stream
will have a concentration of  4000  mg/1 NaCl.  Because the electro-
dialysis unit was not well matched to the reverse osmosis unit in
flow capacity, a large portion of  the reverse osmosis product  [7]
and the reject  [8] was recycled and mixed with the dilution re-
cycle  [4] to form the reverse osmosis feed [91.  A portion of
the product flow  [5] is taken off  continuously.  A portion of
the brine recycle is taken off as  brine product  [3] .  The feed
flow  [1J should be exactly equal to the sum of the brine  [3] and
the reverse osmosis product  [5].

Prior to the run, a standard  brine with a sodium chloride con-
centration of 14,000 mg/1 was run  under the same conditions as
the actual run.  The feed rate was 32 ml/min.  The brine pro-
duction rate was 3 mls/min.   The reverse  osmosis product rate was
29 ml/min.  The dilution stream NaCl  concentration was 4000 mg/1.
The reverse osmosis reject NaCl concentration was 7900 mg/1.  The
reverse osmosis product contained  640 mg/1 NaCl.  The reverse
osmosis split was 50 percent  product-50 percent reject.  The NaCl
rejection was 98.8 percent.   The operating current density was


                               31

-------
         Table 7.    REVERSE OSMOSIS EVALUATION

           Electrodialysis Dilution Recycle Feed
                       200 psi
NaCl, mg/1

Peed Flow, ml/rain.

Product NaCl, mg/1

Product Flow, ml/min.

Reject NaCl, mg/1

Reject Flow, ml/min.

Split
Run*
1
4030
704
486
356
7510
348
51 - 49
2
4210
834
476
404
7411
430
48 - 52

3770
709
503
353
6760
356
50 - 50
*Run 1 - E.D. Recycle from Pretreated Stream 1.
 Run 2 - E.D. Recycle, Stream 1.
 Run 3 - E.D. Recycle from Continuous Run.
                           32

-------
00
u>
          Neutralized Feed
Reject
                                           Brine Recycle
                                                                                Brine
                                                  Electrodialysis
                                   Dilution
                                   Recycle
                                   Revers
                                   Feed
                     2 Osmosis
                                                          .Product
                                                           Water
                                                          Permeate
                                                                   Reiect
                                  Figure  14.   ED - RO  INTEGRATED  SYSTEM
                                              TREATMENT OF  STREAM 1

-------
3.2 amps/dm2.  The limiting current density was found to be 3.5
amps/dm2.   No operational difficulties were noted.

This run was terminated and the feed was switched to the neutral-
ized pretreated Stream 1, with composition similar to that given
in Table 6.  The operating conditions for the experiment are
shown in Table 8.  The limiting current density for this run was
1.8 amps/dm2, and it was operated for 20 hours.

Figure 15 shows the changes that occurred in the critical op-
erating parameters during the run.  The current density had to be
continually adjusted to maintain a value of 2.4 amps/dm2.  As is
noted in Figure 15, the current density continued to decrease.
The adjustment back to 2.4 amps/dm2 caused the voltage to increase,
While this was occurring, the reverse osmosis unit was plagued
with decreasing product water recovery.  Also during the run, the
dilution NaCl concentration increased from 4000 mg/1 to 6700 mg/1.

The brine started at 146,000 mg/1 NaCl and decreased to 136,000
mg/1.  The reverse osmosis product NaCl concentration increased
from 500 mg/1 to 815 mg/1.

After 20 hours, the units were shut down because of the operation-
al problems, and the electrodialysis stack was dismantled.

A thick, oily film was found on the anion membranes.  The material
was collected for analysis.  The analysis of the material in-
dicated that it was 16.8 percent silicon or 36 percent silicon
dioxide.  The pretreated feed was found to contain 69 mg/1 silica.
The reverse osmosis reject stream contained 130 mg/1 silica.

It was apparent that the silica in the feed was fouling both the
electrodialysis membrane and the reverse osmosis membrane.  Also
the operating current density was much lower than it was during
the standard run.  The feed will also have to be treated to
remove silica.

Continuous Electrodialysis-Reverse Osmosis Run with Silica
Removed

Neutralized carbon treated Stream 1 was treated with activated
alumina to remove the silica.  The analysis of the feed was the
same as the previous run, except that the silica concentration
was now 10 mg/1.  The operating conditions were the same as the
previous run.  The limiting current density for this solution was
1.7 amps/dm2.

For a feed rate of 28 mg/min., a product from the RO unit of
21 ml/min. and a brine product rate of 3 ml/min. were obtained.
The reject of RO was 8700 mg/1 NaCl and the product was 400 mg/1
NaCl.  The brine concentration of 14 percent NaCl was ten times
the concentration of the feed.  The ED unit was operated at 0.74


                               34

-------
         Table 8.   ELECTRODIALYSIS - REVERSE OSMOSIS
                    COMBINED TEST, STREAM 1
Electrodialysis Unit:

  Membranes

  Stack Amps

  Stack Volts

  Cell Pairs

  Feed Flow

  Dilution Flow

  Dilution NaCl

  Dilution pH

  Dilution Temperature

  Brine Production


Reverse Osmosis Unit:

  R.0. Membrane

  Pressure

  Feed Rate

  Product Recycle
   Rate

  Reject Recycle
   Rate

  Split

  Product Rate
- A&K 101 Asahi Chemical  Ind.

- 1.6

- 7.3

- 10

- 30 ml/min.

-1.7 1/min.

- 4000 mg/1

- 7.1

- 26ฐC

- ^3 mls/min.




- Dowex XFS-4167.07 Permeator

- 200 psi

- 640 ml/min.

- 317 ml/min.


- 324 ml/min.


- 50 - 50

-27 ml/min.
                              35

-------
U)
CTv
         Volts  4
       cell pair
 85

,80

 75

,70
           Amps
           dm
                                     10
                            15
20
25
                                                                         50
                  40
                                                                         30
                                                                              %  Recovery
                                                                              of Product
                                                                                Water
                                         Hours
            Figure  15.  ED-RO  COMBINED OPERATION - CRITICAL PARAMETERS - STREAM 1
                         WITHOUT SILICA REMOVAL

-------
volts per cell pair and a current density of  1.6 amps/dm2.  Figure
16 plots the same critical operational parameter as shown in the
previous experiment.  The experiment was run  for 70 hours with no
operational difficulties.

The current density remained  stable throughout all of the exper-
iment.  The reverse osmosis recovery increased somewhat toward
the end of the run.   Of greatest interest is  the fact that the
operating current density was still much lower than that during
the standard run.  The probable cause  is the  low level of amines
present in the feed.  This emphasizes  the need for near complete
removal of these compounds.

C.  TOTAL RECYCLE SYSTEM  DESIGN AND COST ESTIMATE

A flow diagram of the total recycle system  of Stream 1 is given
in Figure 17, which  includes  a material balance and flow rates of
each  stream.  The ammonia removal  step is not shown on the flow
sheet, and if required,  its design is  based on conventional steam
stripping.

A summary of  the  estimated capital and operating costs of the
total system  is given in  Table 9.

Details of the design and cost estimates of the unit operations
required  for  the  total  recycle of  Stream  1  are  shown in Appendix D.
                                37

-------
                 .80


                 .75
          Volts  '
       cell pair
u>
00
           Amps

            dm2
                  I
                                                                            50
                                              40
                                                                            30
                                                                                Recovery
                          10
    20
30
 40

Hours
50
60
70
                 Figure  16.
ED - RO COMBINED OPERATION-CRITICAL PARAMETERS
STREAM-1, WITH SILICA REMOVAL

-------
OJ
vo
        (Stream 1)
Waste Water
        Organics - HC1
          Recycle

         4-5% Organics
                                                      Silica
                                                      Removal
                                       2N-HC1
                               Carbon
                               Adsorption
                                                      HC1
                                       Neutralization
                                           -e^J
                                           ion   I
                                                 1.3  %
                                                  ! 1.02 mgd  [
                                                                   fc ^  JNaCl~f4.5 %"i
                                                                       j_ ,09 mgd   j
                                                                                     Brine
                                                                   Electrodialysis
                                                                Reverse Osmosis
                                                                                         Product
--iNaCl <500 ppra
 il .93 mgdj
 T — — • — •-•-
          Ib./hr.
                                  ฉ
TOG
NaOH
Silica
HC1
NaCl
H2O
Total
600
3,000
21
—
300
343,600
374,521
630
3,000
21
—
3,100
402,940
409,721
2
—
—
9,800
—
162,600
172,400
568
—
—
—
—
70,900
71,466
_.
—
—
3,000
—
1,108
8,108
30
—
21
—
5,900
417,899
423,850
30
—
4
—
8,720
720,000
720,720
544
—
72
—
5,700
34,000
39,700
.ป— .
—
—
—
3,020
678,000
681,020
_ซ-
—
—
—
3,870
294,000
296,870
^ MM
—
—
—
150
384,000
384,150
                                Figure  17-  TOTAL  RECYCLE  SYSTEM -  STREAM  1
                                             FLOW DIAGRAM AND MATERIAL BALANCE

-------
  Table 9.   COST FOR TOTAL RECYCLE SYSTEM - STREAM 1



                                         1975 Dollars



Capital Cost;



  Ammonia Removal CSteam Stripping)        $1,156,000



  Carbon Adsorption                         2,330,000



  Neutralization                               87,000



  Silica Removal                              261,000



  Electrodialysis-Reverse Omosis            3,115,550



      Total                                $6,949,550



Operating Cost;



  Ammonia Removal                          $  300,000



  Carbon Adsorption                         2,358,000



  Neutralization                            1,900,000



  Silica Removal                               87,400



  Electrodialysis-Reverse Osmosis             646,300





      Tฐtal                                $5,291,700 $/yr
                            40

-------
                            SECTION  6

                 STREAM 2.  TOTAL RECYCLE SYSTEM


Stream 2 contains 4.2 percent total dissolved solids.  It was
selected to study the electrodialysis-reverse osmosis total
recycle, as it represents the highest concentration that can
possibly be treated with such a system.

A.  PRETREATMENT STUDIES

The characterization of Stream 2 indicated that at least two
steps would be necessary before the waste water could be treated
by the electrodialysis-reverse osmosis process.  These steps are
aluminate removal and recovery  (neutralization) and organic
carbon removal.

Aluminate Removal-Neutralization

The neutralization of Stream 2 caused the precipitation of the
aluminate ion in the form of aluminum hydroxide.  Precipitation
starts at a pH of 11.2.  Analysis of the supernatant after
adjustment to pH 7.0 indicated less than 1 mg/1 of aluminun ion
present.  Therefore virtually all the aluminate is removed at pH
7.  The characterization indicated that Stream 2 had an average
alkalinity of 4950 mg/1 as  sodium hydroxide, a 0.124 N sodium
hydroxide solution.  Therefore 0.124 equivalents Of hydrochloric
acid per liter would be required for neutralization or 10.3
gallons of concentrated HC1  (12 N) per 1000 gallons of Stream 2.

The suspended solids formed during neutralization must be removed
and concentrated.  The concentrated aluminum hydroxide may then
be converted to aluminum chloride for reuse in the process or
sold as a coagulant.  The average suspended solids formed during
neutralization of Stream 2  is 1500 mg/1.

Gravity settling tests were run on the sludge to determine the
design parameters for a gravity settler.  Varying concentrations
of sludge were placed in one liter graduated cylinders.  A 4 rph
stirrer was placed in each cylinder and the interface level was
recorded at various time intervals.  The sludge settles very
poorly.  Addition of varying concentrations of Purifloc A-23
flocculant to the sludge in a standard jar settling test produced
no improvement in settling rate or decrease in sludge volume.
The range of flocculant concentration was from 0.3 to 25 mg/1 of
A-23.   The addition of up to 200 mg/1 of ferric chloride also had
no significant effect on the settling.

                                41

-------
Figure 18 is a plot of the settling velocity versus initial
solids concentration for the gravity settling tests.  Figure 19
is a plot of solids flux versus solids concentration.  From this
graph using the Yoshioka method as described by Vesilind  (1974),
the minimum solids flux can be determined for any desired under
flow concentration and the surface area for the settler can be
determined if the daily flow is known.  Table 10 presents the
design solids flux and settler surface area for various underflow
sludge concentrations.  The design flow was 1 MGD, the initial
solids concentration was 1500 mg/1.  This information will be
used to design a gravity settler for Stream 2.

A comparison was made of various granular activated carbons to
determine which has the greatest capacity for the organics
(Mainly benzene) in Stream 2.  This was done by adding 0.05 grams
of each carbon and mixing with 250 mg/1 of neutralized Stream 2.
The carbon with the greatest capacity was selected for further
evaluation.  The results are shown in Table 11.  The best carbon
was Witco 517.

An adsorption isotherm was run at 25ฐC with Witco 517.  A syn-
thetic Stream 2 sample was used.  Figure 20 presents the isotherm
obtained.  Equilibrium capacity was determined to be 160 mg of
TOC per gram of carbon.

A carbon bed test was run using actual Stream 2 waste water.  A
1-in. I.D. glass column was filled with 175 grams of prewashed
Witco 517 to a height of 2.3 feet.  The bed volume was 690 mis.
The feed rate to the column was 42 mls/min or 2.0 gpm/ft2.
Stream 2 waste water was neutralized, settled and the decant was
fed to the column.  The effluent from the column was analyzed
periodically for benzene and TOC.  The benzene concentration of
the feed was below average for the waste because benzene was lost
to the air during the neutralization and settling step.  Figure
21 is a plot of the breakthrough curve.  Breakthrough was achieved
at 330 hours of operation and exhaustion took place at 370 hours.
The average feed benzene concentration was 29 mg/1.  The break-
through capacity was 134 mg benzene per gram of carbon.  The
exhaustion capacity was 147 mg benzene per gram carbon which is
higher than the isotherm capacity of approximately 100 mg of
benzene per gram of carbon.

The average TOC of the feed during the test was 60 mg/1 and the
average TOC of the effluent was 18 mg/1.  There is apparently a
small amount of TOC which will not adsorb on the carbon.  A carbon
bed will be designed to treat Stream 2 before electrodialysis-
reverse osmosis treatment using this data.  Regeneration of the
carbon will be done thermally.
Vesilind, P. A.  Treatment and Disposal of Wastewater Sludge,
Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan, 1974


                               42

-------
  •P
  m
  -P
  -H
  o
  o
  t-i
  (U
  On
  C
  •H
  H
  -P
  -P
  Q)
  CO
              I
       I
I
I
I
           1000   2000    3000   4000    5000   6000

               Solids  Concentration, mg/1
Figure  18.
GRAVITY  SETTLING - ALUMINUM HYDROXIDE
SLUDGE - STREAM 2
                        43

-------
 "O
 en
 X
 O
 Cfi
      .1
      J
                           I
I
           1000   2000    3000    4000   5000

                Solids Concentration,  mg/1
             6000
7000
Figure 19.   GRAVITY SETTLING - ALUMINUM SOLIDS FLUX  -
             STREAM 2
                            44

-------
Table 10.  GRAVITY SETTLING DESIGN DATA



Initial Solids Concentration. 1500 mg/1
Overflow
Concentration
of Underflow,
mg/1
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
Concentration
Solids Flux,
Ib./ft. Vday
2.87
2.45
2.25
2.06
2.00
1.95
1.89
1.84
... 10
Area,
ft.2
4,359
5,106
5,560
6,073
6,255
6,415
6,619
6,800
LlXjlA
mg/1
Rate
Overflow,
mgd
0.498
0.373
0.299
0.249
0.213
0.186
0.166
0.149
                  45

-------
Table 11.  COMPARISON OF ACTIVATED CARBONS
           REMOVAL OF ORGAN1CS FROM STREAM 2

C0 = 50 mg/1 TOC (Total Organic Carbon)
0.05 gms carbon/250 mis waste water


                              Adsorptivity,
	Carbon	           mg TOC/g C

Witco 517                          84

Calgon FS-300                      76

Barneby Chaney - PC                 72

Westvaco - WV-G                    67

Westvaco WV-W                      61

Hydrodarco HD-3000                 55

Calgon FS-400                      42
                     46

-------
   400
   2,00



u

c  100-
&
    8.0-

    60



    40
U
o
    20-
    10
                    Witco-517
                    Synthetic Stream #2
                    25ฐC
                          \	I
                     4   6   8  10     20

                      Residual TOG,  mg/1
                                                        I
                                                        I   C0 = 92 rag/1

                                                       I jn
                                              40  60  80 100
    Figure 20.  CARBON ADSORPTION ISOTHERM - STREAM 2

-------
CO
                 40r-
                 30
              c  20
              Q)
              N
              0)
                10
                 250
                                                           I
500
750       1000
     Bed Volumes
1250
                                         1
1500
                           Figure  21.   CARBON COLUMN TEST - STREAM 2

-------
B.  ELECTRODIALYSIS-REVERSE OSMOSIS  STUDIES  -  STREAM  2

The electrodialysis unit  used  for  these  experiments was  an Asahi
Chemical Ind. Model SV-4  laboratory  unit.  This  unit  is  larger
than the SV-5 used to  study Stream 1.  it  has  10 cell pairs with
an active membrane area of 2 dm2 as  compared to  5 cell pairs of
0.75 dmz active membrane  area  for  the  SV-5 unit.

The reverse osmosis equipment  was  the  same as  used in the evalua-
tion of Stream 1 treatment.  The SV-4'unit was more closely
matched to the reverse osmosis unit  in capacity  than  the SV-5
unit used previously.

Because of the high concentration  of NaCl  in Stream 2, 4.5 per-
cent NaCl, the units were first tested with  a  standard 4.5 percent
NaCl solution to define the operating  parameters.  The E.D. unit
operated at 0.58 volts/cell pair,  and  a  current  density of 2.5
amps/dm2 producing a dilution  stream of  4900 mg/1 NaCl and a brine
stream of 14.3 percent NaCl.   The  current  efficiency  was 87.4 per-
cent and the limiting  current  density  was  3.65 amps/dm2.  The RO
unit was operated with a  4000  mg/1 NaCl  brine  at 97.4 percent
rejection and 50 percent  product recovery.   The  NaCl  concentration
in the reject was 7400 mg/1 and the  product  concentration was
240 mg/1.

Continuous ED-RO Operation on  Pretreated Stream  2

The electrodialysis unit  was operated  to produce a dilution stream
of 9000 mg/1 NaCl, which  was fed continuously  to the  RO unit.
The reject from the RO unit was 1.5  percent  NaCl and  a product
water of less than 500 mg/1 NaCl was obtained.  This  run was
continued except that  the dilution concentration was  lowered to
approximately 6000 mg/1.  The  same feed  was  used.  The experi-
ment was allowed to run for 120 hours.   The  results are given in
Table 12.  Figure 21 is a plot of  the  critical operating para-
meters.  Table 13 presents a detailed  analysis of the various
streams.

The operation of the units for 120 hours was uneventful.  The
current density was close to the 2.05  amp/dm2  obtained during
the standard run.  The current efficiency  continued to decrease
throughout the run due to a steady decrease  in brine  production
possibly from leakage. Inspection of  the  membrane showed no
fouling or scaling.  The  calcium and magnesium ion present in the
feed were transported  to  the brine along with  the sulfate ion.
Only traces of these ions showed up  in the reverse osmosis product.

C.  TOTAL RECYCLE SYSTEM  DESIGN AND  COST ESTIMATE

A flow diagram of the  total recycle  system of  Stream  2 is given
in Figure 23, which includes a material  balance  and flow rates
of each stream.  Centrifugation of the underflow of the  gravity


                                49

-------
Table 12.   CONTINUOUS ELECTRODIALYSIS-REVERSE OSMOSIS
           RUN PRETREATED, STREAM-2   LOW DILUTION
           CONCENTRATION RESULTS
  Electrodialysis Unit;

    Cell pairs

    Feed NaCl

    Feed Rate

    Dilution Recycle Rate

    Dilution Temperature

    Brine Production Rate

    Dilution NaCl

    Brine NaCl

    Current Density

    Volts/cell pair

  Reverse Osmosis Unit;

    Reverse Osmosis Pressure

    Feed Flow

    Feed pH

    Product Flow

    Reject Flow

    Reject NaCl

    Product NaCl

    Recovery

    NaCl Rejection
10

41,800 mg/1

29 ml/min.

8.0 1/min.

24.0ฐC

7.3 ml/min.

4,700 mg/1

146,000 mg/1

1.95 amp/dm2

0.52



195 psi

42 ml/min.

6.0

23 ml/min.

19 ml/min.

7,850 mg/1

140 mg/1

55%

97.9%
                          50

-------
                                                     90
       0.70
Volts/cp
     A
       0.60
       0.50
     o
Amp/dm2
20
                       40
 60
Hours
80
100
                                                     80  %
                                                     Current
                                                      Eff.
                                                     70
                                   60
                                   50  %
                                    Recovery
                                     R.O.
                                   product
                                                    30
120
       Figure 22.  CONTINUOUS E.D.-R.O. PRETREATED
                   STREAM 2
                              51

-------
Ul
K)
                 Table  13.   CONTINUOUS  ELECTRODIALYSIS-REVERSE OSMOSIS RUN WITH
                            PRETREATED  STREAM 2
                                  Average Analysis  of  Streams
                                                                    Reverse Osmosis
Parameter Feed Dilution
NaCl, mg/1 41,800 4,700
Mg++, mg/1 0.2
Ca++, mg/1 8.5
Fe++, mg/1 0.07
S0if~~, mg/1 90
TOC, mg/1 12
Silica, mg/1 2 —
Brine Product Reject
146,000 140 7,850
<2 0.2
96 0.5
0.6 0.05
317 <1
4 1 —
— — —

-------
 Stream 2,
Waste Wat
                              HCl
            Neutralization

             Centrifuge
                                                                    Electrodialysis
                                                                  Reverse Osmosis
                                                        Product
                                                        Water
                                         Carbon
                                       Adsorption

NaOH
HCl
A1+3
TOC
NaCl
H20
Total
SpGr
Flow
Units
Ib/hr
Ib/hr
Ib/hr
Ib/hr
Ib/hr
Ib/hr
lb/hi-
1.02
mgd
ฉ CD CD (4) (D (D ฉ (D CD ฉ ฉ ฉ
1,720
—
240
30
14,564
332,494
347,500
1.19
1.0
x— „
1,324
—
—
—
2,256
3,580
1.02
.0103
_ ,_
—
240
9
14,454
100,425
105,126
—
.030
ILI-.-,-
	
0
29
13,734
309,638
323,401
1.02
0.934
_—
—
—
6
13,734
309,638
323,378
1.02
0.934
ซ.*_
—
—
8
3,340
75,313
788,444
1.02
0.227
-™
—
—
—
1,700
164,922
166,622
1.01
0.480
_.„-
—
—
—
1,790
380,210
382,000
1.01
1.10
__.
—
—
6
13,730
94,270
108,000
1.10
0.31
— m
—
—
—
30
215,348
215,378
1.0
0.62
mL _ J
—
—
—
15,520
474,480
490,000
1.02
1.410
__ ___L
—
240
1
1,114
25,112
26,282
—
.076
                   Figure
23.  TOTAL  RECYCLE  SYSTEM -  STREAM  2

-------
settler is included as a concentration step and was designed based
on limited data obtained in the laboratory testing.

A summary of the estimated capital and operating costs of the
total system is given in Table 14.

Details of the design and cost estimates of the unit operations
required for the total recycle of Stream 2 are given in Appendix
E.
                               54

-------
 Table  14.  COST FOR TOTAL RECYCLE SYSTEM - STREAM 2
                                        1975 Dollars
Capital Cost:
  Neutralization and Solids Removal      $  250,000
  Carbon Adsorption                       1,017,000
  Electrodialysis-Reverse Osmosis         7,903,000
      Total                              $9,170,000
Operating Costs;
  Neutralization and Solids Removal         123,700
  Carbon Adsorption                         407,000
  Electrodialysis-Reverse Osmosis         1,763,000
      Total                              $2,293,700 $/yr.
                            55

-------
                            SECTION 7

      ECONOMIC EVALUATION — BRINE AND POTABLE WATER REUSE
In order to make a cost comparison of the ED-RO system, calcul-
ations have been made on four different feed water salt concen-
trations.  The study was further expanded by setting permeate
concentrations at 500 and 1500 mg/1 NaCl to demonstrate the
effect of product water quality on process economics.  The cost
does not include pretreatment, as this varies with different
types of streams.  Feed flow is 1 MGD.

Figure 24 shows the relationship of operating cost (including
capital amortization) per 1000 gallons of Stream 1 waste water
treated versus feed brine normality.  The two product water
concentration restrictions are shown, as well as a theoretical
curve based on the higher current density achievable with an
ideal NaCl feed.

Making a change in perspective from a waste treatment plant to
one which produces process water, a similar cost analysis was
made.  Figure 25 is a plot of cost (in dollars)  per 1000 gallons
of permeate versus feed brine normality.  There is a marked
difference at higher normalities between the Stream 1 waste and
the ideal NaCl brine.

To examine a salt production plant, another plot,  Figure 26,
was made of cost per 1000 pounds of NaCl (produced as a 13-14
percent brine) versus feed brine normality.  The three Stream 1
plots are combined in Figure 27.
                               56

-------
01
G
O
rH
    25
O   20

o
o
o
-M
(0
    15

-------
o
o
o
   65^
   60
   55
n  50

o
rH
•H  45

u

   40
    35
   30
 Q)
 g  25
 M
 Q)
 CM
   20
   15


   10


    5
            •  - Permeate,  500 mg/1

            A- Permeate,  1500 mg/1
                            Ideal NaCl
                             Solution

                              1    I    I
     0   .2   .4   .6  .8   1.0  1.2  1.4  1.6  1.8 2.0

           Normality of Feed Brine  (NaCl)
Figure 25.  COST  OF  ED-RO OF BRINE SOLUTIONS -
            PRODUCTION OF PROCESS WATER
                       58

-------
w
tJ
G
o
o
o
(U
M

-------
m
o
Q
65

60


55


50

45


-40


35


30

25


20


15


10

  5

  0
           • - 500 mg/1 NaCl Permeate

           A- 1500 mg/1 NaCl Permeate
           $/1000 Ibs. NaCl
             Recovered
                                        $/1000 Gal.
                                        Permeate
                           $/1000  Gal.
                          Water  Treated
                 _L
I
I
I
I
I
_L
I
       .2  .4  .6   .8   1.0  1.2 1.4  1.6  1.8  2.0

         Normality of Feed Brine  (NaCl)
Figure 27.  COST OF ED-RO  TREATMENT OF STREAM-1
                       60

-------
                          BIBLIOGRAPHY


APHA, AWWA and WPCF, 1971  Standard Methods for the Examination
     of Water and Wastewater, Thirteenth Ed., American Public
     Health Association, Washington, DC.

Battelle-Northwest, "Wastewater Ammonia Removal by Ion Exchange,"
     EPA-Water Pollution Control Series, No. 17010 ECZ 02/71, 1971,

Bell, G. R., J. C. Leineweber, and J. C. Yang, Office of_ Saline
     Water Research &^ Development Progress Report No. 286, 1968.

Bober, T. W. and T. J. Dagon, Journal WPCF, 47(8), 2114-2129, 1975.

Calm, R. P. and N. N. Li, "Separation of Phenol from Waste Water
     by the Liquid Membrane Technique", Separation Science, 9(6),
     505-519.

EPA, 1971.  Methods for Chemical Analysis gjf Water and Wastes
     1971, Environmental Protection Agency, Water Quality Office,
     Analytical Quality Control Laboratory, Cincinnati, Ohio.

Hewes, C. G. and R. R. Davison, AIChE Journal, 17(1), 141-147,
     1971.

Hochhauser, A. M., "Concentrating Chromium with Luquid Surfactant
     Membranes," Ph.D. dissertation, University Microfilms No.
     74-20,365, Ann Arbor, Michigan, 1974.

Li, N. N. and A. L. Shrier, "Liquid Membrane Water Treating,"
     Recent Developments in Separation Science, Vol. 1, CRC
     Press, 1972.

McLaren, J. R. and G. J. Farquhar, "Factors Affecting Ammonia
     Removal by Clinoptilolite," Journal of the Environmental
     Engineering Division, 429, August 1973.

Mercer, B. W. L. L. Ames, W. J. Touhill, W. J. VanSlyke, and
     R. B. Dean, "Ammonia Removal from Secondary Effluents by
     Selective Ion Exchange," Journal WPCF, 42(2), part 2, 1970.

Schiffer, D. K., A. Hochhauser, D. F. Evans, and E. L. Cussler,^
     "Concentrating Solutes with Membranes Containing Carriers,"
     Nature, Vol. 250, August 9, 1974.
                                61

-------
Singer, P. C. and W. B. Zilli, Water Research, 9, 127-134, 1975,

Vesilind, P. A., Treatment and Disposal of Wastewater Sludge,
     Ann Arbor Science Publications, Inc., Ann Arbor, Mich.,
     1974.

Wang, L. K-P, D. W. Goupil, and M. S. Wang, Purdue University
     Engineering Extension Series, Vol. 141, Part 1,  569-578,
     1972.
                               62

-------
Multiply
By
                                                                 To Obtain
   Barrels  (bbls.)                      158.98
   Feet  (ft.)                              .3048
   Feet  per  hour  (ft./hr.)                0.508
   Gallons  (gal.)                         3.785
   Gallons per  day  (gpd)                  3.785
   Gallons per  minute (gpm)               3.785
   Gallons per  minute per               40.74
      square  foot  (gpm/ft.2)
u>  Inches (in.)                           2.54
   Pounds (Ib.)                           0.4536
   Pounds per cubic foot (lb./ft.3)       16.02
   Pounds per day (Ib./day)               0.4536
   Pounds per hour  (Ib./hr.)              0.4536
   Pounds per square foot                4.882
      per day (lb./ft.2)
   Pounds force per square inch          0.07031
      (psia)
    Standard cubic feet (scf)             26.79
                                      Liters  (1)
                                      Meters  (m)
                                      Centimeters per minute (cm/min.)
                                      Liters  (1)
                                      Liters  per  day (I/day)
                                      Liters  per  minute (1/min.)
                                      Liters  per  minute per square
                                        meter (1/min./m2)
                                      Centimeters (cm)
                                      Kilograms  (kg)
                                      Kilograms per  cubic meter (kg/m3)
                                      Kilograms per  day (kg/day)
                                      Kilograms per  hour (kg/hr.)
                                      Kilograms per  square  meter
                                        per day  (kg/m  /day)
                                      Kilograms force  per square
                                        centimeter  (kg/cm2)
                                      Liters  at STP  (1)
                                             o
                                             O
                                            en
                                            H
                                            O
                                            M

                                               >

-------
                           Appendix B
                       ANALYTICAL METHODS


Standard analytical procedures were used for the routine analysis
of pH, total alkalinity, acidity, total dissolved solids (TDS),
suspended solids, chloride ion, organic nitrogen (Kjeldahl
nitrogen), ammonia nitrogen, oil and grease and APHA color.
Sulfate ion in high concentration was analyzed using the benzidine-
dihydrochloride method.  In low concentrations, sulfate ion was
determined by using an ion chromatograph.  Trace metal ions were
determined, using emission spectroscopy.

In the case of Stream 1, a method was required for the analysis
of sodium hydroxide exclusive of the amines present.  A non-
aqueous titration method was developed for this analysis.

The analysis of the organic constituents of the petrochemical
waste waters was more difficult to perform and to interpret.
The characterization of the organic content of the waste waters
required the use of gas chromatographic methods to identify and
quantify the organic compounds present.  A gas chromatographic
method was developed for the analysis of the organics in Stream 1
and Stream 2.  Conditions of GC analysis of Stream 2 are given in
Table B-l.

Other methods were used to monitor the organic content of the
waste waters.  Total organic carbon (TOC), inorganic carbon (1C),
and total carbon (TC) were determined using a Beckman Model 915
carbon analyzer.  The method entails the injection of a micro
sample into a catalytic combustion tube at 950ฐC.  The carbon-
aqueous material is oxidized to carbon dioxide in a carrier stream
of air.  The amount of CO2 produced is directly proportional to
the amount of carbonaceous material in the injected sample and is
measured by a sensitized infrared analyzer.  This determines the
total carbon.  A second injection of the sample is made into a
low temperature oven tube containing phosphoric acid and the in-
organic carbon is released and purged by an air stream to the
infrared detector.   This determines the inorganic carbon.  The
difference between the two values is the total organic carbon.

Total oxygen demand  (TOD)  was also used to determine the organic
content of the waste waters.  An Ionics Model 225 Total Oxygen
Demand analyzer was used.   The determination is made by injecting
a micro sample of the waste water in a combustion tube containing
a platinum  catalyst at 950ฐC.  The catalyst tube is continually
                                64

-------
purged with a flow of nitrogen contaminated with a low concentra-
tion of oxygen.  The oxidation of  the carbon in the injected
sample causes a decrease  in the oxygen level in the nitrogen
stream.  This will register as a change  in electrical output from
a silver-lead fuel through which the nitrogen stream flows.  This
response is proportional  to amount of oxidizable organic carbon
present in the sample.
                                 65

-------
Table B-l.  GAS CHROMATOGRAPHIC ANALYSIS CONDITIONS -
            STREAM 2
Gas Chromatograph

Detector

Injector Temperature

Detector Temperature

Columns

Column Packing


Oven Temperature

Carrier Gas

Carrier Gas Flow Rate

Air Flow Rate

Hydrogen Gas Flow Rate

Sample Injection


Calculation



Retention Times
Hewlitt-Packard 5711A

Dual FID

110ฐC

300ฐC

12 ft. x 1/8, SS

10% Carbowax 20M on Chromosorb
W-HP 80-100 mesh    ,

Isothermal 110ฐC

Helium

30 cc/min. at 60 psi

240 cc/min. at 24 psi

30 cc/min. at 15 psi

3 microliters of 1:1 CS2
extract of the sample

Integration and calculation
performed by HP 3380A
Integrator-Recorder

Benzene - 1.75 min.
Ethylbenzene - 5.05 min.
                          66

-------
                           Appendix C
                         AMMONIA REMOVAL
A.   Literature Review
Several processes are  presently used for ammonia  removal  from
waste waters.  The  concentration of ammonia is  an important  factor
in determining which process  to use.

For ammonia concentrations  of the order of several percent,  stream
stripping in a column  desorber with countercurrent flow is gen-
erally used.  A wet ammonia overhead is obtained  which can be re-
cycled directly or  condensed  to produce an aqueous solution.
Effluent waste water can be expected to contain about 1000 ppm
ammonia.  Secondary stripping is used to reduce the effluent
level to near 50 ppm ammonia.  The secondary stripper overhead
can be used as a steam source for the primary desorber.   This
degree of ammonia removal may be sufficient in  some cases, but
frequently it is not.

To remove ammonia from waste  waters containing  concentrations in
the 100 ppm or less range,  four processes have  been used.  These
are air stripping,  nitrification-denitrification,  selective  ion
exchange, and breakpoint chlorination.

Air stripping is generally  accomplished in a countercurrent  or
crosscurrent flow tower very  similar to a water cooling tower.
Large fans draw air through waste water that trickles through
slats of wood or plastic, desorbing the ammonia.   The pH  of  the
waste water is adjusted to  a  value near eleven, usually with
lime.  This converts all ammonium ions to dissolved ammonia.  The
air requirement is  of  the order of 300 cu.ft. per gallon  of  liquid.
The air stripping process is  simple and reliable,  and removals
range from 60 to 95 percent.   The removal efficiency at low  tem-
peratures is poor because ammonia solubility is inversely related
to the temperature. If lime  is used to raise the pH, scaling
may be a problem.   The discharge of ammonia into  the atmosphere
can be prevented, if necessary, by scrubbing the  ammonia-rich gas.

Nitrification-denitrification is accomplished by  biological  ox-
idation of the ammonia to nitrite or nitrate, followed by bio-
logical reduction to nitrogen gas.  This process  can be carried
out in various biological reactor systems, and  as in all  micro-
biological reactors, the major operational difficulty encountered
is insolubility.  The  nitrification step is aerobic, while the
                                 67

-------
denitrification step is anaerobic.  A substrate, such as methanol,
must be added to the waste for denitrification to occur.  Ammonia
removal efficiencies range from 80 to 98 percent.

In the breakpoint chlorination process, the ammonia is oxidized
to various chloramines, and then to nitrogen gas.  About 10 parts
of chlorine are required for every part of ammonia-nitrogen that
is reacted.  Dechlorination may follow chlorination to remove
residual chlorine, or carbon adsorption to remove unoxidize chlor-
amines.  For high concentrations of ammonia, breakpoint chlorina-
tion is not economical because of chlorine costs.  However,
removal efficiencies of 90 to 100 percent have been reported,
and this process is often used following one of the previously
mentioned processes to reduce the residual ammonia level to near
zero.

B.   Ammonia Removal Tests

     1.   Liquid membrane process

A process by which both the ammonia and the amines could be re-
moved and recovered simultaneously would be of great value in
pretreating this waste water.  One process that offers this
possibility is the liquid surfactant membrane process described
by Li and Shrier  (1972).

A schematic diagram of the conceptual Liquid Membrane Water
Treatment Process is shown in Figure C-l.

Liquid surfactant membranes are formed by first making an emulsion
of two immiscible phases such as a water and an oil.  A surfactant
is added to stabilize the emulsion.  The emulsion is then dis-
persed into a third phase.  The liquid membrane phase refers to
the phase between the encapsulated phase in the emulsion and the
third phase.  The encapsulated phase and the third phase are
usually miscible but are immiscible with the membrane phase.  The
liquid membranes can be modified for specific applications and
the large surface area of the membrane allows rapid permeation
and equilibration during mixing.  Facilitated transport can be
used to enhance this mass transfer.  This involves encapsulation
of a reagent within the membrane that reacts with the permeating
specie and forms a specie that will not permeate.  After suf-
ficient contact time between the emulsion and the third phase,
mixing is stopped and the phases are allowed to separate.  The
emulsion can then be decanted off the top and de-emulsified to
release the permeated specie.
Li, N. N. and A. L. Shrier.  "Liquid Membrane Water Treating",
Recent Developments in Separation Science, Vol. 1, CRC Press,1972.

Cahn, R. P. and N. N. Li.  "Separation of Phenol from Waste Water
by the Liquid Membrane Technique," Separation Science, 9(6),
505-519, 1974.

                               68

-------
Contaminated
 Water Feed
                1200 rpm
       Reagent
       Makeup
      Emulsif ier
                           Chemicals
                             Makeup
100 -  IEMULSIFICAT ION]
200 rpm

    s*J  LM Emulsion,
                     Fresh
                     Emulsion

                       Recycle
                       Emulsion
Mixer
              Settler
                                 Recovered Liquid
                                 Membrane Components
                                (Solvent, Surfactants,
                                 etc.)
   I CONTACTING | (may be multistage)
                                         Demulsif ier
                                         SPENT
                                         EMULSION
                                         TREATMENT
                                                      Spent Reagent
                                                      to Recovery
                                                     'or Disposal
                                                                    •*•Effluent Water
                                                                      to Further

                                                                      Discharge
Figure C-l. SCHEMATIC DIAGRAM OF CONCEPTUAL  LMWT  PROCESS

-------
Liquid membranes have been demonstrated  (Li and Shrier, 1972;
Cahn and Li, 1974) to:

     1.   Remove weak acids such as phenol, H2S, HCN, and
          acetic acid from water.

     2.   Remove weak bases such as ammonia and amines from
          water.

     3.   Purify hydrocarbons and separate toluene from
          heptane and benzene from hexane.

     4.   Remove suspended solids from waste water.

     5.   Remove selectively chromium, cupric, mercuric,
          silver, sulfide, nitrate, and phosphate ions from
          water  (Schiffer, et al, 1974; Hochhauser, 1974).

A synthetic waste water containing 160 mg/1 ammonia and 1000 mg/1
ethylene diamine in pH 12 caustic was treated with liquid surfact-
ant membranes.  The membrane was produced by adding at 10 ml/min,
approximately 80 mis of 0.1 N hydrochloric acid to approximately
160 mis of S-100-N oil containing 2 percent by weight SPANฎ 80.
The mixing was done at 1300 rpm using a Fisher Stedi-Speed Model
12 stirrer.  The mixing continued for 10 minutes after the last
of the acid was added.  A thick milky white emulsion was formed.
The resulting 250 mis of emulsion was mixed with 250 mis of
synthetic waste water at 200 rpm.  Samples were taken of the
aqueous phase at timed intervals.  These samples were analyzed
for ammonia and ethylene diamine.

Table C-l summarizes the experimental conditions and results.
Figure C-2 presents a plot of the Ct/Co versus the time of contact
with the liquid membrane.

The results indicated that the ammonia was removed within 5
minutes and the maximum ethylene diamine removal occurred in 15
minutes.  The ethylene diamine removal stopped short of complete
removal because there was not enough acid to react with the
remaining ethylene diamine.  Membrane breakage and subsequent
neutralization of the acid may account for the fact that only 94
percent of the acid was reacted when the permeation stopped.  The
concentration factor for ammonia was approximately 3 because the
ammonia that was contained in the 250 mis was transferred in the
80 mis of acid within the membrane.
Schiffer, D. K., A. Hochhauser, E. F. Evans, and E. L. Cussler,
"Concentrating Solutes with Membranes Containing Carriers,"
Nature, Vol. 250, August 9, 1974.

Hochhauser, A. M.  "Concentrating Chromium with Liquid Surfactant
Membranes," Ph.D. dissertation, University Microfilms No. 74-20,
365, Ann Arbor, Michigan, 1974.

                                70

-------
Table C-l.  LIQUID MEMBRANE TREATMENT OF SYNTHETIC
           STREAM 1 WASTE WATER
 Volume of liquid membrane taken

 Volume of waste water taken

 Total time of experiment

 Initial ammonia concentration

 Final ammonia concentration

 Ammonia removed

 Initial ethylene diamine
   concentration

 Final ethylene diamine
   concentration

 Ethylene diamine removed

 Ammonia removed

 Ethylene diamine removed

 Total removed

 Total meq of HCl available

 HC1 usage

 Concentration factor-ammonia
250 mis

250 mis

60 minutes

160 mg/1

0 mg/1

100%


1000 mg/1


380 mg/1

62%

2.35 meq

5.17 meq

7.52

8.0

94%

3
                         71

-------
            [C0 =160  mg/1  NH3  + 1000 mg/1 aminej
      10
20      30      40
   Time, minutes
60
Figure C-2.SYNTHETIC STREAM 1 - WASTE WATER
           LIQUID MEMBRANE TREATMENT
                     72

-------
Although the initial  experiment showed some  promise  for  this
technique, attempts to  increase the concentration factor by
encapsulating a higher  concentration of hydrochloric acid in  the
membrane were unsuccessful.   The liquid membrane  formed  was too
unstable.  Experiments  with  waste water which contained  a
higher sodium chloride  level than the synthetic also failed
because of membrane instability.  No experiments  were run to
determine the ease with which the exhaust liquid  membrane could
be de-emulsified  to recover  the ammonia and  amine hydrochlorides.

Further extensive work  will  be required to develop a liquid
surfactant membrane which remains stable in  the presence of
higher HC1 concentrations and higher brine waste  waters.

      2.   Ion exchange

One other method  of ammonia  removal from waste water was tried.
This  was the use  of clinoptilolite as an ion exchanger to remove
ammonium ions from the  water.  This process  has been described
variously in the  literature  in conjunction with removal  of ammonia
from  municipal waters (Mercer, et al, 1970;  McLaren  and  Farquhar,
1973; Battelle-Northwest, 1971).  A supply of clinoptilolite  was
obtained and prepared (McLaren and Fraquhar, 1973).   A small
column was filled with  20 grams of resin. A flow rate of 1 gpm/
sq.ft. was set.   The  feed solution neutralized Stream 1  waste
water.  Figure C-3 shows the results of this experiment.   The
exchange capacity was found  to be 0.28 meq.  of ammonia-nitrogen/
gm. of resin, much less than the capacity (1.6-1.8 meq./g.) given
by the supplier  (McLaren and Farquhar, 1973). A  value of 0.95
meq./g. was obtained  for a water containing  70 mg/1. ammonia-
nitrogen.

Apparently at the higher ion concentration of the waste  water,
the cations compete with the ammonia for exchange sites  and the
capacity of the  resin for the ammonia ion is much too low to  be
used  to remove ammonia  from  Stream 1 waste water.
 Mercer,  B.  W.  et al.   "Ammonia Removal from Secondary Effluents
 by Selective Ion Exchange," Journal WPCF, 42(2)  Part 2,  1970.

 McLaren,  J. R. and G.  J.  Farquhar.  "Factors Affecting Ammonia
 Removal  by  Clinoptilolite," Journal of Environmental Engineering.
 429,  August, 1973.

 Battelle-Northwest, "Wastewater Ammonia Removal by Ion Exchange,"
 EPA-Water Pollution Control Series No. 17010 ECZ 02/71,  1971.
                                 73

-------
  200
tn
e
  100
                      C0 =  218  mg/1 NH3-Nitrogen
                     I
I
I
I
I
I
I
I
I
I
         100   200   300  400   500   600   700   800


           5    10    15   20    throughput     40
                         900  1000  1100  1200  1300  1400  ml.


                               50       Bed volume
                 Figure C-3. EXCHANGE OF NHU + ON CLINOPTILOLITE FROM

                            STREAM  1 WASTE WATER

-------
                           Appendix D
 TOTAL RECYCLE SYSTEM DESIGN AND COST ESTIMATES FOR STREAM 1


The process design of the unit operations required for the total
recycle of Stream 1 are based on a flow rate of 1 MGD wastewater.
A cost estimate of the total system is made based on 1975 costs.

A.   Pretreatment Design

     1.   Carbon adsorption

The bed depth-service time, for 95 percent TOC removal is plotted
in Figure D-l.  One column test was run at the lower rate of 1
gpm/ft2 and the rest were made at  2 gpm/ft2.  The agreement
between the values obtained from the actual 2 gpm/ft2 run and
those calculated from the 1 gpm/ft2 converted to 2 gpm/ft2 was
excellent.  From the results of the electrodialysis-reverse
osmosis tests, it was evident that the TOC removal must be
maximized.  This is why  95 percent removal was selected.  The
design bed versus service time relationship for 95 percent
removal was:

      t =  0.38(x) -  1.25
          where:
               t =  service  time, hours
               x =  bed  depth,  feet
 The  carbon treatment  unit  is  designed  for  a linear flow rate of
 4 gpm/ft2, to handle  1  MGD  of wastewater,  of  1650 mg/1 feed TOC,
 producing a  TOC  concentration of  80 mg/1.  The activated carbon
 is Westvaco  WV-G,  12  x  40 mesh of  25  Ibs/cu.ft. density.

 The  cross-sectional area requirement  is:
      500,000 gpd/(4 gpm/ft2)(1440  min/day) =  87  ft2
      (The area for  a 1,000,000 gpd unit was  too  great,  so
      the  process unit will  be divided into two parallel
      trains.   Each will treat 500,000 gpd.)

 The  carbon bed volume will be:
      87  ft2  x 20 ft.  (bed depth)  = 1740 ft
 The  adsorber volume will be (50% higher for backwash)
      1740 ft3 x 1.5 = 2610 ft3
 The  adsorber dimensions:
      30  ft x 10.5 ft I.D.
                                 75

-------
   14
   12
   10
en
S-)
2
o
0)
0)
o
-H

-------
The bed dimensions will be:
     20 ft x 10.5 ft  I.D.
The weight of the carbon per  bed will  be:
     1740 ft3 x 25 Ibs/ft3 =  43,500  Ibs
The bed service time  will be:
     t = 0.19(20) - 1.25 = 2.5  hours

Three columns in series will  operate for  approximately  8 hours
or one shift.  Each one will  need  to be regenerated once per
shift.  A fourth column will  be needed for  the  regeneration.  To
handle the 1 MGD of Stream 1, two  trains  of four  columns will be
needed.  Three columns will be  in  operation at  one time while
one column in each train is being  regenerated.

The acid requirement  for the  regeneration of the  carbon was
determined to be 1.8  bed volumes of  acid  per regeneration.
However, 0.3 bed volumes are  recycled  back  to the acid tank so
1.5 bed volumes is actually used.  Approximately  10 regenerations
will be required per  day per  train or  20  all together.  Therefore,
30 bed volumes of 30  x 13,000 gal. will be  required per day.
This is 390,000 gallons of 2  N  acid  or 65,000 GPD of concen-
trated acid  (12 N).

Figure D-2 shows a flow diagram and  material balance for one
train of the process.

The capital cost for  the carbon treatment includes adsorbers,
pumps, carbon, tanks  and associated  equipment.  These are shown
in Table D-l.  The operating  cost  includes  utilities, amor-
tization, labor, carbon replacement, and  acid costs.  These are
shown in Table D-2.

     2.   Neutralization

The acid requirement  for the  neutralization of  Stream 1 was
determined to be 22 gal. of concentrated  HC1 per  1000 gals, of
Stream 1.  After carbon treatment, the stream volume was 1,180,000
gal.  After neutralization, the stream rate will  be 1,206,000.
It will require 26,000 gal. of  concentrated acid  per day, or 9.5
Mgal/yr.  The capital and operating  costs for neutralization are
given in Table D-3.

     3.   Silica removal

The design and costs  are presented for silica removal.  A 1.2
MGD stream after neutralization must be treated with an alumina
column.  Table D-4 is the design bases.   Figure D-3 gives the
process design and material balance.   The process consists of
two trains of two columns each. Each  column is 16 ft.  long x
11.5 ft. diameter.  The bed volume is  1675  ft  and each contains
92,000 Ibs. of F-l alumina.   Before  regeneration, 190 bed
volumes are treated with 1.5  bed volumes  of caustic.  Table D-5
gives the cost estimate for this process.

                                77

-------
       0.5 mgd
to r~L

n 1 ฎ
u x 1
Feed
Tank

•••ซ••



" -*"*.



Ld
Mi v
(?) Tank
oo _______

NaCl 150 1,550 1,550
1
'
l_
















•••••n
^^^^H




0



ฉ
1



L









••muM









ฉ
\

\







i









(D
^v











ฉ ฉ ฉ
1,400
HC1 -- — -- 4,900
NaOH 1,500 1,500 1,500
TOC 300 330 15
700 4,200
500
1
280
H20 171,800 201,470 201,470 81,300
Organics
30
5
35,450 29,670 12,850 11,400 57,300
410
Total 173,750 205,350 205,035 86,200
Sp. Gr 1.01 1.01 1.01 1.03

Flow, gpd 500,000 590,000 590,000 234,000

•j __ 	
36,140 31,600 13,550 15,600 57,300
1.03
1.03
1.03 1.19 1.0
104,000 91,000 39,000 32,700 465,000
Figure D-2. TOTAL RECYCLE SYSTEM STREAM  1 - CARBON  TREATMENT
            PROCESS DESIGN AND MATERIAL  BALANCE

-------
B.   Electrodialysis-Reverse Osmosis Design

The basis for the design to handle the pretreated Stream 1 waste
is given in Table D-6.  The material balance is given with the
flow diagram presented in Figure D-4.

For the reverse osmosis unit, the capital cost is calculated as
0.65 dollars per gallon per day permeate and the operating cost
as 0.60 dollars per  1000 gallons of permeate.  For the electro-
dialysis unit, 2 presses 1800 cell pairs each are required.
The cost per press of $500,000 times a factor of 2.5 gives the
total plant cost.

A summary of the capital and operating costs of the reverse
osmosis-electrodialysis hybrid is given in Table D-7.
                                 79

-------
   Table D-l. 1-MGD STREAM 1 - ACTIVATED CARBON TREATMENT
              ESTIMATE OF DIRECT CAPITAL COST

                                                  Installed
                      Item                          Cost
Adsorption Columns (8) at $200,000 each
  30' x 10.51 dia. - Monel-Vertical Vessel        $1,600,000

Pumps (8)                                             15,000

Pumps - Acid Service  (2), Monel                       20,000

Carbon,  Initial Charge - Westvaco WV-G,
  12 x 40 Mesh, 348,000 Ibs. at $.50                 174,000

Feed Equilization and Storage Tank
  1/2 Day Volume  (10,100 bbl.)                        78,000

Acid Storage Tank (2,100 bbl.)
  8-Hour Capacity Rubber Lined Carbon Steel           54,000

General Services  (10%)                               190,000

   Direct Capital                                 $2,131,000

   Allocated Capital - Electric Power                 50,000

   Factory Expense (7% of Direct Capital)            149,000

          FIXED CAPITAL                           $2,330,000
                             80

-------
Table D-2.  1-MGD STREAM 1 - ACTIVATED CARBON TREATMENT -
           ESTIMATE OF OPERATING COST
         Item
Raw Materials
  HC1, 38%
  Carbon Replacement
Operating Labor
Maintenance

Power
River Water
Factory Expense

Taxes and Insurance

Depreciation
 Quantity

9 x 107 Ibs.
34,000 Ibs.
Cost per
  Unit
 $0.02
  0.50
        4 Man-yrs.
      6% of Process
    Equipment Capital
60 M Gal.
  0.025
    2.6% of Direct
    Capital
    1.1% of Direct
    Capital
    10% of Direct
    Capital
 Cost per
   Year
$1,800,000
    17,000
   100,000

   128,000
    20,000
     2,000

    55,000

    23,000

   213,000
$2,358,000
TOC Removed per year            =
Organics Recycled per year      =
Cost per Ib. Organics Recycled  =
Value of Organics at 50* per  lb.=
           2.45 x 106 Ibs.
           6.0 x 106  Ibs.
           $.39/lb.
           $3,000,000  $/yr.
                            81

-------
Table D-3.  1-MGD STREAM 1 - NEUTRALIZATION
           ESTIMATE OF CAPITAL AND OPERATING COSTS


              Item                        Cost
Capital Cost;

  Mixing Tank - Rubber Lined
    8-Hour Capacity, 400,000 gal.      $   54,000

  Pump - Acid Service, Monel               10,000

  Pump                                      2,000

  General Service (10%)                     6,000

     Direct Capital                    $   72,000

  Allocated Capital, power                 10,000

  Factory Expense (7%)                      5,000

         FIXED CAPITAL                 $   87,000

Operating Cost;
                                         Cost  $/yr.
  Raw Materials, HC1, 38%
    9.4 x 107 Ibs. at $0.02            $1,880,000

  Maintenance  (6%)                          4,000

  Power                                     5,000

  Factory Expense (2.6%)                    2,000

  Taxes and Insurance (1.1%)                1,000

  Depreciation  (10%/yr.)                    7,000

                                       $1,900,000  $/yr,
                          82

-------
    Table  D-4. SILICA REMOVAL - STREAM 1
Design Basis:

  Linear Flow Rate

  Design Flow

  Design Silica
   Concentration

  Design Product
   Concentration

  Capacity

  Bed Depth

  Alumina

  Density
-  4 gpm/ft.2

-  50 mg/1


-  <10 mg/1


   30 minutes

   20 mg Silica/g Alumina

-  16 ft.

   Alcon F-l

-  55 lbs./cu.  ft.
                     83

-------
Feed  (k Total Flow)
    (1)
      In  Sodium Hydroxide


1


2

^ฉ

•_ Trpat^d S-i-r^am 1
ฉ
               Spent Caustic
NaOH
Silica
H20
Total

Flow
          Units
          Ibs/hr
            gpd
  10.5
208,490
208,500

600,000
           70
   2.1
208,490  1639
208,492  1700

600,000  5000
   70
  8.4
1621.6
1700

5000
Figure D-3.  TOTAL RECYCLE SYSTEM - STREAM 1
            SILICA REMOVAL, DESIGN BASIS AND
            MATERIAL BALANCE
                           84

-------
   Table D-5.  ACTIVATED ALUMINA TREATMENT OF STREAM 1 -
              ESTIMATE OF CAPITAL AND OPERATING COSTS


  	Item                          Cost
Capital Cost:

  Four Adsorber Columns at $36,000 each
   (16' x 115' dia. - Carbon Steel)               $144,000

  Caustic Mix Tank
    1-Day Capacity - Carbon Steel                   9,000

  Two Pumps                                         4,000

  Alumina, 370,000 Ibs. at $0.15/lb.               56,000

  General Services  (10%)                           22,000

     Direct Capital                              $235,000

     Allocated Capital                             10,000

     Factory Expense  (7%)                          16,000

                                                 $261,000

Operating Cost;                                   $/yr.

   Raw Materials

     100% Caustic, 600,000 Ibs.                   $ 30,000

     Water,  1,825,000 gallons                          40ฐ


   Maintenance  (6% of  Capital)                      16,000

   Power                                             5'000

   Factory Expense (2.6%)                            7,000

   Taxes  and  Insurance (1.1%)                        3,000

   Depreciation  (10%/yr.)                           26'000

                                                 $ 87,400
                           85

-------
   Table D-6.  1-MGD STREAM 1 - ELECTRODIALYSIS-REVERSE
              OSMOSIS DESIGN
Basis:

  Design NaCl Feed Concentration

  Design Brine Product


  Design R.O. Product

  Feed Rate

  Volts/cell pair

  Current Density

  R.O. Product Recovery

  NaCl Rejection

  Number of SS-0 Press

  Cell Pairs

  Operation
-  14,000 mg/1  (0.24 N)

-  >140,000 mg/1
   Maximum  2.4 0 N

-  <500 mg/1 (0.009 N)

-  1.2 mgd

-  0.74

   1.6 amp/dm2

-  48%

-  95%

-  2

-  3131

-  330 day/yr.
                          86

-------
                               Brine
                               Recycle
 Neutralized
 Feed
        Reject
                                 Electro-dialysis
                       Dilution
                       Recycle
                   CD
                   TR
 Reverse
 Osmosis
AFeed

Permeate
                                                      Brine
                                Product
                                'Water
                                               Reject
                              Reverse Osmosis
Ib/hr    (&       (&      (&       (V      15;      (6)

NaCl     5,850    8,720   5,700    3,020      150    2,870

H20    418,000  712,000  34,000  678,000  384,000  294,000

Total  423,850  720,720  39,700  681,020  384,150  296,870
     Figure Q-4.  ELECTRODIALYSIS-REVERSE OSMOSIS OF
                 PRETREATED STREAM-1, FLOW DIAGRAM
                 AND MATERIAL BALANCE
                             87

-------
Table D-7. COST ESTIMATE SUMMARY - ELECTRODIALYSIS-
           REVERSE OSMOSIS TREATMENT OF STREAM 1
 Reverse Osmosis;

   Capital                         $  615,550

   Operating Cost                     187,500 $/yr.

 Electrodialysis;

   Capital                         $2,500,000

   Operating Cost

     Power                            133,530
     Membrane Replacement              87,000
     Total Labor                       38,250
     Miscellaneous                     50,000

   Depreciation                       250,000

      Total                        $  458,780 $/yr.

 ED-RO;

   Total Capital                   $3,115,550

   Total Operating Cost            $  646,280 $/yr.
                         88

-------
                            Appendix  E
   TOTAL RECYCLE SYSTEM DESIGN AND  COST  ESTIMATES  FOR STREAM  2


A total recycle system is designed  for the  treatment  of  1 MGD of
Stream 2 waste water.  The  pretreatment  steps  before  the electro-
dialysis-reverse osmosis hybrid  are the  neutralization,  solids
removal7 and carbon  adsorption.   A  cost  estimate of each step is
made based on 1975 costs.

A.  Pretreatment Design Bases

    1.  Neutralization and  solids removal

A flow diagram and a material balance is given in  Figure E-l.
The feed contains 680 mg/1  of Al   ion that results in 150 mg/1
of suspended solids, A1(OH)3  after  neutralization.  The  primary
settling tank underflow concentration of 500 mg/1  at  a flow of 0.3
MGD requires a surface area of  5600 ft  . The  sludge  is  thickened
in a centrifuge at 1100 G's acceleration, to a 2 percent compacted
sludge.  The volume  of the  sludge is  reduced to 75,000 GPD.

A capital and operating cost  of  this  pretreatment  step is given
in Table E-l.

    2.  Carbon adsorption

The design basis  and parameters  of  the carbon  adsorption system
for Stream 2 are  given  in  Table  E-2.   The flow diagram and mate-
rial balance are  shown  in  Figure E-2. A summary of the  capital
and operating costs  is  given  in  Table E-3.

B.  Electrodialysis-Reverse Osmosis Design

The design basis  of  the  ED-RO hybrid for recycle of fresh water
and concentrated  brine  from the  pretreated  Stream  2,  is  given in
Table  E-4.  A flow diagram and material  balance is shown in
Figure E-3.  •

For the reverse  osmosis  unit,  the capital is calcultated as 0.65
dollars per day  permeate  and the operating  cost as 0.6 dollars
per 1000 gallons  of  product water.   For  the electrodialysis unit,
6 presses,  1800  cell pairs each, are required.  The cost per  press
of  $500,000  times a  factor of 2.5 gives  the estimated capital cost.
Membrane replacement is  based on 7  years lifetime. A summary of
capital and operating costs of the  reverse  osmosis-electrodialysis
hybrid for Stream 2  recycle is given in  Table  E-5.
                                 89

-------
                     Waste Water
/ C3J
vo
o
         Al
         NaCl
         H2O
         TOC
         HC1
         Total
         Total

         NaOH
(D ,
^-^
ฉ

V Primary /
\Settler/
Clarified
Waste




HCl \ /

ฉ


Solids Disposal /g\ X*^"\ /g\
/->y Ponc^ ^ ^— ' 1 1 v^

Units Ci)
lb./hr. 240
lb./hr. 14,564
lb./hr. 332,494
lb./hr. 30
lb./hr.
lb./hr. 347,500
M gpd 1.0
Sp Gr 1.02
lb./hr. 1,720
V 1
\ 	 y Centrifuge
ฉ ฉ ฉ
0 240
10,394 4,454
2,256 234,325 100,425
21 9
1,324
3,580 244,740 105,1246
0.0103 0.7072 0.3031
1.19 1.02
1.0103

ฉ ฉ
0 240
3,340 1,114
75,313 25,112
8 1
—
78,844 26,282
0.227 0.076
1.02
— —

ฉ
0
13,734
309,638
29
—
323,584
0.934
1.02
—
                     Figure  E-l. NEUTRALIZATION AND SOLIDS REMOVAL - STREAM 2
                                  Flow Diagram and Material Balance

-------
 Table E-l.  NEUTRALIZATION AND SOLIDS REMOVAL - CAPITAL
            AND OPERATING COST ESTIMATES - STREAM 2

Capital Cost;

  Mixing Tank - rubber lined, 4 x 105 gallons   $ 54,000

  Pumps, Monel                                    16,000

  Gravity Thickener                               75,000

  Centrifuges                                     50,000

  G&A  (10%)                                       20,000

    Direct Capital                              $215,000

  Allocated Capital, power                        20,000

  Factory Expense  (7%)                            15,000

    FIXED CAPITAL                               $250,000

Operating Cost;                                   $/yr.

  Raw Materials
   HCl 37%, 3.4 x 106 Ib./yr.                   $ 68,000

  Maintenance  (6%)                                15,000

  Power                                           10,000

  Factory Expense  (2.6%)                           6,500

  Taxes and Insurance  (1.1%)                       2,700

  Depreciation  (10%)                              21,500

                                                $123,700 $/yr,
                               91

-------
  Table  E-2. CARBON ADSORPTION OF STREAM 2 - DESIGN BASIS
Feed Concentration TOC

Effluent TOC

Flow Rate

Columns - 4, 2 trains


Bed Volume

Weight Carbon (Witco 517)

Service Time

Capacity

Regeneration Capacity
-  90 mg/1

-  18 mg/1

   4.0 gpm/ft.2

-  20' x 18' diameter
   each

-  2,430 cu. ft.

-  60,750 Ib./bed

-  713 hours

-  13.5 Ib. TOC/lb. carbon

   8,200 Ib. carbon/day
                             92

-------
  Two Trains
* t
Loading
Cycle

A

B
Regeneration
Cycle
1
t Q
Units 1 2
TOG lb./nr. 74.5 3
NaCl lb./hr. 6,867 6,867
Water lb./hr. 754,819 154,819
Total lb./hr. 161,700 161,700
Figure  E-2. FLOW DIAGRAM AND MATERIAL BALANCE FOR
            CARBON ADSORPTION TREATMENT - STREAM 2
                         93

-------
  Table E-3.  CARBON ADSORPTION - CAPITAL AND OPERATING
             COST ESTIMATES - STREAM 2


Capital Cost;

  Adsorption Columns
    Four, 20' x 14' diameter                   $  224,000

  Initial Carbon
    244,000 Ibs. at 50<=/lb.                       122,000

  Thermal Regeneration Unit
    8,200 Ib./day capacity                        492,000

  General Services  (10%)                           84,000

      Direct Capital                           $  922,000

  Allocated Capital, power and steam               20,000

  Factory Expense  (7%)                             65,000

      FIXED CAPITAL                            $1,017,000

Operating Cost:                                   ... ,
_ฃ	2	                                    $/yr.

  Utilities                                    $   40,000

  Labor and O.K.                                   70,000

  Carbon Make-up (7%)                              100,000

  Maintenance (6%)                                  60,000

  Factory Expense  (2.6%)                           26,000

  Taxes and Insurance (1.1%)                       11,000

  Depreciation  (10%)                              100,000

                                               $  407,000
                            94

-------
  Table E-4. ELECTRODIALYSIS - REVERSE OSMOSIS
             TREATMENT - STREAM 2
Design Basis;

  Feed NaCl Concentration     -  4.24%

  Feed Normality              -  0.74N

  Design Brine Concentration  -  2.39N

  Design Product Quality      -  <500 mg/1

  Feed Rate                   -  0.934 mgd

  E.D. Current Density        -  1.95 amp/dm2

  E.D. Volts/cell pair        -  0.55

  R.O. Product Recovery       -  55%

  R.O. NaCl Rejection         -  97%

  No. of SS-0 Presses         -  6

  No. of Cell Pairs           -  10,800
                       95

-------
                              Brine
                              Recycle
Neutralized
Feed
        Reject
                           ฃ3—
                                                          Brine
 Dilution
 Recycle
                   I'
                    1C
                    I Feed
H Reverse
Dsmosis
                                                        Product
                                                        'Water
                 Permeate
                                                Reject
                             Reverse  Osmosis
        Units
 NaCl   lb./hz:   13,734   15,520   13,730    1,790      30   1,700

 H20   Ib./hr.  309,638  474,480   94,270  380,210 215,348  164,922

 Total  Ib./hc  323,378  490,000  108,000  382,000 215,378  166,622

 Flow   mgd     0.934   1.410   0.31    1.10    0.62    0.480
 Figure  E-3. ELECTRODIALYSIS-REVERSE OSMOSIS TREATMENT -
              STREAM  2
              Flow Diagram and Material Balance
                               96

-------
Table E-5.  COST ESTIMATE SUMMARY - ELECTRODIALYSIS
           REVERSE OSMOSIS TREATMENT OF STREAM 2
 Reverse Osmosis:

   Capital                         $  403,000

   Operating Cost                     122,760

 Electrodialysis:

   Capital                         $7,500,000

   Operating Cost

     Power                            377,800
     Membrane Replacement             294,000
     Total Labor                       68,250
     Miscellaneous                    150,000

   Depreciation                       750,000

       Total                       $1,640,050  $/yr.

 ED-RO;

   Total Capital                   $7,903,000

   Total Operating  Cost            $1,763,000  $/yr.
                           97

-------
                                    TECHNICAL REPORT DATA
                             (Please read Instructions on the reverse before completing)
  REPORT NO.
  EPA-600/2-79-021
                                                             3. RECIPIENT'S ACCESSION-NO,
4. TITLE AND SUBTITLE
 TOTAL RECYCLE  SYSTEMS FOR PETROCHEMICAL WASTE
 BRINES CONTAINING REFRACTORY  CONTAMINANTS
                           5. REPORT DATE
                             January  1979 issuing date
                           6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
   M. A. Zeitoun
   C. A. Roorda
                                                             8. PERFORMING ORGANIZATION REPORT NO.
G. R. Powers
9. PERFORMING ORGANIZATION NAME AND ADDRESS
   Dow Chemical Company
   Freeport,  Texas  77541
                           10. PROGRAM ELEMENT NO.

                                 1BB61Q
                           11. CONTRACT/GRANT NO.
                                                                   S 803085-01-0
 12. SPONSORING AGENCY NAME AND ADDRESS
  Robert S.  Kerr Environmental  Research
  Office of  Research and Development
  U.S. Environmental Protection Agency
  Ada, Oklahoma 74820
         Lab. - Ada,  OK
13. TYPE OF REPORT AND PERIOD COVERED
      FTNAT.	
                           T4. SPONSORING AGENCY CODE
                                EPA/600/15
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
     Petrochemical wastewaters  containing relatively high concentrations of
     salt  and  refractory organics  were selected  to study their  feasibility for
     total recycle.   A combination of reverse osmosis and electrodialysis was
     operated  as  a hybrid system using the pretreated wastes to produce reusable
     water and a  concentrated brine.

     The combined electrodialysis-reverse osmosis  system is not considered
     economically feasible when applied to industrial wastewaters  containing
     relatively high concentrations of salt.
17.
                                 KEY WORDS AND DOCUMENT ANALYSIS
                   DESCRIPTORS
                                               b.IDENTIFIERS/OPEN ENDED TERMS
                                         c. cos AT I Field/Group
    Activated Carbon  Treatment
    Electrodialysis
    Circulation
    Brines
               Reverse Osmosis
               Reuse
               Petrochemical Industry
                     68D
 8. DISTRIBUTION STATEMENT

    Release to Public
              19. SECURITY CLASS (ThisReport)
              21. NO. OF PAGES
                   108
                             (This page)
                                         22. PRICE
                                                 Unclassified
EPA Form 2220-1 (9-73)
                                              98
                              U. S. GOVERNMENT PRINTING OFFICE: 1979 — 657-060/1591

-------