PB81-172637
Removal of Phenolic Compounds from Wood  Preserving Wastewaters
Edward C.  Jordan Co.,  Inc.
Portland,  ME
Prepared for

Industrial Environmental Research Lab.
Cincinnati, OH
Mar 81
                        U.S. DEPARTMENT OF COMMERCE
                     National Technical Information Service

-------
                                        EPA 600/2-81-043
                                        March 1981
        REMOVAL OF PHENOLIC COMPOUNDS
      FROM WOOD PRESERVING WASTEWATERS
                     by

               Bruce K. Wallin
              Arthur J. Condren
                Roy L. Walden

         Edward C. Jordan Co., Inc.
           Portland, Maine  04112
           Contract No. 68-03-2605
         Work Directives 2(1) and 5
              Project Officers

            Donald L. Wilson and
               Brian Westfall
        Food and Wood Products Branch
Industrial Environmental Research Laboratory
           Cincinnati, Ohio  45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
           CINCINNATI, OHIO  45268

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
   EPA-6QO/2-81-043
                  PIENT'S ACCESSION NO.
                  PB81   172637
4. TITLE AND SUBTITLE
                      REMOVAL OF PHENOLIC COMPOUNDS
                    FROM WOOD PRESERVING WASTEWATERS
                                                           5. REPORT DATE
                                                               •Mi-rr.'h 1QS1
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

   Bruce K. Wallin,. Arthur J. Condren, Roy L. Walden
             3. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
                                                           10. PROGRAM ELEMENT NO-.
                  Edward C.  Jordan Co., Inc.
                  P.O.  Box 7050, Downtown Station
                  Portland,  Maine 04112
             	1BB610	
             11. CONTRACT/GRANT NO.
                  68-03-2605
                  Work Directive Nos.  21 & 5
 12. SPONSORING AGENCY NAME AND ADDRESS
               Industrial Environmental Research  Laborator
               Office of  Research and Development
               U.S.  Environmental Protection Agency
               Cincinnati,  OH  45268
             13. TYPE OF REPORT AND PERIOD COVERED
             r  Final:  11/20/78-5/20/80
             14. SPONSORING AGENCY CODE

                   EPA 600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
     Laboratory  and pilot-scale studies were  undertaken to develop economically
feasible  technologies for the treatment of wastewaters from wood preserving  operations.
Of prime  concern was the removal of phenol and  its  chlorinated derivatives,  in particu-
lar, pentachlorophenol.   Screening analysis of  the  wastewater indicated  that penta-
chlorophenol was the only chlorinated derivative  consistently present in  concentrations
of approximately 100 mg/1.
     Treatment technologies investigated  for  the  treatment of these wastewaters
included:   1. adsorption; 2. biological oxidation;  3.  chemical oxidation;  4.  coagula-
tion; 5.  extraction; and 6. pH adjustment.
     Each of the above,  along or in combination,  was  capable of yielding  a measurable
reduction in the concentration of total phenols and pentachlorophenol in  the untreated
wastewater.
     Two  technologies yielded consistently high levels of treatment:
          1.   pH adjustment of the wastewater, followed by adsorption with  bentonite
               clay and  final polishing by the  polymeric adsorbant, XAD-4; and
          2.   pH adjustment of the wastewater, followed by extraction with  a mixture
               of #2 fuel oil and a co-solvent  such as still bottoms from aaiyl alcohol
               production.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                                                                           COSATi Field/Croup
18. DISTRIBUTION STATEMENT
        Release to Public
                                              19. SECURITY CLASS (This Report)
                                                  Unclassified
                           21. NO. OF P£>GES
20. SECURITY CLASS (This pa^e)
    Unclassified
                                                                         22. PRICE
EPA Form 2220-1 (R»v. 4-77)   PREVIOUS EDITION is OBSOLETE

-------
                                   DISCLAIMER
     This report  has  been reviewed  by the Industrial Environmental  Research
Laboratory,  U.S.  Environmental  Protection Agency,  and  approved  for  publica-
tion.  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 recom-
mendation for use.
                                     -11-

-------
                                    TORE WORD
     When energy  and material resources are  extracted,  processed,  converted,
and used, the  related  pollutional impacts  on our  environment  and  even on our
health often require that  new and increasingly more  efficient  pollution con-
trol  methods  be  used.  The  Industrial Environmental  Research Laboratory  -
Cincinnati (lERL-Ci) assists  in  developing and demonstrating new and improved
methodologies  that  will meet  these  needs  both efficiently  and economically.

     Identification  and evaluation  of  treatment  techniques  for  removal  or
reduction of  pollutant discharges in  an industry (in this  instance  the wood
preserving industry) provides.useful  information  relative to the treatability
of the pollutants  of interest.  Regulatory agencies  can  use such  information
to assess the applicability of the technology in controlling the pollutants of
interest.  The Food and Wood Products Branch,  lERL-Ci,  can be contacted for
further information on the subject.
                                                  David  G.  Stephan,  Director
                                Industrial  Environmental Research  Laboratory
                                                              Cincinnati,    OH
                                     -111-

-------
                                    ABSTRACT


     Laboratory and pilot-scale  studies were  undertaken to  develop economical-
ly feasible technologies for the treatment  of wastewaters from wood preserving
operations.  Of prime  concern was  the removal  of phenol and  its chlorinated
derivatives,  in particular,  pentachlorophenol.   Screening  analysis  of  the
wastewater  indicated that  pentachlorophenol  was the only  chlorinated  deriva-
tive consistently  present in concentrations of approximately 100 mg/1.

     Treatment  technologies  investigated  for  the treatment  of  these  waste-
waters included:

     1.   adsorption;
     2.   biological oxidation;
     3.   chemical oxidation;
     4.   coagulation;
     5.   extraction;  and
     6.   pH adjustment.

     Each  of  the  above, alone  or in combination,  was capable  of  yielding a
measurable  reduction in the  concentration of total phenols  and pentachloro-
phenol in the untreated wastewater.

     Two technologies yielded consistently  high  levels  of treatment:
                  •
     1.   pH adjustment  of the wastewater,  followed by adsorption with ben-
          tonite clay  and  final polishing  by the polymeric  adsorbant,  XAD-4;
          and

     2.   pH adjustment of  the  wastewater, followed  by extraction with a mix-
          ture of  #2 fuel oil and a  co-solvent such as  still bottoms from amyl
          alcohol  production.

     Total annual  operating costs for systems treating  a typical 10,000 gpd of
wastewater  were calculated to  be  $40,000  and $23,600, respectively,  for the
two aforementioned technologies.

     This report was submitted in fulfillment of Contract No.  68-03-2605, Work
Directives Nos. 2,  Part I,  and  5, by the Edward C.  Jordan  Co., Inc. under the
sponsorship of  the  U.S. Environmental Protection Agency.   This  report covers
the period November 20, 1978 to  May  20,  1980, and work  was  completed as of May
20, 1980.
                                      -iv-

-------
                                    CONTENTS

Foreword	    iii
Abstract	     iv
Figures	     vi
Tables	     ix
Abbreviations and Symbols	    xii
Acknowledgments	,	    xiv

     1.   Summary	     1
     2.   Conclusions and Recommendations	     5
     3.   Introduction 	     7
     4.   Literature Review	     8
     5.   Treatability Studies 	    11
     6.   Selection of Treatment Alternatives	    30

References	    36
Appendices

     A.   Review of Literature on Phenolic Treatment Technologies. .    37
          1.   Biological Oxidation	    37
          2.   Foam Fractionation	    57
          3.   Solvent Extraction	    60
          4.   Physical/Chemical Oxidation  	    70
          5.   Carbon Adsorption 	    88
          6.   Stripping Operations	100
          7.   Resin Adsorption	110
          8.   Electrochemical Oxidation 	   116
          9.   Ionizing Radiation	118
          10.  Elimination of Pollutant Discharge	121
     B.   Analytical Methodologies 	   125
          1.   Analytical Procedures for Total Phenolics 	   126
          2.   Analytical Procedures for Pentachlorophenol - Mill A.   127
          3.   Analytical Procedures for Pentachlorophenol - Mill B.   129
          4.   Analytical Procedures for Pentachlorophenol - Gas
               Chromatography/Mass Spectrometry	132
          5.   Quality Control/Quality Assurance Pentachlorophenol
               Analysis	134
                                       -v-

-------
                                     FIGURES

Number                                                                l*£

  1            Phenolic structures  	  .  	        3

  2            Coagulation/adsorption, process  flow diagram ...       32

  3            Solvent extraction with oil  separation flow
               diagram	       35

  4            Biological oxidation process schematic for coke
                    plant wastes	       39

  5            Biological oxidation process schematic for wood
                    preserving wastes	.	       40

  6            Removal of 2,4-dichlorophenol  and  2,4-dichlorophen-
                    oxyacetic acid  from solution  in aeration
                    'basin effluent  by continuous  aeration. ...       42

  7            Removal of 2,6-dichlorophenol  and  2,6-dichloro-
                    phenoxyacetic acid from solution in aeration
                    basin effluent  by continuous  aeration. ...       43

  8            Removal of 2,4,5-trichlorophenol and 2,4,5-tri-
                    chlorophenoxyacetic acid  from solution in
                    aeration basin  effluent by continuous
                    aeration .	       44

  9            Removal of 2,4,6-trichlorophenol and 2,4,6-tri-
                    chlorophenoxyacetic acid  from solution in
                    aeration basin  effluent by continuous
                    aeration	• .       45

  10           Change in pentachlorophenol  concentration in
                    aerated solution in aeration  lagoon
                    effluent	       46

  11           Schematic of full scale biological treatment sys-
                    tem	  ....       49
                                       -vi-

-------
                                                                    Page

             Phenolic resistance to biodegradation 	       54

             Process schematic for solvent extraction of phe-
                  nols from coke plant wastewaters	       61

14           Dual solvent process schematic-separate cycles.  .       63

15           Dual solvent process schematic-linked cycles.  .  .       64

16           Chem-pro solvent extraction process schematic  .  .       68

17           Decomposition of phenol by wet-air oxidation.  .  .       71

18           Percentage of phenol oxidized vs.  ultrasonic in-
                  tensity	       73

19           Ozonation of phenol	       74

20           Oxidation of phenol by ozone	       75

21           Oxidation of phenol by chlorine	       78

22           Oxidation of phenol by chlorine dioxide 	       79

23           Effect of initial hydrogen peroxide/phenol ratio
                  on the oxidation of pure phenol	       81

24           Effect of hydrogen peroxide/phenol ratio on chem-
                  ical oxygen demand reduction of pure phenol.       82

25           Process flow diagram-plant "A"	       90

26           Process flow diagram-plant "B"	       92

27           Total installed cost of two-stage adsorption con-
                  tacting equipment	       96
 a
28           Total installed cost of carbon reactivation and
                  handling system	       97

29           Vapor pressure of various volatile organic com-
                  pounds 	      102

30           Air stripper schematic	      105
                                    -vii-

-------
Figures (cont'd)

Number

  31           Steam stripper schematic
  32           Flow diagram-stripper column integral with evapor-
                    ator ....................       107

  33           Stripping column design details .........       108

  34           Effect of pH on adsorption of p-nitrophenol .  .  .       Ill

  35           Rohm & Haas adsorption process flow diagram ...       114

  36           Process flow diagram closed pulp mill ......       122

  37           Process flow diagram Rapson/Reeve salt recovery
                    process ...................       123
                                      -viii-

-------
                                     TABLES


Number                                                                Page

  1            Phenolic Compound Identification	       2

  2            Individual Treatment Techniques 	 ....      12

  3            Combined Treatment Technologies 	      13

  4            Acid Addition and Filtration	      14

  5            Typical Coagulation Study Results 	      15

  6            Extraction Study Results	      16

  7            Combined Acid Addition Pretreatment/Chemical Oxi-
                    dation  	      17

  8            Combined Acid Addition Pretreatment/Coagulation
               Results	      18

  9            Combined Acid Addition Pretreatment/Extraction
                    Results	      18

  10           Combined Coagulation/Adsorption Results  	      19

  11           Acid Addition Study Results 	      20

  12           Biological Treatment Results	      22

  13           Solvent Extraction Results	      23

  14           XAD-4 Adsorption Results	      24

  15           Acid Addition/Extraction Results  	       25

  16           Acid Addition/Bentonite Clay Adsorption	       26

  17           Acid Addition/Bentonite Adsorption/XAD-4 Adsorp-
                    tion	       28

  18           XAD-4 Resin Regeneration 	       29
                                       -ix-

-------
Tables (cont'd)

Number

  19           Typical Wood Preserving Facility Wastewater
                    Characteristics
  20           Proposed Process No.  1  Estimated  Costs  .....         33

  21           Proposed Process No.  2  Estimated  Costs  .....         34

  22           Operating Results ,  Biological  Treatment of Wood
                    Preserving Wastes  .............         41
  23           Biological Pilot Plant Process  Parameters.  ...         48

  24           Full Scale (Biological) Treatment Facility  Design
                    Parameters .................         50

  25           Comparison of the General Characteristics of Three
                    Bioreactors ................         51

  26           The Biochemical Oxidation of Pure Phenols by the
                    Activated Sludge Process ..........         55

  27           Capital and Operating Costs  for Various Foam
                    Separation Systems ..........  ...         58

  28           Effect of Feed Concentration on Solvent Extrac-
                    tion Process.  .  ..............         60

  29           Equilibrium Distribution Coefficients for Various
                    Solvents ..................         62

  30           Sequential Extraction of Lube-Oil Refining  Waste-
                    water .....  . ............  .         66

  31           Economic Evaluation of Phenol Removal from  Process
                    Condensate ....... .  .........         67

  32           Catalytic Oxidation of Lacquer Manufacturing
                    Wastewater ......... . .......         72

  33           Hydrogen Peroxide Treatment of Phenol Wastes .  .         80

  34           Reaction of Phenol with Fenton's Reagent ....         83
                                      -x-

-------
                                                                    Page

             Effect of Catalysts on Phenol Oxidation	         84

             Phenol Adsorption Systems	         89

             Operating Data-Plant "A"	         9.1

             Operating Data-Plant "B" 	         93

             Effects of Carbon Regeneration 	         94

             Impurity Removal Decrease after Regeneration .  .         94

41           Major Annual Cost Considerations for Adsorption
                  Options	         98

42           Characteristics of Various Condensate Treatment
                  Systems at 5 Bleached Kraft Mills 	        103

43           Pollutant Load by Contaminated Condensates .  .  .        104

44           Removal of Phenol from Contaminated Condensates
                  by Steam Stripping	        104

45           Effect of Salt Content upon Capacity of Amberlite
                  XAD-4 for Phenol and m-Chlorophenol Adsorp-
                  tion	        112

46           Adsorption of Phenolic Compounds on Amberlite XAD-
                  4 at 25°C and Flow Rate of 0.5 gpm/cu ft.  .        113

47           Effect of Gamma Irradiation and Oxygen Pres-
                  sure on pH, Suspended Solids and COD of
                  Various Pulp Mill Effluents	        119
                                     -xi-

-------
 ABBREVIATIONS
                         LIST OF ABBREVIATIONS  AND SYMBOLS
 BOD
 BOD5
 COD
 DI
 ECD
 EPA
 ft
 ft2
 ft3
 FID
 g
 gal
 GC
 GC/MS
 8/1
gpm
tp
in.
kg
kW
kWh
1
Ib
bl/day
min
ml
mg
mg/1
ml/1
mug/1
MLSS
MLVSS
N
nm
O&M
POTW
ppm
sec
SVI
ug/1
w
wt
—biochemical oxygen demand
—5-day biochemical  oxygen demand
—chemical oxygen demand
—de-ionized
—electron capture detection
--U.S.  Environmental Protection Agency
--feet  (foot)
—square feet
—cubic feet
--flame ionization detection
--gram
--gallon
—gas chromatography
—gas chromatography/mass  spectroscopy
—grams/liter
•-gallons  per day
•-gallons  per minute
•-horsepower
•-inches
•-kilogram
•-kilowatt
•-kilowatt-hour
•—liter
•-pound
•-pound/day
•-minute
•-milliliter
•-milligram
•-milligram/liter
•-milliliter/liter
•-millimicrograms/liter
•-mixed liquor suspended solids
•-mixed liquor volatile suspended solids
•-normality
•-logmeters
•-operation and maintenance
•-publicly owned treatment works
•-parts per million
•-seconds
•-sludge volume index
•-micrograms/liter
•-watt(s)
•-weight
                                       -xii-

-------
SYMBOLS

°C                  degrees Centigrade
CH2C12,              dichlorome thane
°F                  degrees Fahrenheit
HC1                 hydrochloric acid
H2S04               sulfuric acid
H20                 water
NaOH                sodium hydroxide
±                   plus or minus
                                     -xiii-

-------
                                 ACKNOWLEDGMENTS


     This report was prepared by Dr.  Bruce K. Wallin, Dr. Arthur J.  Condren,
and Roy  L.  Walden of  the  Edward C.  Jordan Company in Portland, Maine.   Key
staff  members  participating  in the project  were  Robert  A.   Steeves  and
Frederick A.  Keenan.

     Laboratory analyses  during the  preliminary phase  of  the  project  were
conducted by EMS Laboratories, Indianapolis,  IN, under the  direction  of Mr.  C.
Stephen  Gohmann.   All  remaining analyses  were  performed by the  Edward  C.
Jordan Company laboratory under  the  direction of Dr. Bruce  Wallin.  Throughout
the project,  the  Gulf  South  Research Institute  New  Orleans, LA,  under  the
direction of  Dr. Roger  Novak,  provided gas  chromatography/mass  spectroscopy.
                                     -xiv-

-------
                                    SECTION 1

                                  INTRODUCTION
     The U.S.  Environmental  Protection Agency has been involved  in extensive
investigations of toxic compounds being discharged from industrial facilities.
Primary emphasis has been on the 65 "priority" pollutants,  which are contained
in  the Settlement  Agreement  of 1976  and in  the Clean  Water  Act of  1977
(PL 95-217).  Phenolics and  their chlorinated derivatives  are a  part  of  this
group and are commonly found in the wood products industry's wastewater streams.

     The initial  objective of this study was to  evaluate  the treatability of
2,4,6-trichlorophenol,  parachlorometacresol,   2-chlorophenol,   2,4-dichloro-
phenol, and pentachlorophenol in. the wood products industry's wastewater.   The
wood preserving industry was selected for this program because wastewater  from
wood preserving facilities is usually low in volume but high in concentrations
of  chlorinated phenolics.   A chemical  screening  of  the  wastewater revealed
that  it had  a high  organic  content,  but  more significantly,  it contained
pentachlorophenol in  concentrations exceeding 100 mg/1.   The program's  focus
then  shifted  primarily to pentachlorophenol  to the virtual  exclusion  of all
else,  since only  trace  levels  of  other chlorinated  phenolics  were  found.

     In  conjunction with  the treatability aspect, the program  was intended
also  to explore atypical  pretreatment schemes  that would reduce chlorinated
phenolics,  namely pentachlorophenol,  in typical wood  preserving wastewater to
levels  at which the wastewater could be  discharged to  a  POTW without causing
an  upset.   Owing to  constraints imposed by time and  financial  resources, it
was  not the  objective of  this  program to  investigate  all aspects  of  each
treatment scheme  (e.g.,  residual catalyst after PCP removal,  toxicity  of the
removal  concentration,  adsorptive  capacity  of  the regenerated  resin).   The
objective here  was  also not to  improve  the phenol-contaminated  wastewater to
drinking water  quality,  nor  was it within the bounds  of the study to make any
conclusions concerning  the  toxicity of the residual pentachlorophenol.concen-
tration in the wastewater that would be discharged to  the POTW.

     The physical/chemical  properties of  pentachlorophenol,  as  distinguished
from those  of  phenol,  were important in  considering the  pretreatment systems
which  would be  capable  of  reducing  concentrations  to an  acceptable  level.
Pentachlorophenol consists  of  a benzene  ring  (C^H,)  with  all  six hydrogen
                                     -1-

-------
 sites  substituted by  one  hydroxyl group (OH) and  five  chlorine atoms (C15)-
 The  resulting compounds,  C-C1 OH,  is mildly  acidic,  boils at  309°C,  and is
 soluble  in 50°C water  at 30 mg/1.  Oils or emulsions in wastewater can provide
 a  "carrier"  effect,  allowing pentachlorophenol to far exceed its normal solu-
 bility  in water.   Table 1 summarizes the physical  properties  of several phe-
 nolic compounds.  Figure 1 presents phenolic structures for several compounds.


                                     TABLE 1

                PHYSICAL PROPERTIES OF SEVERAL PHENOLIC COMPOUNDS
 Compound
Solubility in HO
   mg/1 @ 25°C
                                                                     Ka x 10
                                                                            10
Phenol
o-Chlorophenol
m-Chlorophenol
p-Chlorophenol
2 , 4-Dichlorophenol
2 , 4 , 6-Trichlorophenol
Pentachlorophenol
o-Cresol
m-Cresol
p-Cresol
p-Chloro-m-Cresol
2,4, 6-Trichloro-m-Cresol
182
173
214
220
210
246
309
191
.201
202
196
265
93,000
28,000
26,000
27,000
4,500
900
30(50°C)
25,000
26,000
23,000
insol.
si. sol.
1.1
77
16
6.3
Large
Very Large
Very Large
0.63
0.98
0.67
Unknown
Unknown
Ka = .thermodynamic acid dissociation constant
     A review  of  available  literature assisted the investigators in identify-
ing  treatment  techiques for investigation.  Investigations were divided into
two phases:  preliminary bench-scale treatability studies and an evaluation of
batch treatment techniques.

     Unlike phenol, which is  quite unstable and easily  oxidized either chem-
ically or  biologically, pentachlorophenol  is  stable and  resistant to oxida-
tion.  In some instances, though, when wastewater containing pentachlorophenol
in concentrations  which a biomass  can tolerate  is  run through a biological
treatment system,  pentachlor.opb.enol is adsorped onto the biofloc.  Disposal of
the sludge, now laden  with  pentachlorophenol,  then presents a hazardous waste
problem.   Alkylinization  of  the  sludge,  a relatively  common practice, would
more than likely  release  the  pentachlorophenol into .the environment.   If the
sludge were  incinerated, 2,3,7,8-tetrachlorodibenzo-p-dioxin,  a  thermal de-
gradation product  of pentachlorophenol, could be released into the atmosphere.
                                     -2-

-------
PHENOLIC  RESISTANCE

 TO  BIODEGRADATION (2)
       PHENOL
       0-CHLOROPHENOL
       2,4  DICHLOROPHENOL
       2,4,6 TRICHLOROPHENOL
       PENTACHLOROPHENOL
a:
o
UJ
a
o

CD
                                     Ul
                                     O
                                     z
ui
ct

a
UJ
en
<
ui
cc

-------
      These  rather  ominous  prospects make such treatment techniques as solvent
 extraction  and  batch polymeric  resin adsorption,  two schemes  investigated
 during  this study, more appealing from the standpoint of avoiding the creation
 of  a  hazardous waste problem while improving water quality.

      In the preliminary bench  scale studies the  investigators  traveled to a
 wood  preserving  facility on the west coast to evaluate treatment technologies,
 described  in the  literature,  designed to lower  pentachlorophenol concentra-
 tions.   Treatment  schemes  tested included pH reduction, acid  cracking,. chem-
 ical  coagulation,  chemical oxidation, ultrafiltration,  resin  adsorption,  and
 solvent extraction.

      After  conducting the preliminary bench scale  studies,  the investigators
 reported to  representatives of the EPA.   At this time, mid-course corrections
 were  suggested.    It  was  suggested  that  the program  emphasis be  shifted to
 focus on only those  pretreatment systems that would  both lower pentachloro-
 phenol  concentrations to  levels acceptable for  discharge to  a- POTW  and be
 economically feasible for wood preservers.

      To  meet the revised criteria,  pretreatment schemes had to be both effi-
 cient and  economically feasible, which essentially  eliminated from consider-
 ation the more  elaborate treatment technologies (e.g., reductive degradation,
 electrochemical  oxidation, ion  exchange,  rotary vacuum filtration with activ-
 ated  carbon).   The effectiveness of  these  cannot be denied,  but  the cost of
 installing and operating any one of them makes  it economically impractical for
 wood preservers.

     Batch treatment  systems were set up at the  southern  facility to provide
 some  indication  of which methods were cost effective.  Technologies tested at
 the southern facility  included  batch biological treatment, chemical and poly-
meric coagulation,  resin adsorption,  acid  cracking, bentonite  clay, solvent
extraction,  and  filtration.  As  at  the west coast  facility,  a combination of
these were  tested.   The  size  of  the  batch  treatment reactors  ranged from
several  liters  to  50 gallons,  depending on  the  pretreatment system  being
tested.
                                     -4-

-------
                                    SECTION 2

                             SUMMARY AND CONCLUSIONS

     Processes  for  the  treatment  of  wastewaters  from  two  wood  preserving
facilities were  investigated at  the  laboratory and pilot  scale  levels.   In-
cluded  were  various  chemical,  physical,  and biological operations  selected
specifically  for  the removal  of  phenol and  its  chlorinated  derivatives.   Of
the treatment  technologies  investigated,  two systems consistently lowered the
concentrations of  pentachlorophenol in  the  wastewater from  100  mg/1  to  less
than 1 mg/1.

     In the  first  system, the wastewater was first acidified to a pH of 4.0 ±
0.1, then bentonite  clay was added.   A polymeric  adsorbant,  amberlite XAD-4
was used  in the final  polishing  process.   In the second system  a  mixture of
No. 2  fuel oil and a cosolvent  (amyl  alcohol still bottoms) was used  to ex-
tract  pentachlorophenol from the waste  stream.   Reductions were  consistently
in excess of 99 percent.  Since No. 2 fuel oil is used often in the  preserving
process,  as  it was at  this facility,  it  is  conceivable that a facility could
operate  this  extraction  process  without  incurring  any additional  chemical
expense,  except possibly  for the cosolvent.   Trials with No.  2 fuel oil alone
yielded removal efficiencies  in  the vicinity of 97 percent, which may be high
enough  to  allow  the wastewater to be  discharged  to  a POTW.  For both systems
to function consistently, the wastewater first had to be subjected to free oil
separation and flow  equalization.  The following summarizes  the findings of
other  investigations, based on  reductions in total phenol and/or pentachloro-
phenol concentrations.

pH Adjustment

     Lowering  the  pH  of the wastewater with sulfuric acid was found to induce
the  formation of  colloidal material.    Subsequent  removal  of  this colloidal
material  resulted  in  slight reductions' in total  phenol  concentrations  as
measured  by  the  modified lowry  procedure described  in Appendix  B-l.   Penta-
chlorophenol  concentrations,  however,  were  consistently  reduced  from approx-
imately 100 mg/1 to less  than 20 mg/1.

Biological Oxidation

     Biological oxidation was not found to be an effective treatment technique
because bioadsorption rather  than biotransformation was found  to be the pri-
                                     -5-

-------
mary removal mechanism.  Removal rates continued to diminish as the adsorptive
capacity of the biomass was approached.

Chemical Oxidation

     At high  doses,  chlorine yielded substantial reductions in  the compounds
of  concern.   Hydrogen peroxide, on  the other  hand,  had little  effect.   The
high  chemical  demand  observed  precludes  the use  of  chemical oxidation  as a
viable treatment alternative.

Coagulation

     Coagulation  with  alum,  ferric  chloride,  and/or  polymers  resulted  in
modest pollutant  reductions.   These  reductions were not deemed  sufficient to
justify coagulation as a sole treatment  technology.

Applicability of Technologies

     In turning now to the applicability of the programs'  findings to the  wood
preserving  industry  as a whole, it  should be understood that the  utility of
any single pretreatment option is contingent upon both the volume and chemical
make-up of  the waste  stream.   Simply put, because wood preserving processes
are  so  variant,  each  facility  must  be  evaluated in terms of  its  preserving
process, its waste stream,  and  the capital available  for investing in a  pre-
treatment system.  The systems  devised  for wood preservers  in connection  with
this study  would probably not be transferable to the  leather tanning industry
or  to  the paper  industry,  not because  a  No.  2 fuel oil -  cosolvent mixture
would  not  remove  pentachlorophenol  from  leather  tanning  or paper industry
effluent just as effectively as it would from wood preserving wastewater, but
because neither  of these  industries  uses No.  2 fuel  oil  in its  production
process (non-combustion) as  do  some  members of the wood preserving industry.
The advantage of the  fuel oil extraction process is that the pentachlorophenol
can be  removed from  the  wastewater  without creating an additional waste and
without bringing  large  capital and  operating expense  to  bear on  the  wood
preserver.
                                     -6-

-------
                                    SECTION 3

                                 RECOMMENDATIONS
     If results  obtained during  these  investigations are  to  be verified,  a
continuously  flowing  pilot-scale  system must  be  established.  Because  this
study was  done on  a  quick response  basis  with limited  financial  resources,
researchers  could  not  investigate all  aspects of  each pretreatment  scheme
(e.g., residual catalyst  after  pentachlorophenol  removal, toxicity  of removal
concentrations, utility or  necessity of  catalysts   other  than amyl  alcohol
still bottoms).   Instead,  the  project provided an overview of the  many  eco-
nomically achievable approaches  to removing pentachlorophenol  from  wastewater
and identified two systems especially successful in lowering pentachlorophenol
concentrations in wood preservers' wastewater to levels that would  allow it to
be discharged to the POTW without causing an upset.
                                     -7-

-------
                                    SECTION 4

                                LITERATURE REVIEW

      The  treatment  of toxic compounds is a new  and  rapidly developing field.
 A detailed literature review  was  prepared on  the removal  of  phenolics from
 wastewaters to assess the state-of-the-art.

      Applicable  treatment  processes  identified  in  the  literature were  as
 follows:

           1.   biological oxidation;

           2.   foam fractionation;

           3.   solvent extraction;

           4.   chemical oxidation;

           5.   Carbon adsorption;

           6.   gas stripping;

           7.   resin adsorption/ion exchange;

           8.   coagulation/precipitation;

           9.   electrochemical oxidation;

           10.   ionizing radiation;  and

           11.   elimination of discharge.

     None  of  the above  exhibited  universal applicability,  although each was
useful in  specific  instances.   A detailed review of each process can be found
in Appendix A, and a number of these are summarized below.

1.   Biological oxidation has  been shown capable of 99-percent phenol removal
under  proper  conditions  (pH,  detention time,  nutrient  supplements,  etc.).
Although phenol bio-oxidation  is a generally recognized phenomenon, the abil-
                                     -8-

-------
ity  of micro-organisms  to  chemically modify  certain substituted phenols  is
controversial.

2.   Foam  fractionation  has considerable value as  a  pretreatment technology,
although it is not promising as an independent treatment system due to its low
efficiency for  phenol removals.   Its  ability to  limit the effects  of  shock
loadings and  remove a portion of the  wastewater's toxicity  allows  for more
stable, and efficient use of existing biological treatment systems.

3.   Solvent extraction is extremely effective because the solvent used can be
tailored to the removal of specific compounds.   Again, greater  than 99 percent
phenol removal  has  been  reported.   Recovery of chemicals can offset a portion
of  the operating costs.   Capital  costs, however,  can prove  excessive  since
considerable equipment  and instrumentation is required  for  efficient process
control.   Additional  downstream effluent polishing may be  required  prior to
final  discharge.

A.   Chemical  oxidation  can  completely remove  phenolics  from  wastewater.
However, the required doses of oxidizing agents are generally so high that the
processes  cannot  economically  compete  with  other processes,  except in special
applications.  Similarly, high capital costs for the wet air oxidation process
make this process uneconomical.

5.   Carbon  adsorption  has been demonstrated  commercially to  achieve greater
than  99-percent phenol  removal.   Reasonable  service  life,  however,  requires
some  form  of  wastewater  pretreatment.   System operation is also complex.  The
economical use  of carbon necessitates regeneration of  spent material,  accom-
plished  either  onsite or  through  a contract  regeneration service.   A second
use  of carbon  is the direct addition of  this  material to an activated sludge
system.  Removals of BOD5,  COD and TOC improve, and  the  biological  system's
ability  to react to and  recover  from  shock  loadings is increased.   Carbon
adsorption alone  is too  expensive  for  the  treatment of pulp,  paper  and wood
products industry wastewaters.   When combined with the proper pretreatment and
other  processes,  however,  it can provide effective  and economical treatment.

6.   Gas stripping  failed  to remove phenolics in all  studies  reported in the
literature.  On the other hand, steam stripping is viable.  Steam stripping is
frequently used as an odor removal/chemical recovery 'system.   When this is the
case,  phenol  removals can  occur but probably not  to  acceptable  levels  since
the  operating  parameters would  be tailored to  the unit's  primary  function.

7.   Resin adsorption can also provide greater than 99 percent phenol removal.
Under  certain  conditions,  recovery  of  various marketable chemicals,  such as
phenol,  is  also  possible.   Several  full scale systems  have been constructed
for  chemical  processing  plants.   Again,  for  the  treatment  of wood  products
industry effluents,  some wastewater pretreatment is  necessary to protect the
resin  (i.e.,  methanol).   Operation of the system may be complex and expensive,
if conventional regenerating substances are used.
                                     -9-

-------
8.   Coagulation/precipitation processes  for phenolic  removals  were reported
by the SUNY College of Environmental Science and Forestry in a report entitled
"Toxicity Reduction Through Chemical and Biological Modification of Spent Pulp
Bleaching Liquors"  (EPA-600/2-80-039).  It  was  reported that coagulation/pre-
cipitation processes remove the higher molecular weight phenolics efficiently.
The majority  of toxicity, however,  was attributable  to  the lower  molecular
weight species.  Appendix  A  contains  a review of the completed  phases of this
study through February 1978.

     The  remaining  technologies  investigated,   electro-chemical  oxidation,
ionizing  radiation,  and  elimination  of  discharge,  are  all reported  to  be
highly efficient  in terms of  phenol  removal, provided  the proper  operating
conditions are  maintained.   In most  cases,  however, the achievement of  high
removals was  economically infeasible.
                                    •10-

-------
                                    SECTION 5

                              TREATABILITY STUDIES
INTRODUCTION

     The objective  of  the  program was to determine the treatability of pheno-
lic compounds and  its  chlorinated derivatives in the  wood  products industry.
Of particular  interest was  pentachlorophenol.   The wood preserving industry
was selected for  this  program because wastewater from wood  preserving facili-
ties  are  typically low  in volume but  high in  concentrations  of chlorinated
phenolics, particularly pentachlorophenol.

     Wastewater treatment studies were conducted at two wood preserving facil-
ities.  Preliminary studies were  undertaken  at a  west coast  facility while
more  in-depth  efforts  were  completed  at  a  facility  located  in  the  south.
Table  2 presents  the  individual  treatment  techniques investigated  at each.
                                     -11-

-------
                    TABLE 2.  INDIVIDUAL TREATMENT TECHNIQUES
                                                   Facility location
     Type of treatment                            West coast     South
A.   Acid addition                                     X           X

B.   Biological oxidation                                          X

C.   Chemical oxidation
     1 .   Chlorine                                     X

D.   Coagulation
     1.   Alum                                         X           X
     2.   Ferric chloride                              X
     3.   Polymer                                                  X

E.   Extraction                                        X           X

F.   Resin adsorption
     1.    Amberlite XAD-2                              X
     2.    Amberlite XAD-4                                          X
     3.    Bentonite clay                                           X
     Combinations  of the  above  individual  treatment  techniques  were also
investigated and Table 3 presents those undertaken.
                                     -12-

-------
                    TABLE 3.  COMBINED TREATMENT TECHNOLOGIES
Preliminary
                 Type of treatment
Intermediate
Final
Facility location
   West
   Coast  South
Acid addition
Acid addition

Acid addition
Acid.addition

Acid addition
Acid addition
Acid addition
Acid addition
Acid addition
Alum coagulation
Alum coagulation
Ferric chloride coagulation
Alum coagulation   Polymer coagulation
Acid addition      Polymer coagulation
Acid addition      Bentonite clay
                    adsorption
                 Chlorine oxidation          X
                 Hydrogen peroxide
                 oxidation                   X
                 Alum coagulation            X
                 Ferric chloride coagula-
                 tion                        X
                 Polymer coagulation                X
                 Extraction                  X      X
                 XAD-2 resin adsorption      X
                 XAD-4 resin adsorption             X
                 Bentonite clay adsorption          X
                 Polymer coagulation                X
                 XAD-2 resin adsorption      X
                 XAD-2 resin adsorption      X
                 XAD-4 resin adsorption             X
                 XAD-4 resin adsorption             X
                 XAD-4 resin adsorption             X
     Due  to differing wood  preserving processes  and  chemicals at  these  two
facilities,  wastewater  characteristics  varied  substantially.   As  a  result,
treatment  techniques  applicable  to the wastewaters from one facility were not
necessarily transferable to the other.  Therefore,, results of the treatability
studies  for each facility are presented in the order of type of treatment in-
vestigated, as in Tables 2 and 3.

     The  primary pollutants  of  concern in these  studies were  phenol  and the
chlorinated  derivatives  thereof.   Total phenols were measured  by a modifica-
tion  of  the  Lowry procedure,  and chlorinated  phenols  were measured  by gas
chromatography.  Appendix B gives specific analytical procedures.

INVESTIGATIONS AT A WEST COAST FACILITY

     The  wood preserving facility  located on  the west coast  uses the boul-
tonizing  process  of  wood preparation and both organic and inorganic chemicals
to  impart  preserving and  fire  retardency  properties  to  the  wood products
processed.   The  resultant  wastewater  volumes  approximating 10,000  gpd,  are
subjected  to treatment using  the following processes:
                                     -13-

-------
     o    wastewater storage/flow equalization;

     o    phase (free oil) separation employing a Lamella separator;

     o    chemical coagulation;

     o    filtration employing paper cartridge filters;

     o    ultrafiltration; and

     o    hyperfiltration (reverse osmosis).

     The  above  treatment  procedure  was intended to remove all organic and in-
organic pollutants  from  the  wastewater,  so that the  resultant  water could be
used as boiler  feed water.   The pollutants separated from the wastewater were
to  be  burned in  the facility's boiler, thereby recovering  potential energy.
Due to problems with membrane fouling, the latter two unit operations were not
continuously  operative  throughout the treatability studies  at  this  facility.

     The  following  subsections  describe  the  studies  undertaken at  this fa-
cility.

Acid Addition

     Preliminary  investigations  indicated  that alteration of the wastewater's
pH  through  the  addition  of concentrated sulfuric  acid  directly to the waste-
water  with  constant mixing  resulted in the  formation of particulate matter
that could be readily removed by filtration through Whatman No.  41 paper  (mean
pore diameter =  0.25 microns).   Table 4  presents typical  results  of pH ad-
justment  to a value of 2  and subsequent  filtration on both total phenol (as
2,4-dichlorophenol) and pentachlorophenol  concentrations.

	TABLE 4.  ACID ADDITION AND FILTRATION	

                     	Total phenol,* mg/1	     Pentachlorophenol, mg/1
Sample description   Raw    Treated    % Removal    Raw    Treated    % Removal
Raw waste
Lamella effluent
UF filtrate
9,800
1,300
2,500
8,400
1,300
2,300
14
0
8
14
6
40
12
4
16
14
33
60

*as 2,4-dichlorophenol

     There was indication that pH adjustment to 2 may be excessive; additional
laboratory studies have shown that the solubility of pentachlorophenol at a pH
of 3.7  and at  22°C approximates  15  mg/1.  With  raw wastewater pentachloro-
                                     -14-

-------
phenol  concentrations  of 80  to  100  mg/1,  reductions in excess  of  80 percent
were realized after mild acidification to pH 2 to 4, and filtration.  This was
presumably a solubility phenomena.

Chemical Coagulation

     Coagulation  studies using  a Phipps  and  Bird jar  test apparatus  were
conducted  using a rapid  mix time of  one minute,  a flocculation  time  of 10
minutes, and  a  settling time of  one hour.   The resultant supernatant samples
were  then  filtered through  Whatman  No.  41 paper prior  to  analysis.   Optimum
conditions using alum as the coagulant were a pH of 6.0 and a concentration of
150  mg/1;  the optimum conditions for  ferric  chloride  were  a pH of 8.0  and a
concentration of  300  mg/1.   Typical  results on the coagulation of the Lamella
separator  effluent,  as  shown  in Table 5, indicate  the  effectiveness  of both
alum  and ferric  chloride  for the reduction  of both total  phenol  and penta-
chlorophenol.   Based  on the  limited  testing it was noted that  as  the penta-
chlorophenol concentration decreased, the removal efficiency increased.

                   TABLE 5.  TYPICAL COAGULATION STUDY RESULTS
Coagulant
    Total phenol,* mg/1         Pentachlorophenol,  mg/1
Raw    Treated    % Removal    Raw    Treated    % Removal
Alum
Alum
Ferric
Ferric
4,600
300
chloride 4,600
chloride 430
4
N
4
N
,200
.A.
,250
.A.
9 12
' 20
8 12
90
.5
.0
.5
.0
5
10
3
70
.0
.0
.0
.0
60
50
76
.22

*as 2,4-dichlorophenol
N.A. - not analyzed
Chemical Oxidation

     Oxidation of the organics in the wastewater was investigated using sodium
hypochlorite as  the  oxidant.   To achieve a  slight  chlorine residual after 30
minutes.of  reaction,  the applied dose, as  chlorine,  approximated 3,000 mg/1.
Such a  dose usually reduced the total phenol  concentration in Lamella separ-
ator effluent samples by approximately 70 percent (from 10,700 to 3,000 mg/1),
and the pentachlorophenol  concentration  by 99 percent  (from  6.0 to less than
0.05 mg/1).  The  high  oxidant  demand of  the wastewater  was  felt to  be  of
sufficient magnitude to terminate further chemical oxidation studies.
                                     -15-

-------
Extraction
     Solvent extraction experiments were undertaken using  a  4:1  sample:solvent
(v/v)  ratio.   Lamella  separator effluent samples were extracted with  various
organic  solvents  by  rapid shaking  in  a  separatory funnel  for 30  seconds,
followed by  quiescent phase  separation.   Table  6  presents  results of  these
studies.

                       TABLE 6.   EXTRACTION STUDY RESULTS	
                         Total phenol.* mg/1          Pentachlorophenol.  mg/1	
Extractant           Raw    Treated    % Removal     Raw    Treated    % Removal
1,1,1-Trichloro-
ethane
Isobutyl alcohol
Carbon tetrachloride
Kerosene
Freon 113
Ethyl acetate

1,300
1,300
1,300
1,600
1,600
1,600

500
80
800
1,450
1,350
250

62
94
38
9
16
84

6.0
6.0
6.0
16.0
16.0
16.0

2.0
N.A.
N.A.
15.0
20.0
6.0

67
-
-
6
-
63

*as 2,4-dichlorophenol
N.A. - not analyzed


     Although  substantial  reductions  in both  total  phenol  and  pentachloro-
phenol  were  realized with  certain extractants,  additional studies  were  not
undertaken because of the high solubility of these solvents in water.

Resin Adsorption

     Adsorption studies  using Rohm and  Haas Company's XAD-2 polymeric resin
were undertaken on Lamella separator effluent.   Resin columns with a height to
diameter  ratio of  four were  used in these studies  and the  wastewater  was
passed through the  columns  at a rate of 0.1 bed volumes per minute.  Five bed
volumes of wastewater were  applied during each trial.   Results were usually-a
reduction  in total  phenol  concentration  of 98  percent  (from 4,600  to  100
mg/1), and a pentachlorophenol reduction in excess of. 99 percent (from 12.5 to
less than 0.05 mg/1).
                                     -16-

-------
     The  resin  was  reused after regeneration with  isopropyl  alcohol.   Regen-
eration was  at a  rate  of 0.3 bed volumes  per  minute, and a  total  volume  of
three bed volumes of isopropyl alcohol was used  per regeneration.

Acid Addition/Chemical Oxidation

     Chemical oxidation  of the  organics  in Lamella  separator effluent which
had  been  subjected  to  pretreatment  by  the  acid  addition process was  also
investigated.   Both  sodium hypochlorite  and  hydrogen peroxide were investi-
gated as oxidizing agents.  Chlorine demand of the wastewater, after 30 minutes
of  reaction,  approximated 3,000  mg/1,  and  hydrogen  peroxide,  5,000  mg/1.
Table 7 presents  typical  results of pH adjustment  to a value of  2 and subse-
quent chemical  oxidation.

    TABLE 7.  COMBINED ACID ADDITION PRETREATMENT/CHEMICAL OXIDATION RESULTS

                        'Total phenol,* mg/1-         Pentachlorophenol, mg/1
Chemical oxidant     Raw**  Treated    % Removal    Raw    Treated    % Removal
Sodium hypochlorite
Hydrogen peroxide
10
10
,700
,700
3
11
,000
,600
72
6.
6,
.0
.0
6.
4.
0
5
0
25

*as 2,4-dichlorophenol
**before pH adjustment


     Comparison  of  the  results presented in Table  7  with those presented for
individual  treatment by either acid addition  or  chemical oxidation indicates
no benefit to combining these two methods of wastewater treatment.

Acid Addition/Coagulation

     Coagulation  of  Lamella  separator  effluent  that  had been  subjected to
pretreatment  by the  acid  addition process previously  described  was  also in-
vestigated.   Both  alum  and   ferric  chloride  were  employed  as  coagulants.
Optimum alum  dose  approximated 150 mg/1 at a pH of 6.0.  For ferric chloride,
the  optimum dose was  300 mg/1 at  a pH of 8.0.  Table  8 presents  results of
these studies.
                                     -17-

-------
	TABLE 8.  COMBINED ACID ADDITION PRETEEATMENT/COAGULATION RESULTS	

                   Total phenols,* mg/1          Pentachlorophenol, mg/1
Coagulant      Untreated  Treated  % Removal   Untreated  Treated  % Removal

Alum             4,700     5,400        -         12.0      3.0        75

Ferric chloride  4,700     4,300        6         12.0      5.0        58


*as 2,4-dichlorophenol

     Comparison  of  the results  presented  in Table 8 with  those presented in
Tables 4 and 5 indicates an increase in pentachlorophenol removal when the two
processes are combined.

Acid Addition/Extraction

     Extraction  of  Lamella  separator effluent  which  had  been  subjected to
pretreatment  by the  acid addition  process was  also investigated^   The ex-
traction technique employed  was  as previously described, and results' of these
studies are as follows:

        TABLE 9.  COMBINED ACID ADDITION PRETREATMENT/EXTRACTION RESULTS


Total
phenol,* mg/1
Extractant Untreated Treated
1,1,1-Trichloro-
e thane
'Isobutyl alcohol
Carbon tetra-
chloride
Kerosene
Freon 113
Ethyl acetate
1,300
1,300
1,300
1,600
1,600
1,600
550
100
800
1,400
1,400
N.D.
% Removal
58
92
38
13
13
>99
Pentachlorophenol ,
Untreated Treated
6.0 2.0
6.0 N.A.
6.0 N.A.
20.0 10.0
20.0 12.0
20.0 <1
mg/1
% Removal
67
-
_
50
40
• >95
*as 2,4-dichlorophenol
                                     -18-

-------
     Comparison of  the results  presented  in Table 9 with  those  presented in
Table 6 indicates supplemental removal can be realized for certain extractants
if the wastewater is first subjected to acid addition treatment.

Acid Addition/Adsorption

     Pretreatment of Lamella separator effluent with the acid addition process
followed by treatment with XAD-2 resin was also investigated.  Greater than 99
percent removal of  total  phenols and pentachlorophenol was realized with this
combined process.   Wastewater subjected to this treatment contained 4,700 mg/1
total phenols and 12.0 mg/1 pentachlorophenol.

Coagulation/Adsorption

     XAD-2  adsorption of  Lamella separator  effluent  pretreated  by  alum and
ferric  chloride  coagulation  was  also  investigated.   Coagulation conditions
were as previously  reported,  and Table 10 presents  results of these efforts.

	TABLE  10.  COMBINED COAGULATION/ADSORPTION RESULTS .	

              	Total phenol,* mg/1	      Pentachlorophenol, mg/1
Coagulant      Untreated  Treated  % Removal   Untreated  Treated  % Removal
Alum
Alum
Ferric
4,200
5,400
chloride 4,300
N.D.
300
300
>99
94
93
5.0
3.0
3.0
N.D.
N.D.
N.D.
>99
>99
>99

*as 2,4-dichlorophenol
N.D. - none detected

Comparison of  these results with those presented for adsorption alone or acid
addition/adsorption  combined treatment  where  removals  were  already greater
than  99  percent  indicates  no supplemental removal of either  total  phenol or
pentachlorophenol.

INVESTIGATIONS AT A SOUTHERN FACILITY

     Wood  preserving at  the southern  facility  was  undertaken  with organic
compounds, primarily  pentachlorophenol.   The  wood is preconditioned using the
steaming  process  whereby the wood is steamed  to prepare  it for preservative
impregnation.  Production of the preserved product results  in the generation
of  approximately  10,000  gpd  of wastewater.   The  wastewater  is  subjected to
free oil  removal  before it is  conveyed  to a  storage/evaporation pond.  Some-
times discharge from this pond enters the POTW  system.
                                     -19-

-------
     The  following  subsections describe  the studies  undertaken at  this  fa-
cility.

Acid Addition

     Acid addition studies similar to those  conducted at the west coast facil-
ity were conducted on the wastewaters from this  facility.   Upon decreasing the
pH of  the wastewater by  sulfuric acid addition,  the  formation of colloidal
matter was  noted.   Filtration through Whatman  No.  41 paper would  not remove
this material;  however,  the material  could be  removed by  filtration of the
acidified sample through Gelman Type  AE paper.   The  Gelman type AE paper has a
mean pore  diameter  of 0.45  microns, whereas the Whatman type  41  has a mean
pore diameter of  0.25 microns.  Results  of  these acid  addition/ Gelman paper
filtration studies are presented in Table  11.

                     TABLE 11.  ACID  ADDITION STUDY  RESULTS
                 pH
                               Pentachlorophenol,  mg/1
Trial
Initial
Final
Untreated
Treated
% Removal
1
2
3
6.0
6.0
6.0
2.0
2.0
4.0
50
128
128
4.5
10.0
7.0
91
92
95

     The  above data  indicate a  substantial reduction  in  pentachlorophenol
through pH adjustment.

Biological Treatment

     Biological treatment   (batch  type)  studies  used three  50-gal  aerobic
reactors  which were operated on  a  24-hr  fill and  draw basis.   Wastewater
containing approximately 35 mg/1 pentachlorophenol was added to  the  bioreac-
tors according to  the following schedule:
                    Reactor"

                       A

                       B

                       C
                                        Volume  added daily,  gal
Wastewater
25
20
15
Tap water
0
5
10
Total
25
25
25
                                     -20-

-------
     Supplemental nutrients  were added,  as  required, to  insure  a  trace con-
centration in  the  supernatant withdrawn from the  three  bioreactors  each day.
MLVSS in the bioreactors  increased during the study from an average of 610 to
1,735 mg/1.

     After a suitable  period of bioacclimation, more in-depth sample analyses
were performed, and Table 12 presents certain results thereof.

     All bioreactors were degrading certain constituents in the wastewater, as
evidenced  by  both COD  removal and  oxygen  uptake rate  measurements.   On the
other hand,  the pentachlorophenol  data indicate  a  lesser amount of removal.
The  predominant removal mechanism  seems to be adsorption  on the biofloc and
not  biodegradation  since  total  pentachlorophenol  levels  prior to  settling
remained constant.   This phenomenon  is evidenced by the  increasing effluent
concentrations  with  respect  to time; decreasing the mass of pentachlorophenol
in the  bioreactor  feed tended only to postpone the time for saturation of the
biomass, and not prevention  of its occurrence.

Chemical Coagulation

     A  number  of jar tests  were-undertaken using  alum, ferric chloride, cati-
onic  polymers, and anionic  polymers.   Following  settling  the supernatant was
filtered through Whatman No.  41 paper.   These chemical coagulation studies did
not  lead to  any measurable  reduction in the pentachlorophenol content of this
wastewater.

Extraction

     At this  facility,  pentachlorophenol is dissolved in  an  85:10  mixture of
No.  2 fuel oil and still bottoms  from  the  production of amyl alcohol.  Batch
solvent extraction  studies  using No. 2  fuel oil  alone,  and the 85:10 mixture
were  undertaken.   Wasterextractant  volumes  were  constant  at  4:1,  and  the
results for two sets of tests  are presented in Table  13.
                                     -21-

-------
                     TABLE 12.  BIOLOGICAL TREATMENT RESULTS

Reactor/
date
Reactor A
5/28
5/29
5/31
6/2
6/4
6/5
Reactor B
5/28
5/29
5/31
6/2
6/4
6/5
Reactor C
5/28
5/29
5/31
6/2
6/4
6/5
0 Uptake
(mg/l/hr)

11.0
6.0
20.0
48.0
33.6
22.0

10.8
6.0
31.5
27.0
30.0
32.0

16. 0
9.6
20.0
22.0
23.0
23.0
0 (mg)
hr/gr/MLVSS

12.6
5.5
13.6
33.6
18.8
14.9

39.3
5.7
22.5
18.8
17.2
17.3

23.5
9.1
12.0
13.3
12.0
12.2
MLVSS*
(mg/1)

870
1,100
1,470
1,430
1,790
1,480

275
1,060
1,400
1,440
1,740
1,850

680
1,050
1,660
1,660
1,910
1,880
COD, (mg/1)
Raw

3,920
3,475
3,275
4,055
N.A.
3,815

1,080
2,780
2,620
3,245
N.A.
3,050

1,405
2,085
1,965
2,435
N.A.
2,290
Final

755
1,080
1,325
N.A.
1,970
N.A.

1,030
1,080
1,365
N.A.
950
N.A.

975
920
1,000
N.A.
1,075
N.A.
Pentachlorophenol
(mg/1)
Raw

37
30
N.A.
N.A.
30
35

24
24
N.A.
N.A.
24
28

35
30
N.A.
N.A.
18
21
Final

0.27
0.15
0.71
18.0
26.0
25.0

0.58
0.95
0.83
0.46
6.60
13.0

0.19
0.25
0.26
0.60
4.50
8.90
N.A. - not analyzed
*Mixed liquor volatile suspended solids.
                                     -22-

-------
                      TABLE 13.  SOLVENT EXTRACTION RESULTS
                                   Pentachlorophenol, mg/1
Extractant                    Untreated      Treated        % Removal
No. 2 Fuel oil
85 : 10 mixture
45
45
1.2
0.3
97
99

     The 85:10 mixture of No. 2 fuel oil and amyl alcohol still bottoms yielded
a higher level of treatment than No. 2 fuel oil alone.

Adsorption

     Resin  adsorption studies were  also undertaken using Rohm  and Haas Com-
pany's XAD-4  nonionic resin,  using a Phipps and Bird jar test apparatus.  The
XAD-4 resin was chosen for the southern facility as compared to the XAD-2 used
at  the west coast facility.  The XAD-4  has  comparable porosity and capacity,
but  the  pore  diameter is smaller.  The investigators thought that there would
be some selectivity towards pentachlorophenol with the XAD-4 and thus increase
the  capacity.   Initial  investigations studied the addition  of  the resin dir-
ectly  to the  wastewater.  No apparent  reduction  in pentachlorophenol concen-
tration  occurred  after  1 hour of reaction time in the presence of 15 grams of
resin per liter of wastewater.  Subsequent studies were conducted in which the
resin  was  added at 10 g/1  and  the reaction time was  varied.   The results of
this study are presented in Table 14.
                                     -23-

-------
                    TABLE 14.  XAD-4 RESIN ADSORPTION RESULTS

Run no.
1
2
3
4
5
6
Contact time
(hrs)
24
24
23
96
20
44
Pentachlorophenol (mg/1)^
Untreated
70
35
58
58
17
17
Treated
1.4
1.9
2.0
9.0
12.0
5.2
% Removal
98
95
97
84
29
69

     In  all  cases  except Run No. 5, removal of  pentachlorophenol was 69 per-
cent  or  greater.   Regression  analysis  showed  no significant  relationship
between  either  contact  time  and percent removal  or input  concentration and
percent  removal.

     Adsorption of pentachlorophenol in the  wastewater onto bentonite clay was
also investigated.  Before  the  pH of the wastewater was  adjusted,  the bento-
nite stayed  in  suspension,  and  therefore this method of  individual treatment
was not pursued.

Alum/Polymer Coagulation

     Limited coagulation trials  were conducted  using combinations of alum and
polymers,  followed  by solids separation.   These tests were  conducted  at the
investigators'  suggestion,  following discussions  with chemical suppliers after
the studies  at  a west coast  mill.   This  type  of treatment provided  less re-
moval  than  other  techniques and,  therefore subsequent  investigations  were
discontinued.

Acid Addition/Polymer Coagulation

     As  was  noted,  the  addition of acid to the  raw  wastewater induced the
formation  of colloidal  material which  had an  associated  pentachlorophenol
content.   Attempts to  coagulate  this colloidal material with an anionic poly-
mer had  limited  success.   A 46-percent pentachlorophenol reduction was real-
ized,  from  70  to 38  mg/1,  when 4 mg/1  of  polymer was added to  a  wastewater
                                     -24-

-------
sample at a pH of 4.0.  Only a pin floe could be developed, and the floe had a
tendency to float rather than settle.

Acid Addition/Extraction

     Extraction  of  samples subjected  to  acid  addition  pretreatment  was also
investigated.  The  methodologies and  extractant materials  for  these studies
were  as  previously  described.   Table  15  presents  typical  results  of this
combined treatment technique.

	TABLE 15.  ACID ADDITION/EXTRACTION RESULTS	

                                   	Pentachlorophenol, mg/1	
Extractant               pH        Untreated      Treated        % Removal
No. 2 Fuel oil
85:10 mixture
4
4
45
45
0.5
0.2
99
>99
No. 2 fuel oil cosolvent
     Comparison of these data with those presented in Table 13 indicate only a
slight  increase  in pentachlorophenol removal over that achieved by extraction
alone.

Acid Addition/Adsorption

     A  limited number of trials were conducted in which acid pretreated waste-
water was  also  subjected to XAD-4 adsorption.   A  pH of 4 in the pretreatment
step led  to'a final effluent pentachlorophenol  concentration ranging from 0.4
to  1.1  mg/1,  independent of the  concentration present  in the raw wastewater.

     A  combination of acid pretreatment/bentonite clay adsorption also yielded
positive  results  since  the lower pH of the waste not only allowed for adsorp-
tion of the pentachlorophenol onto the clay, but  also, the subsequent gravity
phase  separation of the  clay from  the  wastewater.   Results  typical of this
type of treatment are presented in Table 16.
                                     -25-

-------
               TABLE 16.  ACID ADDITION/BENTONITE CLAY ADSORPTION
                                          Pentachlorophenol, mg/1
Trial no.                          Untreated      Treated        % Removal
1 26
2 33
3 35
4 50
5 50
6 50
7 58
2.6
4.1
3.4
7.0
5.0
6.3
5.0
90
88
90
86
90
87
91
     The above results were  achieved  at optimum conditions of a pH of 4.0 and
a bentonite clay  concentration  of  2 g/1.   At these conditions, rapid settling
occurred.   Reductions in pentachlorophenol  concentration  following  solids
removal averaged  85 to 90 percent  in 7 trials.   Acidification, bentonite clay
adsorption, solids  removal and subsequent  adsorption on XAD-4  resin reduced
pentachlorophenol levels   to  less than 1 mg/1 in.13 trials,  as reported subse-
quently .

Coagulation/Adsorption

     XAD-4 resin  adsorption  was also  tried in conjunction with alum and poly-
mer coagulation,   with limited success.   The minimum pentachlorophenol concen-
tration obtained  with this combination of treatment techniques was  1.7 mg/1.
This was deemed inadequate and  further investigations in this  area  were dis-
continued.

Acid Addition/Coagulation/Adsorption

     One study was  completed which involved acid  addition to  a  pH  of  4.0,
coagulation with  4  mg/1   of  an  anionic polymer,  and XAD-4 adsorption of the
supernatant.   Raw waste  pentachlorophenol  concentration  was  70 mg/1  and the
final treated  effluent, 1.7 mg/1.   Resin dose approximated 10  g/1 and reaction
time,  24 hours.
                                     -26-

-------
Acid Addition/Bentonite Adsorption/XAD-4 Adsorption

     To  insure  a high  level of  treatment,  studies involving  acid addition,
bentonite clay  adsorption,  and finally XAD-4 adsorption were undertaken.  The
optimum conditions established were:

                1.   pH adjustment to 4.0 with sulfuric acid.

                2.   Bentonite addition at 2 g/1;

                    a.   30  seconds rapid mix;

                    b.   30  minutes flocculation;

                    c.   60  minutes settling; and

                    d.   decant supernatant.

                3.   XAD-4 addition at 10 g/1;

                    a.   24  hour mixing;

                    b.   5 minutes settling; and

                    c.   decant supernatant.

     Table  17  gives  results  of  this  three-phased treatment  approach.  The
average  effluent concentration from  this process  was  0.21 mg/1 pentachloro-
phenol.

     Further  work on  this process indicated that the XAD-4  resin dose could be
reduced  to about 3 g/1 without  substantially altering  the  effluent quality
reported above.

XAD-4 Resin Regeneration

     Spent XAD-4 resin was found  to be effectively  regenerated using the 85:10
mixture  of No.  2 fuel oil and  amyl alcohol still bottoms  used at this facility
for  dissolving  pentachlorophenol prior  to  its  use in  the  wood preserving
process.   Regeneration  was   accomplished using  a 1:4 regenerant to  resin volu-
metric  ratio  and a reaction time of  one hour.  Table 18 indicates the  levels
of regeneration' achieved.
                                      -27-

-------
TABLE 17.   ACID ADDITION/BENTONITE ADSORPTION/XAD-4 ADSORPTION
Pentachlorophenol, mg/1
Trial no.
1
2
3
4
5
6
7
8
9
10
11
12
13
Untreated
33
50
50
50
58
58
58
58
58
10
58
35
26
Treated
0.20
0.08
0.05
0.22
0.06
0.36
0.75
0.06
0.05
0.10
0.06
0.55
0.17
% Removal
>99.
>99.
>99.
>99.
>99.
• >99.
>98.
>99.
>99.
>99.
>99.
>98.
>99.
                            -28-

-------
                       TABLE 18.   XAD-4 RESIN REGENERATION

Trial No.
1
2
3
4
5
6
Pentachlorophenol

Adsorbed, g/g Recovered, g/g
2.2
2.2
2.2
2.2
2.2
2.2
1.6
1.5
1.5
1.3
1.6
1.3
Regeneration
Efficiency (%)
73
68
68
59
73
59

     Of  the  2.2  grams of  pentachlorophenol  adsorbed per  gram o.f  resin,  an
average of 62  percent could be recovered and  reused,  if desired,  in the wood
preserving process.
                                     -29-

-------
                                    SECTION  6

                       SELECTION OF  TREATMENT ALTERNATIVES

     It must be emphasized that the characteristics  of  wood  preserving waste-
waters are extremely variable.   With such variance, it is difficult to identi-
fy one  specific  treatment process  which can be applied uniformly  across  the
industry.   Two  batch treatment  processes,  however,  evolved from  this  study
which  were  capable  of,  consistently yielding  a highly  treated  wastewater.

     To allow for  optimum removal of pollutants,  the  wastewaters  from the wood
preserving processes must first  be  subjected to:   1) free  oil  separation in a
Lamella-type  separator;  and 2)  wastewater flow  equalization.  For the purposes
of estimating  capital  and operating  costs,  wastewaters from  a typical  wood
preserving facility were  approximated and are presented  in  Table  19.

     TABLE 19.   TYPICAL WOOD  PRESERVING  FACILITY WASTEWATER CHARACTERISTICS


               Volume                        10,000 gpd
               Temperature                   20  to 25°C

               pH                            5.0 to 5.5
               Penta chlo ropheno1
                    Average                   3.6 Ib/day (43  mg/1)
                    Peak                     10.7 Ib/day (129 mg/1)
     The following  subsections  describe the  two recommended  batch treatment
processes and the associated capital  and operating costs.

Proposed Process No. 1:   Acid Addition/Bentonite Adsorption/XAD-4 Adsorption

     Proposed batch treatment process no.  1  involves the  following operations
following free oil separation and wastewater  flow equalization:

     1.   Wastewater pH adjustment to 4.0+ using sulfuric acid.   The require-
          ment approximates 0.4  gal  of  66° Baume sulfuric  acid  per 1,000 gal
          of wastewater.
                                     •30-

-------
     2.    The addition  of  15  to  20  Ib of  bentonite clay  per  1,000 gal  of
          wastewater.   Maintain  suspension  of the clay in the wastewater  for
          30 minutes with mixing; then allow gravity separation to occur  for 1
          hour.   Remove  clay sludge from bottom of reactor by pumping.

     3.    The addition of 25 Ib of XAD-4 resin to the acidic  bentonite  treated
          wastewater.   Maintain  resin  in  suspension for  12 hr with  mixing;
          then pass wastewater-resin  mixture  over a sidehill screen for  resin
          separation.

     4.    Neutralize, the wastewater to  a neutral pH value prior  to discharge
          to a POTW for further treatment.

     The  resin,  after sidehill separation, is  approximately 50-percent  water
(w/w) and will  require  drying  before regeneration.  This  can be  accomplished
by using  a  vibratory  fluid bed-type dryer.   After  drying, the resin is  to be
regenerated using  a  1:4  resin-to-regenerant ratio (v/v),  the regenerant  being
a mixture of No. 2 fuel oil and a co-solvent such as amyl  alcohol  still bottoms
After one hour of  mixing  the  resin/regenerant mixture should be  conveyed to
the  sidehill  screen for  resin separation.    The  resin is to  be  stored  until
required  for  reuse while  the  spent  regenerant  will be returned  to the wood
preserving process  for either  reuse or disposal.. This system is  graphically
depicted in Figure 2.

     Table  20  presents  estimated construction  and  operating costs (September
1979) for the  10,000  gpd treatment system utilizing  acid  addition, bentonite
adsorption, and XAD-4 resin adsorption processes.  Based  on  an equipment life
of 20 years  and an interest rate of 10 percent,,the equivalent uniform annual
cost was.calculated to be approximately $20,000.

Proposed Process No. 2:   Acid Addition/Extraction

     Proposed batch treatment  process No.  2 involves the  following operations
following free oil separation and wastewater flow equalization.

     1.    Wastewater pH  adjustment to  4.0+  0.1  with  sulfuric acid,  require-
          ments approximating 0.4 gal of 66° Baume sulfuric acid per 1,000 gal
          of wastewater.
                                     -31-

-------
                                  COAGULATION/ADSORPTION  PROCESS FLOW DIAGRAM
LO
NJ
                              OIL
                          SEPARATION
                                                                      COAGULATION
                                                                           a
                                                                      ADSORPTION
                                                                                   BENT.
                                                                HgSO,
NoOH
                                         BENTONITE
                                         SLUDGE
                                                                                                                       REGENERANT
                                                                   SPENT
                                                                   REGENEKA
                                                                                HYDRO- SEIVE
                                                                                                      RESIN
                                                                                                  REGENERATION
                                                                                                               FIGURE 2

-------
                TABLE 20.   PROPOSED PROCESS NO.  1 ESTIMATED COSTS
               Construction costs

                    Site work                               $  2,000
                    Mechanical                                81,100
                    Structural                                43,200
                    Electrical                                 8,100
                    Engineering, administration,
                      contingencies                           33,600

                                   Total                    $168,000

               Annual operating expenses

                    Amortization                            $ 20,000
                    Labor                                     14,000
                    Power*                                       800
                    Chemicals                                  5,200

                                             Total          $ 40,000/yr
*Power - $0.05/kWh
     2.   The addition of 250 gal of extractant per 1,000 gal of.wastewater,  a
          typical  extractant  being No. 2  fuel  oil and  a co-solvent  such as
          amyl alcohol  still  bottoms.   Mix the wastewater and  the  extractant
          for one  hour; then  allow phase separation  to occur for  about two
          hours.

     3.   Decant the  spent  extractant  and return to the  wood preserving pro-
          cess for either reuse or disposal.

     4.   Neutralize the treated wastewater to a neutral pH .prior to discharge
          to a POTW for further treatment.

     Table  21  presents estimated construction and  operating costs  (September
1979) for a 10,000 gpd treatment system utilizing acid addition and extraction
processes.  Based  on  an equipment life of 20 years and an interest rate of 10
percent, the equivalent uniform annual cost of this  system  was calculated to
be approximately $24,000.
                                     -33-

-------
                TABLE 21.  PROPOSED PROCESS NO. 2 ESTIMATED COSTS
               Construction costs

                    Site work                               $  1,600
                    Mechanical                                57,300
                    Structural                                34,200
                    Electrical                                 8,500
                    Engineering, administration, &
                      contingencies                           25,400

                                             Total          $127,000

               Annual operating expenses

                    Amortization                            $ 15,000
                    Labor                                      5,300
                    Power*                                       300
                    Chemicals                                  3,000

                                             Total          $ 23,600
*Power - $0.05/kWh
**Chemical cost for fuel oil required above current use only.
                                     -34-

-------
             SOLVENT  EXTRACTION  WITH OIL SEPARATION

                             PROCESS  FLOW  DIAGRAM
   OIL
SEPARATION
                                          SOLVENT
                                         EXTRACTION
                     FLOW
                 EQUALIZATION
                               -9-
                SOLVENT
SOLVENT
              -M-
                                      SOLVENT      HtS°4
                                                    •M-
                                                     -M-
                                                                      H2S04
                                                               AUX. RECYCLE
                                             No OH
                                                                                           NoOH
                                                                                        TREATED
                                                                                        EFFLUENT
                                                                                  FIGURE 3

-------
                                  REFERENCES
1.    National Resources Defense  Council et al.  versus Russell Train,  United
     States District Court  for  the District of Columbia,  (8 ERC  2120),  June 7,
     1976.

2.    Ingols, R.S.,  et  al.   "Biological activity of Halophenols."   Journal of
     Water  Pollution Control Federation, 45(2):359-364,  1973.

3.    Development Document for Effluent  Limitations  Guidelines  and  Standards
     for the Timber Products Processing Point Source Category.   EPA  440/1-79/
     0230,  U.S.  Environmental  Protection  Agency,  Washington,  D.C.,  1979.  427
     pp.
                                     -36-

-------
                                 APPENDIX A
             REVIEW OF LITERATURE ON PHENOLIC TREATMENT TECHNOLOGIES

1.0  BIOLOGICAL OXIDATION

     Numerous investigators have  demonstrated  the feasibility of biologically
oxidizing phenols (1, 2, 3, 4, 5).  In the laboratory, Radhakrishnan and Sinha
Ray  (3)  found that  the  bacteria  B.  Cereus is capable  of  completely metabol-
izing phenol  at a  calculated detention time  of 26  hours.   The  system para-
meters developed for continuous flow cultures were:
          Heterogeneous metabolism rate   0.0022 mg phenol/hr/mg bacteria
          True yield coefficient
81.8%
          Maximum growth rate @40°C
0.628/hr
          Minimum N to phenol ratio
1:10
Kirsch and Etzel (1) had similar success when biodegrading pentachlorophenate.
They found  that  up to 68 percent  of  the  radioactive pentachlorophenate added
to  an  acclimated  culture  was  recovered  as  radioactive carbon  dioxide  in 24
hours.

     Kostenbader  and Flecksteiner  (2)  have  reported  on the  Bethlehem  Steel
Corporation's  experience  with biological oxidation  of  coke  plant wastewaters
at  their  Bethlehem,  Pennsylvania facility.   Their treatment  process  is  shown
schematically on Figure 4. The system was first put on line in September 1962,
and  it  was learned  that  existing  aeration  capabilities limited  the plant's
capacity to 2,700  Ib of phenol/day.  As a result, they increased the aeration
capacity  in 1964,  which  in turn increased the plant's  treatment capacity to
greater than  4,000  Ib  of phenol/day.  For  the  next  2-1/2  years,  the  plant
                                     -37-

-------
processed an average daily  flow  of 112,000 gallons at  a  phenol concentration
of 1,390  mg/1  (1,300 Ib  phenol/day).   The  system  consistently removed  99.9
percent of  the influent phenol  which resulted  in  less than 0.1 mg/1  in the
effluent.    It  should be noted, however, that  the plant was operating  at ap-
proximately 35  percent of its design phenol removal  capacity.

     Miller (4) reported similar  results  when treating wastes  from  a cross-tie
creosoting  operation.   The  treatment scheme  for this  facility,  as  shown on
Figure 5,  consisted  of nutrient  addition, aeration, clarification,  and  land
application.  Actual plant  operating  data  are  shown in Table 22.   Evaluation
of  this  data  clearly   shows  that 99 percent  phenol reductions consistently
achieved by biological treatment.

     There was  also one report of a system  treating  a  combination of municipal
sewage and  a herbicide waste  high in BOD  and chlorophenols.(6)  This  system
was unique  in  that a completely  mixed aerated  lagoon was  installed  between an
existing conventional activated sludge treatment  facility  and  existing stabil-
ization ponds.   The  purpose of  the aerated  lagoon was  to  avoid  hydraulic
overloading  of the  conventional  plant  by accepting all  raw  wastewaters in
excess of the  conventional  plant's design  capacity and reduce  the  BOD to the
stabilization ponds.  The conclusions of this  study were  essentially that the
organisms present  in domestic sewage will remove  complex  phenolic  compounds
when  sufficient  biomass and nutrients  are available.  The phenolic removals
actually achieved are shown graphically  on  Figures 6 through 10.  It is inter-
esting to note that  the joint treatment of  chlorophenolic  waste and domestic
sewage was  reported  to produce  essentially the  same biological  data  as  a
system treating only domestic sewage.

     The  chemical  industry  has  also had experience with biological  oxidation
of phenols  as  evidenced by the  work of Capestany et  al(7)  Wastewater  from
this facility was found to contain 1,000  mg/1 of  phenol.  After  reviewing  past
work  concerning  phenol  removal,   Capestany concluded that such wastes  can be
treated  biologically  with   an  acclimated,  nutrient-supplemented,   activated
sludge system.   The  same conclusion  was  reached by  Keith, (8) based  on his
work with two Georgia Kraft container board mills.

     Capestany's  'initial  pilot plant work found that  adequate removals  were
obtained only when phenol  feed concentrations  were less than 250  mg/1.  Sub-
sequent  investigations,  however,  revealed that  sulfate  deficiencies  were
limiting  the  biological  oxidation process.   The  pilot  plant  was   then  re-
started using the  process parameters  shown in Table  23.  The  effluent phenol
concentration immediately  dropped below 1.0  mg/1 and  remained there for the
duration  of the  pilot  study.  As a  result of  this  success,   the  full scale
treatment facility shown  on Figure  11  was constructed  in 1975.    The plant
design parameters  are. given in Table 24.  Since  going  on  line, this facility
has had  great  difficulty in maintaining adequate mixing as evidenced  by the
changing  of aerator  impellers, the  installation of draft tubes, and the in-
stallation of a supplemental compressed  air mixing system.   Even so,  the plant
                                     -38-

-------
  BIOLOGICAL  OXIDATION   PROCESS  SCHEMATIC

              FOR  COKE   PLANT  WASTES  (2)
        WAL
   STEAM
              FRESHWAL
   PHOSPHORIC
      AGIO
BPW
           DILUTION WATER
                                                         WASTE SLUDGE TO
                                                         SEWAGE PLANT
ACTIVATED SLUDGE
AERATION  SYSTEM
     260,000 go I.
     35,000 cu. ft.
                           OXYGEN FROM SURFACE AERATORS
                                                         EFFLUENT TO STORM SEWER
                                                                       FIGURE 4

-------
BIOLOGICAL OXIDATION PROCESS SCHEMATIC
FOR WOOD PRESERVING WASTES (4)
A VAPOR DRYING OUf>LEX AERATION
... ^ . EFFLUENT TANK 8 CLARIFIER

DECANTER S^A'1
"^S >»-
^s^ EQUALIZING Q
	 • TANK *"
^f
B CREOSOTE / B -1
*" DECANTER (4 -DAYS RETENTION) I
- -- EXCESS '

C SEPTIC C-1 CHLORINE N8P
TANK CELL SOLUTION
' A-1 B-1 C-1 D
LAGOONS
. ^ 1/4 ACRE
>. I/5AC"F ^

^^" 1/4 ACRE
SLUDGE T
IRRIGATION
8 SOAK FIELD
(NO RUNOFF)
E F
FLOW l0'000 2'000 l'000 13,000 13,000 
-------
TABLE 22.  OPERATING RESULTS, BIOLOGICAL TREATMENT OF WOOD
                   PRESERVING WASTES (4)

Influent Aeration tank Clarifiers Lagoon
Date of
startup
3/72
4/72
-5/72
6/72
8/72
; 2/73
3/73
5/73
9/73
11/73
Settleable Pro-
COD Phenol pH solids tozoa
2150 145 6.6 - 0
2900 - 7.0 - Yes
7.2
7.0
1850 140 7.3
1600 140 6.9 600 "
6.5
1100 80 . 7.0 - "
3200 300 7.2 900 "
6.1
Total
organic
carbon Phenol Phenol
80
80 0.7. 0.2
70 0.4 0.3
70 0.1 0.1
55 <0.1 <0.1
<.05 <.02
60
<0.5
<1,0
33
COD Phenol'
90
65
130
45 0.03
-
40 0.004
-
35
65
-


-------
200
150
100
REMOVAL OF 2,4-DICHLOROPHHNOL AND 2,4-DICHLORO-
 PHENOXYACETIC ACID FROM SOLUTION IN  AERATION
    BASIN EFFLUENT BY CONTINUOUS AERATION (6 )
                       Initial Conditions;

                        64  mg/1 2,4-DCP
                       174  mg/1 2,4-D Acid
                       Aeration Lagoon Effluent

                       pH of acid-phenol mixture
                        adjusced to 7.00 just before
                               .with effluent

                       Temperature: 20-21°C.

                       Constant slow stream of air
                        bubbled through mixture.

                       Control samples same as above
                        without Aeration Lagoon    	
                        Effluent.   Only distilled
                        water as solvent.
 50
        2.4-DCP Concentration
                                    2,4-D Acid Concentration
                                    Distilled Water Control
                           6    7    8    9   10
                           Time  In Days
                                                FIGURE 6
                                 42

-------
                                           i
200
REMOVAL OF 2 , 6-DICHLOROPHENOL AKD 2,6-DICHLORO-

 PHENOXYACETIC ACID FROM SOLUTION IN AERATION

    BASIN EFFLUENT BY CONTINUOUS AERATION  (6 )
150
     2,6-D Acid Concentration
100
                      Initial Conditions;

                       64 mg/1 2,6-DCP
                      178 mg/l 2,6-D Acid

                      Aeration Lagoon Effluent

                      pH of acid-phenol mixture
                       adjusted to 7.00 just before
                       mixing with effluent.
                                  Temperature:   20-21°C.
                      Constant stream of  air
                       bubbled through mixture.

                      Control samples same  as  above
                       without Aeration Lagoon
                       Effluent.  Only distilled
                       water as solvent.
        Distilled Water Control
 50
                    2,6-DCP  Concentration
           J	1
                                   J	L
                                                J	LJ
                                789

                              Time in Days
                                           10  11
                                       II1
                                                    ^  13  14  15
                                                  FIGURE  7
                                  43

-------
mg/1


 60
REMOVAL OF 2,4,5-TRICHLOROPHENOL AND 2,4,5-TRI-
   CHLOROPHENOXYACETIC ACID FROM SOLUTION IN
   AERATION BASIN   EFFLUENT   BY   CONTINUOUS

                   AERATION  (s )

             Initial Conditions:

             50 mg/1 2,4,5-T Acid
             .1.8. 8 mg/1 2,4, 5-TCP

             Aeration Basin Effluent

             pH of acid-phenol mixture adjusted to
              7.00 just before nixing with effluent.
             Temperature: 20-21CC.

             Constant slow stream of air bubbled.
              through mixture.

             Control samples sane as above without
              Aeration Basin Effluent.  Only
              distilled water as solvent.
 50
 40
 30
 20
 10
        I    I    I    I    I    I  V I  J, I    I   I
                                         Distilled Water Control
                                  2,4,5-T Acid Concentration
                                         Distilled Water Control
        123456789  10
                             Time in Days
                                                  FIGURE  8
                                   44

-------
mg/1

  70




  60




  50




  40




  30




  20




  10
       REMOVAL OF 2,4,6-TRICHLOROPHENOL AND

2,4,6-TRICHLOROPHENOXYACETIC ACID FROM SOLUTION IN

  AERATION BASIN EFFLUENT BY CONTINUOUS AERATION (6)


                          Initial Conditions:

                          18.5mg/l 2,4,6-TC?
                          53.0mg/l 2,4,6-T Acid
                          Aeration Lagoon Effluent

                          pH of acid-phenol mixture
                           adjusted to 7.00 just before
                           mixing with effluent.
                          Temperature: 20-21°C.
                          Constant slow stream of air
                           bubbled through mixture.

                          Control samples same as above
                           without Aeration Lagoon
                           Effluent,  'only distilled
                           water as solvent.
2,4,6-T Acid Concentration
    Distilled Water Control
                    V.U11U.LU-L «^



                   *— <*-	^
                   2,4,6-TCP Concentration
            2   3   4   5   6   7   8   9  10   11   12   13   14  15
                               Time in  Days
                                                   FIGURE 9
                                   45

-------
90 H
          CHANGE IN PENTACHLOROPHENOL CONCENTRATION IN'

         AERATED SOLUTIONS IN AERATION LAGOON   EFFLUENT ( 6 )
               Experiment 2
          Experiment 3
                Experiment
                          5678
                          Time in Days
11  12
                                                FIGURE 10
                                 46

-------
consistently removed 99  percent  of the influent phenol  and 97 percent of the
influent BOD.

     Halladay  et  al   (9)  extensively  investigated  the phenolic  conversion
capabilities of 3 types of bioreactors:

     1.   continuously stirred (CSTBR);

     2.   packed bed (PBBR);  and

     3.   fluidized bed (FBBR).

The results  of  these  investigations are given  in.  Table  25.  Evaluating these
results led to the following conclusions:

     A.   The  CSTBR can  treat the highest  influent feed  concentrations,  is
          relatively easy  to operate,  and the retention time  may  be varied.
          It  does,  however,  require   the  largest  volume,  is  susceptible  to
          shocks and washouts, and is  slow to recover from upsets.

     B.   The PBBR has higher degradation rates and lower retention times than
          the CSTBR, and recovers quickly from shock loads.   It does, however,
          develop  excess  biomass which  tends  to  stop  flow when the biomass
          sloughs.

     C.   The FBBR has degradation rates  and  retention times  similar  to the
          PBBR  and has  low  pressure  drop  characteristics.   The  FBBR does,
          however, yield poor results  with compounds requiring long retention
          times.  Solid-liquid disengagement is also difficult.

It  should  be noted  that the  lowest  effluent phenol  concentrations  were ob-
tained with the FBBR type bioreactor.

     Rotating biological  contactors (RBC's) serve  as yet  another alternative
for the biological  treatment of pulp and  paper mill wastes.   Egh and Mueller
(10) reported on a laboratory study whereby bleached kraft mill effluents were
tested  at  retention times varying  between 2 and  16 hours.   BOD_  removals  in
all cases  exceeded 90 percent.   In addition,  all  acute toxicity to  fish was
removed at retention times as short as two hours.

     Ingols  et  al(ll), in  their work with  halaphenols,  demonstrated that  as
phenol  becomes  more  substituted,  there  is an increased  resistance  to bio-
degradation.  Figure 12  (7)  diagrams  how  this  resistance increases with both
the level  of substitution  and position of  the chlorine atom.   His results,
however, showed  that in  the absence  of other  organic  sources low  concentra-
tions  of  the simpler  compounds   can be  degraded  biologically in  one  to ten
days.
                                     -47-

-------
  TABLE 23.  BIOLOGICAL PILOT PLANT PROCESS PARAMETERS (7)
Raw waste characteristics




     Phenol




     BOD




     COD




     pH




Nutrient ratios




     BOD:N:S04




Process parameters




     Retention time




     Volumetric loading




     MLSS




     SVI




     Sludge loading




     Sludge production
1,000 mg/1




4,000 mg/1




6,000 mg/1




8.5








100:5:5








24 hr




3.92 Kg BOD/m3




6,050 mg/1




160




0.6 g BOD/day/g MLSS




0.3 g MLSS/day/g BOD removed
                           -48-

-------
         SCHEMATIC   OF  FULL  SCALE BIOLOGICAL

           TREATMENT    FACILITY OF  CAPESTANY(7)
WASTE
   NH
                                                                       AERATOR
                                                                       94,00090!.
                                                                           DIGESTOR
                                                                           206,000gol.
                                                                           TO DRYING BEDS
                                                                               FIGURE  II

-------
              TABLE 24.  FULL SCALE (BIOLOGICAL) TREATMENT FACILITY
                           DESIGN PARAMETERS (7) 	
General

Flow

Influent BOD

Influent phenol

Aeration basin

Volume

Detention time

Oxygen requirements
(@5°C & 2 mg/1 residual)

Energy requirements

Clarifier

Overflow rate

Solids loading

Digester

Volume

Sludge age

Energy requirements
 72,000 gpd

  3,109 Ib/day

  1,943 ib/day



 93,000 gal

     31 hr

    136 Ib/hr


  42/75 hp
    360 gal/ft

     16 lb/ft2
200,000 gal

     20 days

     15 hp
272,521 I/day

  1,410 kg/day

    881 kg/day



352,006 1



   61.7 kg/hr


31.3/55.9 kW



 14,656 1/m2

     78 kg/m2



757,000 1



   11.2 kW
                                     -50-

-------
              TABLE 25.   COMPARISON OF THE  GENERAL CHARACTERISTICS OF THREE BIOREACTORS  (9)
            Conditions
     CSTBR
     PBBR
     FBBR
  Maximum phenol degradation  rate
  for  C. = 500 mg/liter,  0.99
  conversion
l.Og of phenol/day/1
bioreactor
volume
4.7g of phenol/day/1
bioreactor
volume
8.5g of phenol/day/1
bioreactor
volume
  Maximum phenol  degradation  rate
  0.99  conversion of  any  feed
  Maximum phenol  degradation  rate
  any  conversion
  Normal  effluent  phenol  cancan-
s'1 tration at  maximum conversion
2.67g/l d
C. = 1,400 mg/1
Q* = 300 ml/h
 J_i

2.67 g/1 d
C. = 1,400 mg/1
Q* = 300 ml/h
 J_i

0.25-1.00 mg/1
4.7 g/1 d
C. = 500 mg/1
Q^ = 875 ml/h
 Lt

6.0 g/1 d
C. =800 mg/1
Q* = 700 ml/h
 L

0.25-1.00 mg/1
11.2 g/1 d
C. = 260 mg/1
Q^ = 18,000 ml/h
 Li

21.2 g/1 d
C. = 240 mg/1
Q* = 42,000 ml/h
 Li

0.01-0.50 mg/1
   TOC  reduction
   for  0.99  conversion
0.90
0.90
 0.95
  Retention time necessary for
  sizeable  thiocyanate  conversion
20 h
NDa
ND
   Highest C.  successfully degraded

   Resistance  to step hydraulic
   shocks
   Recovery from step hydraulic
   shocks
1,400 mg/1
Poor
Slow
1-5 days
850 mg/1
Good
Fast
24 h
2200 mg/1
Fair
Fastest
24 h
   Resistance to step organic
   carbon shocks
Poor
Good
Good

-------

Conditions
Recovery from step carbon
shocks
Facility for aeration
Compatibility with degassing
TABLE 25 (cont'd)
CSTBR
Slow
1-5 days
Good
Good

PBBR
Fast
24-48 h
Fair
Fair

FBBR
Fastest
24 h
Good
Good

 Not determined
C. = Feed concentration in mg/1




Q_ = Liquid feed rate in ml/h

-------
     Preliminary  investigations  by  Ashmore et  al  (12)  also  suggested  that
certain  dihydric  phenols  are resistant  to bio-oxidation.   This  conclusion,
however, was based on intermittent phenol additions.   When a study of continu-
ous treatment  under steady  state  conditions was  made,  "none  of  the phenols
investigated showed undue  resistance to attack."  Table 26  shows  the results
of  this study.   It will  be noted  that  the degree  of  purification  of  both
monohydric and dihydric phenols was normally greater than 95 percent.  Another
interesting  finding was  that high phenol  concentrations  cause  filamentous
sludge  bulking.    Ammonium  chloride was  found  to  suppress the  filamentous
growth.
                                     -53-

-------
PHENOLIC  RESISTANCE

 TO  BIODEGRADATION (II)
      PHENOL
      0-CHLOROPHENOL
      2,4 DJ CHLORO PHENOL
      2,4,6 TRICHLOROPHENOL
      PENTACHLOROPHENOL
                                    EC
                                    O
                                    Ul
                                    a
                                    Q
                                    Ul
                                    
-------
              TABLE 26.   THE BIOCHEMICAL OXIDATION OF PURE PHENOLS
             	BY THE ACTIVATED SLUDGE PROCESS (12)	
                                           Permanganate value (mg/1)
Compounds used
       Amounts in
     influent(mg/1)
Influent
Effluent   Percent
  mean   purification
Phenol
0-Cresol
m-Cresol
p-Cresol
1,680
1,120
1,120
1,120
1,340
1,625
1,580
1,535
17
10
8.4
6.9
99.4
99.4
99.5
99.5
Equal parts of
 three creso'ls

Catechol
4 Me-catechol
3 Me-catechol
Resorcinol
Phenol
0-Cresol
m-Cresol
Catechol
3 Me-catchol
4 Me-catechol
Resorcinol
4 Me-resorcinol
5 Me-resorcinol
Quinol

Total
As above
          1,120

          1,400
          1,400
            840
          1,400

Synthetic spent liquor
          56.0
           4.2
           4.2
         231.0
          94.8
          78.8
         179.2
          16.1
          16.1
          19.6

         700.0

Synthetic spent liquor
         700.0
  1,580

  1,875
  1,570
    950
  2,325
    9.5

   21.7
   55.4
  120
   18.4
99.4

98.8
96.5
87.4
99.2
  kept in neutral solution
  105
    6.1
    5.9
  310
  107
   88.3
  298
   20.6
   19.6
   30.8
  990
    9.4
99.1
  kept in alkaline solution
  990            34.2      96.5
                                     -55-

-------
                                 REFERENCES


1.    Kirsch,  E.J. and  J.E.  -Etzel.   "Microbial  Decomposition of  Pentachloro-
     phenol."    Journal of Water  Pollution  Control Federation   45(2):359-364,
     1973.

2.    Kostenbader,  P.D.   and  J.W.  Flecksteiner.   "Biological Oxidation  of  Coke
     Plant  Weak Ammonia Liquor."  Journal  of Water Pollution Control  Federation
     41(2):199-207,  1969.

3.    Radhakrishnan,  I.   and  A.K. Sinha  Ray.   "Activated  Sludge Studies  with
     Phenol Bacteria." Journal of Water Pollution Control Federation  46(10):
     2393-2417, 1974.

4.    Miller,  M.D.   "Exemplary  Waste  Treatment  System." Presented at  American
     Wood Preservers Association  Conference,  1974.

5.    Thompson,  W.S.   "Wood  Preservatives and the  Environment." Presented  at
     Conference of American  Wood  Preservers Association,  1974.

6.    EPA Project 12130  EGK 06/71.

7.    Capestany, G.J.   "The  Influence of Sulfate on  Biological Treatment  of
     Phenolbenzaldehyde Wastes."  Journal  of Water Pollution Control  Federation
     49 (2):256-261,  1977.

8.    Keith, L.H.   "Chemical  Profiles  of Kraft-Paper Mill  Treated Wastewaters."
     Abstract  Bulletin  of the Institute of  Paper Chemistry,    (45)   3,   1974.

9.    Holladay,  et al.   "Biodegradation  of Phenolic  Waste Liquors  in  Stirred
     Tank,  Packed Bed,  and Fluidized  Bed  Bioreactors."  Journal of  Water Pollu-
     tion Control Federation 50(11):2573-2589, 1978.

10.   Egh, L.  and J.C.  Mueller.   "Rotating  Biological  Disc Treatment of  Kraft
     Mill  Effluents."   Journal of Water  and  Pollution  Control,  113(5):25-29.

11.   Ingols,  R.S.  et  al.   "Biological  Activity  of  Halophenols."  Journal  of
     Water  Pollution Control Federation 38(4):629-635,  1966.

12.   Ashmore,  A.G. J.R. Catchpole, and R.L. Cooper.  "The Biological Treatment
     of  Carbonization  Effluents  -  Investigation into  Treatment by The  Acti-
     vated  Sludge  Process." Water Research,  Pergamon Press Vol.  1:605-624,
     1967-   (Printed in Great Britain.)
                                     -56-

-------
2.0  FOAM FRACTIONATION

     Foaming  is  generally  considered a  nuisance  by  the  pulp and  paper in-
dustry.  There  is  benefit,  however,  in that foam  formation  tends  to concen-
trate some  of  the  pollutants of interest at  the gas/liquid interface.  Leach
et al  reported that  this  process  normally removes only one  source  of pollu-
tion, e.g.,  suspended solids,  color,  etc.  Therefore, foam separation must be
used in conjunction with other unit processes to effect complete treatment.(1)
It has also  been reported  that foam  separation,  though readily applied to
bleached kraft whole mill effluent, is ineffective on newsprint and groundwood
mill effluents.(2)

     Removal of pollutants  in  the foam fraction of effluents has been studied
for  several years.(3, 4)   The process requires a  large  gas-liquid  interface
area (30 to 50 m /I) which may be met by conventional aeration systems such as
turbines,  jet  aerators,  and/or porous diffusers.   Foam removal  also improves
the  operating  efficiency and stability of  standard biological treatment sys-
tems by reducing toxicity and shock loadings.

     Mueller  et  al  (3)  noted  an  increase  from between 60 and  70 percent to
over 95  percent removal of toxicity  on an activated  sludge  pilot plant with
upstream foam  separation.   The removed foam  (generally 2  to  3 percent of the
influent volume) was  subsequently treated by biological oxidation employing at
least  three  days  aeration  time.   Rubin et al(5) noted 10 to 45 percent reduc-
tion (11 to 53 mg/1) in effluent  COD values  following foam separation, indi-
cating that organic  compounds  other  than  the  ABS  (alkyl  benzene  sulfonate)
under  scrutiny were  being  removed with the foam.  In addition, an increase in
effluent pH suggested that weakly acidic compounds (such as the phenol group)
were being  removed preferentially.

     More  recently,   Grieves  et al  (6)  studied the  removal  of  phenols  from
aqueous solutions over the pH range 10 to 12.  Synthetic phenol solutions were
used.  Foam fractionation was reported to remove 40 percent of the phenol with
stoichiometric  surfactant  concentrations.  Removal  was  found to  be strongly
influenced by ionic strength.

     Ng et al (4) stressed the pH dependency of foam fractionation.   Detoxifi-
cation of  whole mill effluents  was  found  to be effective only  at  pH values
greater than  7.   Temperature did not appear to have a great effect within the
range  25°  to 40°C.   Efficiency of the  fractionation system  was  found to be
improved by using  a two-stage system.  Ng also reported that  interfacial area
requirements  and   thus  foam volumes,  increased with the  increase   in toxic
compound concentration.

     Operating  costs  for a  foam separation  facility  were  estimated at about
$1.78  to $2.71/ton  at  a 25  mgd effluent  flow rate, including  a  biological
facility for treatment of the collapsed foam.  Capital costs were estimated at
$1.0 to  $1.7  million (1976).  Ranges of costs represent differing foam gener-
                                     -57-

-------
ation and  collapse  methods as shown  in  Table  27.   Ng concluded  that if sus-
pended solids removal is not required, use of a turbine system was the process
of  choice.   The  capability to remove solids is part of  dissolved  air flota-
tion.

             TABLE 27-   CAPITAL AND OPERATING COSTS FOR VARIOUS FOAM
               	      SEPARATION SYSTEMS (1976) (4)	^___
Foam separation
   system	
 Capital
   ($)
       Operating
$/1.000 gal.     $/ton
Porous media

Turbine aeration

Dissolved air
 flotation
1,201,346

1,012,546

1,664,646
   6.29

   5.34

   8.13
2.10

1.78

2.71
                                     -58-

-------
                                 REFERENCES
1.    Leach,  J.M., J.C.  Mueller,  C.C.  Walden.  "Identification and  Removal  of
     Toxic Materials  from Kraft  and  Groundwood Pulp Mill  Effluent,"  Process
     Biochemistry. 10(1):  7-10, 1976.

2.    Council  of  Forest  Industries  of  British Columbia.   "Pollution  Control
     Objectives of the Forest Products Industry:.   1976.

3.    Mueller,   J.C.,  J.M.  Leach, C.C.  Walden.   "Detoxification of  Bleached
     Kraft  Mill  Effluents  -  A  Manageable  Problem."  Tappi, 60(9):  135-137,
     1977.

4.    Ng, K.S., J.C.  Mueller, C.C. Walden.   "Foam Separation for  Detoxification
     of Bleached Kraft Mill Effluents." Journal Water Pollution  Control Feder-
     ation,  48(3): 458-472, 1976.

5.    Rubin,  E. ,  R.  Everett,  J.J. Weinstock,  and H.M.   Schoen.   "Contaminant
     Removal  from Sewage  Plant  Effluents  by Foaming."  Report AWTR-5,  U.S.
     Dept. of Health, Education and Welfare, December 1963.

6.    Grieves,   R.B.,   W.   Charewic,  S.M.  Brien.   "Separation of Phenol  from
     Dilute Alkaline Aqueous Solution by Solvent Extraction, Solvent Sublation
     and  Foam Fractionation."  Analytica  Chimien  Acta,  73(2): 293-300,  1974.
                                     -59-

-------
3.0  SOLVENT EXTRACTION

     Phenol removal  by solvent extraction  has  been used extensively  for the
treatment  of  refinery  and  coke  by-product wastes.(1,  2,   3)   Mulligan  (2)
reported of an extraction process  developed  by the Jones and Laughlin Corpora-
tion which  was  capable of 99.7 to  99.9  percent phenol  removal  when treating
coke wastes.  This process is  shown  schematically on Figure  13 and is reported
to produce  an effluent containing  1 to 4  mg/1 phenol  from  an  influent feed
containing 1,500 mg/1.  Even  though significant removals are obtained, _it was
Mulligan's  opinion  that some  form  of polishing  treatment  would be  required
prior to  direct discharge to  a  stream.   This study  also evaluated the effect
of feed  concentration on- the  cost  of phenol  recovery.  For evaluation pur-
poses,  it was assumed  that the solvent had  a  high distribution coefficient
(low solvent  to  wastewater  ratio)   and  that it  was easy  to strip from  the
aqueous  raffinate.  Unfortunately the study did not include  capital,  mainte-
nance or labor expense, each of  which significantly affects  the cost of recov-
ery.   The results of  this evaluation are  shown  in Table  28.   Mulligan con-
cluded that where  solvents with  a  high distribution coefficient are available,
phenol recovery through extraction can be economical.

                     TABLE 28.   EFFECT OF FEED CONCENTRATION
	ON SOLVENT EXTRACTION  PROCESS (2)	

                     Energy consumption	Solvent losses  @0.15%	
Phenol concen-    Steam consumption    Cost     Solvent  losses         Cost
tration (%)	Ib/lb phenol    $/lb phenol  Ib/lb phenol	$/lb  phenol
1.0
.0.1
0.01
5
50
500
$0.018
0.15
1.50
0.02
0.09
0.90
$0.004
0.018
0.18

     Earhart et al  (1)  have  done extensive work with  the  recovery of various
organic pollutants  through  solvent extraction.  Important process parameters
are summarized below:

     1.   Solvent solubility  should be minimal;

     2.   The  solvent's  equilibrium distribution  coefficient DO  should  be
          high  (Table   29  contains  experimental values  for  K.^  for  various
          solvents and  solutes);                          „
                                     -60-

-------
             PROCESS SCHEMATIC  FOR SOLVENT  EXTRACTION
                OF PHENOLS  FROM COKE PLANT WASTEWATERS (2)
                  EXTRACTION
                    COLUMN
 FEED-
SURGE TANK


1

— v
-€?
CONDENSER
3-

CONDENSE


•


PHENOL
                                     SOLVENT
                                     RECOVERY
                                     COLUMN
                                                          STILL
                                                      RECYCLE  SOLVENT
                                                 RAFFINATE
                                                                                       FIGURE 13

-------
     3.    The ratio KS/W should be between 1.5  and 3.0,  where
                    K= wt fraction of solute in solvent
                        wt fraction of solute in water

                    S = mass  flow rate of solvent

                    W = mass  flow rate of water;  and

     4.   The solvent circulation rate should be minimized.

                TABLE 29.   EQUILIBRIUM DISTRIBUTION COEFFICIENTS
_ FOR VARIOUS SOLVENTS (20 to 25°C)  (1) _

                                                  n-Butyl Methyl Isobutyl
Solute           Isobutane  Isobutylene  Benzene  Acetate _ Ketone
Phenol
o-Cresol
M Cresol
o-Ethylphenol
0.2 0.7
4.8
2.7
__ __
2.9
16
10
63
65
—
150
__
110
--
260
__
o-Chlorophenol      —          —         --         290          490
     When removing phenols,  it has been  determined that lower values  of S/W
can be  used with polar  solvents  than with  volatile hydrocarbons.   Unfortu-
nately, with polar solvents, there  is  considerable carryover in the effluent.
These solvents do, however, have  a  high distribution coefficient  for extrac-
tion into volatile hydrocarbon  solvents.   As a  result,  Earhart et al proposed
the processes shown on Figures  14 and  15  for phenol removal.  In each case, a
polar solvent such as butyl acetate is used  for phenol  extraction and a vola-
tile hydrocarbon  suc'h  as isobutane is used  to  remove  residual butyl acetate
from the aqueous  fraction.   The  solvents are  then regenerated in- two distilla-
tion columns to isolate  the recovered  pollutants and recycle the  solvents.   A
mini-pilot plant was used to test this process  for 'treating wastewater from a
lube oil refining operation.  Two sets of sequential extractions  with n-butyl
acetate and isobutylene were carried out  at  a water flow rate of 3.21 gal/hr.
Solvent flowrates  for  each pair  of runs were:

     Run No.        n-butyl  acetate           Iso-butylene

       I              0.37 gal/hr              0.55 gal/hr
       H             1.11 gal/hr              -0.55 gal/hr


                                     -62-

-------
DUAL SOLVENT  PROCESS  SCHEMATIC
               SEPARATE  CYCLES (I)
      LOADED P.S
                                                    PURIFIED WATER
                                                  FIGURE 14
                               63

-------
           DUAL  SOLVENT PROCESS  SCHEMATIC
                              LINKED  CYCLES (I)
WASTE
WATER
                 LOADED MIXED SOLVENT    ^
             MIXED
            SOLVENT
           EXTRACTOR
            VOLATILE
            SOLVENT
           EXTRACTOR
                                                     V.S. - P.S.
                                                     SPLITTER
                                                RECYCLE P.SS.
                                                     P.S.-POLLUTANTS
                                                         SPLITTER
                                          POLLUTANTS
                                                      J»EC_YCL_E_ Yi£-	I
                                                   	^- V.S. VAPOR
                                       HOLDING
                                        TANK
                                                                 PURIFIED WATER
                                                                FIGURE 15
                                            64

-------
Table 30 summarizes the results of the above pilot plant studies.

     Grieves et  al  (4)  compared the relative  efficiencies  of solvent extrac-
tion,  solvent  sublation,  and foam fractionation  for  phenol removal.   Extrac-
tions with  amyl  acetate were found to provide 80 to 95 percent phenol removal
over a  pH  range  of 7.0  to  10.7.   Removals sharply declined when  the pH was
above 11.0.  Solvent  sublation with amyl acetate over  the  aqueous  phase in a
cylindrical  foam separation column  was  found to provide  removals  similar to
solvent  extraction.   However,   sublation  had  two  distinct advantages  over
solvent extraction

     •1.   There  is  no  equilibrium  dissolution of the  solvent  in the aqueous
          phase.

     2.   It permits the use of a lower phase/volume ratio.

     Lorton  (3)  evaluated the  economics of three alternative  methods of re-
moving  phenols from coal gasification process condensate  - biological oxida-
tion,  solvent  extraction,  and  adsorption.   Of the three alternatives, solvent
extraction  appeared to  be the most economical.  The evaluation was made using
a  hypothetical  flow  rate and  phenol  concentration typical  for gasification
processes  producing  phenols.   Manufacturers of commercial  processes  then
developed process  designs  and  cost estimates  for use  in the evaluation.  The
results  are given  in  Table  31.  The recommended process,  as  shown on Figure
16, was a product of the Chem-Pro Equipment Corporation.  This system has been
proven  commercially at  the Jones and Laughlin Steel Mill in Pittsburgh, Penn-
sylvania. (1, 2)   Though liquid extraction was the recommended process, actual
pilot plant data should be obtained for each of the above alternatives before
making a  firm  selection.
                                     -65-

-------
                   TABLE 30.  SEQUENTIAL EXTRACTION OF LUBE-OIL REFINING WASTEWATER  (1)
o>

Butyl-acetate Extraction
Concentration, ppm
Component KD
Feed Prod. ^Removal
Isobutylene Extraction
Concentration, ppm
KD Feed Prod. %Removal
Overall
% Removal
S/W for Butyl Acetate Extraction = 0.100:
Methyl Ethyl Ketone 4.6
Phenol 65
o-Cresol
ri-Butyl Acetate
TOD
COD
12
8


52
54
,200 5,900 52
,800 104 98.8
890 6.5 99.3
0 7,100
,900
,500
2.5 5,600 3,600 36
0.7 310 230 26
4.8 24 2.3 91
168 7,100 11.0 99.8
9,350
8,570
69
99.1
99.9

82
84
a
S/W for Butyl Acetate Extraction = 0.30:
Methyl Ethyl Ketone 4.6
Phenol 65
o-Cresol
n-Butyl Acetate
TOD
COD
12
8


' 52
54
,200 2,500 82
,800 77 99.1
890 4.3 99.5
0 6,800
,900
,500
2.5 2,800 1,890 33
0.7 230 190 17
4.8 18.0 2.8 84
168 6,800 15.2 99.8
5,110
4,690
88
99.3
99.9

90
91

-------
                TABLE 31.   ECONOMIC EVALUATION OF PHENOL REMOVAL
                          FROM PROCESS COMPENSATE (3)

•v
Incremental capital requirements
Extraction
1.
2.
3.
4.
5.
Initial plant investment
Chemicals & catalysts
Royalties
Startup costs
Working capital
Subtotal
base
base
base
base
base
base
Adsorption
$9,100,000
1,700,000
200,000
200,000
100,000
$11,300,000
Bio-oxidation
$22,500,000
0
0
600,000
200,000
$23,300,000
                                     Incremental operating costs

6.  Utilities                  base         $  500,000     $   900,000

7.  Chemicals & catalysts      base            300,000         600,000

8.  Labor                      base            100,000         400,000

9.  Administration & overhead  base            100,000         200,000

10.  Supplies                  base            100,000         200,000

11.  Taxes & insurance         base            100.000       .  500,000

     Subtotal                  base         $1,200,000     $ 2,800,000


12.  Credit for sale of        base               0        (-1,100,000)
    phenol byproduct

          Total                base        $12,500,000     $27,200,000

                                   9
**NOTE:  Plant capacity is 250 x 10  Btu/day of pipeline quality gas
         from western coal at a 90% stream factor
                                     -67-

-------
CTJO
a m
"i
0 =
                                 CHEM-PRO   SOLVENT   EXTRACTION  PROCESS  SCHEMATIC  (3)
                    PMCI91 COMANUTI
                                                       1
r~D
                                                                                                                        ntmtr «^i

                                                                                                                        IMIMU

                                                                                                                        TOF-llll-
                                                                                                                          TO t MMMII MCV1HT W«N 0»« WILL M MriMII


                                                                                                                         itrouLiHt IN TM tot.HINT MMiifa is KMMMJ PILOT *IMI
                                                                                                                          •ILL MM we i mi LIICUHOOO or rouiiMi, MI CMH-MO H
                           PRELIMINARY
rxNa aiwMft
r= „ filMM MM
awm ' - .
tM/M
11* TM
MLVMT 	
OTU IM/HB
_- •*«* -
~Moan
tMHHUTI
'"'"•
MB

'•!.«•*
\»»
fioam
oMffmm
t,OIO
•oo
• 0
— " "
H*.IU
1,0(0
pwa*
CMWsrMJl
IM.OOO
*,01(
IM
	
»*.<••
l»
MML
MOOUCI
«.oo«
_ 100
. .._>»
*.
-------
                                  REFERENCES
1.    Earhart et al.   "Recovery of Organic Pollutants via Solvent Extraction."
     Chemical Engineering Progress,  73(5):  67-73,  1977.

2.    Mulligan, T.J.  and R.D.  Fox "Treatment of Industrial Wastewaters."
     Chemical Engineering, 83(22): 49-65, 1976.

3.    Lorton, G..A-  "Removal of Phenols from Process Condensate." U.S.
     Department of Energy Contract No. EX-76-C-01-224Q October 1977,  24 pp.

4.    Grieves, R.B. et al.  "Separation of Phenol from Dilute Alkaline
     Aqueous Solutions by Solvent Extraction, Solvent Sublation and Foam
     Fractionation." Analytical Chemica Acta 73(2): 293-300, 1974.
                                     -69-

-------
4.0  PHYSICAL/CHEMICAL OXIDATION

     Numerous  researchers  have  demonstrated the  technical  feasibility  of
physical/chemical oxidation of  phenols.   Their work has been  centered prima-
rily around use of the following oxidants:

     o    oxygen

     o    ozone

     o    chlorine

     o    chlorine dioxide

     o    hydrogen peroxide

     Additionally, oxidation  by ultrasonic, ultraviolet radiation,  and cata-
lytic contact treatments has been demonstrated.

     Oxygen  -  Oxygen has been used with and without catalysts as well as in
combination with  ultraviolet  radiation  or ultrasound.(1)  One such use is the
wet air  oxidation process  in which the  waste  temperature  is increased to the
point where  air (oxygen) reacts with oxidizable material  autogenously.  Wil-
helm and Ely (2)  reported a 99.99+ percent COD removal (Figure 17) when using
this process.   The Pulp and Paper Institute of Canada  reported several addi-
tional methods  of oxidizing phenols  with oxygen.(1)  Waste from a lacquer and
varnish  manufacturer was  oxidized using  oxygen  (in  air)  and acid  treated
pyrolusite  (MnO.)  as a  catalyst.  The  results are listed in  Table  32.  Suc-
cessive uses of the  catalyst resulted in a decrease in its activity.  It was,
however, possible to regenerate  the  catalyst  with a 1.5 to  2.0  percent sul-
furic acid solution.

     Examples  of the  photo-oxidation of  dilute phenol  solutions  were also
included in  the  Canadian  report.(1)  A waste containing 200  mg/1 phenol was
irradiated with ultraviolet light (300 to 400 nm) using 5,000 mg/1 of ZnO as a
catalyst.  Ninety percent of the phenol was oxidized to CO   within 44 to 68
hours,  depending on  the depth  of the  waste  stream.   Even  though  dissolved
oxygen was found to be necessary for oxidation, aeration did not significantly
affect the phenol oxidation rate.  When TiO^  was  substituted  as  the catalyst
under  the  same  conditions,  a  maximum  80  percent phenol reduction occurred
after 72 hours  irradiation.   Similar results were  also  obtained  when 300 g/1
of beach sand was used as the catalyst.

     Ultrasonic  irradiation  (ultra  high frequency  sound) was also found to
produce  significant   phenol  reductions.(1,  3)   Wastes  containing  100 mg/1
phenol were reported to be completely oxidized in four hours  when subjected to
a  field  of  800  KHz  at  33  watts/cm .   When  the  waste  concentration was in-
creased  to  500  mg/1,  a 70-percent  reduction was  achieved in 6  hours.   As
                                     -70-

-------
         DECOMPOSITION  OF  PHENOL

           BY  WET-AIR  OXIDATION (?)
      100
        10
o
111
(T
O
Z
III
        1.0
       O.I
     0-01
                  20       40       €0      80


                        COD REDUCTION, %
100
                                             FIGURE 17
                             71

-------
shown on Figure  18,  phenol  reduction was found to  be  proportional to irradi-
ation intensity.  Even though  significant  phenol  reductions were obtained, it
was pointed out  that this process appears  to oxidize phenols to products with
a higher oxygen  content  and not to carbon  dioxide.   Consequently, this treat-
ment may not necessarily reduce toxicity.

     Ultrasonic  irradiation in  the  presence  of  various  catalysts  was  also
investigated by  Chen.(4)  In experiments with metal oxides,  platinum,  rhubi-
dium and Raney-nickel, an inhibition of sono-oxidation  was  noted.  Howevsr, as
indicated by TOG analysis,  more complete oxidation  occurs  since approximately
eight percent of the products of oxidation  was carbon dioxide (CO ).

     TABLE 32.   CATALYTIC OXIDATION OF LACQUER MANUFACTURING WASTEWATER (1)


     Phenol concentration (raw waste)                        8,000 mg/1

     Formaldehyde concentration (raw waste)                   1,400 mg/1

     pH (raw waste)                                               7.0

     Aeration period                                             2 hr

     Percent removals

     Phenol                                                     99%

     Formaldehyde                                        87 to 95%
     Ozone - Ozone has  been shown to be extremely  effective  in the oxidation
of  many complex organics  including phenols.   Eisenhauer  reported that  when
ozone initially  reacts  with phenol,  catechol is formed.   This compound,  in
turn, degrades  to carboxylic  acids and  carbon  dioxide,  as  shown on  Figure
19.(1, 5)

     However, Dence,(6) reporting  on the  ozonation of spent  chlorination and
caustic extraction liquors,  stated that "in no instance does ozonation cause a
substantial reduction in the toxicity  .  .   .  and under alkaline conditions, a
decided  increase  is  produced."  He  goes  on  to  state that "it  seems  safe to
assume  that  the phenolic  components  of  the  liquors are not  responsible."
Apparently, under alkaline  conditions, ozone  partially decomposes  to hydroxyl
groups and could form more  toxic products.

     The phenol  content of coke  plant effluent has  been  reduced  from 2,000
mg/1 to  less than  1  mg/1  with an  average  ozone  dose  of 1.7 g/g of phenol. (5)
Cleary and Kinney  (7)  also  worked with coke  wastes.   Their results,  as shown
on Figure 20, led to  the following conclusions:
                                     -72-

-------
        PERCENTAGE  OF  PHENOL  OXIDIZED


                               VS

                ULTRASONIC  INTENSITY (I)
Q
UJ
X
o
o
z
Ul

a.
ui
     60-
     50-
     40-
     30
Ul
o

£    20
                      10              20            30


                         ULTRASONIC INTENSITY, wafts/cm2
40
                                                  FIGURE 18
                                  73

-------
                  OZONATION  OF PHENOL ( 1,5)
OH
                                                             -f- COOH	HOOC
                                                        COOH
                                                                   FIGURE 19

-------
OXIDATION  OF PHENOL  BY OZONE (7)
  200
400       600      800



   OZONE DOSAGE, PPM
                                         1000
1200
                                       FIGURE 20
                        75

-------
     1.    Ozone oxidation of phenol  is  maximized at pH 11.8.

     2.    The required ozone dose  is independent of temperature.

     3.    Undertreatment with ozone does not  form additional toxic compounds.

     4.    No pretreatment of the waste  is required.

     Similarly, phenols  in  refinery wastes have been reduced from  11,600 to
2.5 mg/1 with  an  applied ozone dose of 2.0  g/g  of phenol.(5)  Gould and Weber
(8) evaluated phenol oxidation with  ozone and  found that  near-complete removal
was achieved when four  to  six moles of ozone  were consumed per mole of phenol
initially present.  They also determined that  when a plot of  COD  reduction vs.
ozone dosage is made, there  is a sharp  transition from one linear relationship
to another  at a much lower  rate.    The break  point between  the  two  functions
was found to correspond to the point of near-complete phenol  removal.

     Little work has been done on  the influence  of catalysts  on phenol removal
with ozone.  The Pulp and Paper Research Institute of Canada  (1)  reported that
the oxidation  rate  is  increased twofold when  Raney-nickel is  used as a cata-
lyst.

     Many researchers reported that  the cost of  ozonation is  excessive.  Dence
(9) states  that in  the  case of spent liquors, "other components  compete .  .  .
successfully with phenols for the ozone, requiring excessive amounts of ozone
for phenol reduction."  Eisenhauer(S),  in reporting on the work of Peppier and
Fern, stated that the cost of ozonation is four  to seven  times that of biolog-
ical  oxidation.   Thompson  (10) similarly  disregarded ozonation as  a viable
treatment  alternative  due  to  high  capital and  operating costs.   Gould  and
Weber  (8)  stated that  ozonation for  complete  removal  would be  exceedingly
costly and  time consuming.   In general, most  researchers  said that  ozonation
would only be  economically feasible  as  a polishing step  after some  other
treatment process.

     Chlorine - Chlorine is  a powerful oxidant  and  as such  will readily oxi-
dize phenol.  Thompson  (10)  reported that  chlorination of phenolic  wastes to
an excess  actually  breaks the benzene  ring to  form organic  acids.   There is
similar evidence  to show that chlorine oxidizes pentachlorophenol  to chlor-
anil.  Thompson did  not feel, however,  that  chlorination  alone will adequately
treat wood preserving wastes. Too many oxygen demanding  substances,  including
residual phenolic fractions, sometimes  remain  in  the  effluent, even  after
massive chlorine dosages.  Frequently a ratio  of chlorine to  pentachlorophenol
of 700:1 is  required for near-complete oxidation.  This  reduces  the economic
feasibility of such  a treatment alternative.  Thompson did feel,  however, that
chlorination could be used  as a polishing  step  for biological treatment plant
effluents.
                                     -76-

-------
     Cleary and Kinney  (7)  investigated chlorine oxidation  of  coke wastes at
both the laboratory and pilot plant levels.

     They found that once the waste's ammonia demand is satisfied, substantial
phenol reductions  can be obtained with small  increases  in the chlorine dose.
The  reaction  reaches a  point,  however, where  increasingly larger  doses  are
required for  small increases in phenol destruction.  The relationship between
chlorine dose and phenol remaining is shown on Figure 21.

     They also  found that phenols  can be destroyed at  pH  values between 1.8
and  11.0.   To prevent  the formation of chlorophenols and nitrogen trichloride
the  pH should be  between 7.0 and 10.0.  Temperature should also be reduced to
45°C to  prevent  chlorate formation.   There is some indication that pH affects
the  amount  of chlorine  demand  which  is  satisfied prior  to beginning phenol
destruction.  It seems to be easier to oxidize phenols at a high pH when other
chlorine consuming compounds are present.

     Based  on the > above, complete  destruction  of phenol  with  chlorine  is
technically feasible.   It does,  however, require an excess of several hundred
mg/1 of  chlorine,  to accomplish complete destruction.

     Oxidation  by chlorination,  however,  may  produce harmful  side effects.
Dence et al,  (6)  in current research at SUMY-Syracuse, have reported distinct
increases  in the  toxicity of  spent  chlorination  liquors  following chlorine
treatments, even though up to 70 percent phenol reduction was noted at dosages
of 0.8 mgCl/ml  spent liquor.  Similar results occurred when hypochlorous acid
was  used as  treatment.   Phenol  reduction of 82 percent was  experienced at a
dosage  of   0.37  mg  HOCl/ml  spent liquor.   Similarly, a  sodium hypochlorite
treatment  at 2 mg/ml  of liquor  produced  at  95  percent reduction  in phenol
concentration in nine hours at 60°C.  A decided increase in toxicity was again.
noted in the 'treated samples.

     Chlorine Dioxide -  Chlorine dioxide is another strong  oxidant which can
be used  for  treating  industrial wastewater.  Unlike  chlorine,  however, this
chemical has  been reported to be selective for cyanide, phenol, sulfides, and
mercaptans.   It does not react with many  other  organics such as alcohols and
organic  acids.(11)   Cleary  and  Kinney (7) have  found  in their work with coke
plant wastewaters  that pretreatment  was not  required.   They found  that  the
best  results  were obtained when a 2  to  1 mixture  of chlorine  to chlorine
dioxide  was  applied  together.   When chlorine dioxide  was  used alone, a small
chemical dosage  was found to produce a large phenol  reduction.   As shown on
Figure  22,  additional  phenol  removal  required increasingly  higher chemical
doses.   Mulligan   and  Fox  (11)  reported  that  approximately  1.5  Ib chlorine
dioxide  was required to  convert one" pound  of phenol  to  quinone.   They also
reported that "because  of its cost, approximately  $4/lb, its use will be as a
selective polishing treatment  for destruction  of  trace amounts  of specific
compounds in industrial effluents."
                                     -77-

-------
         OXIDATION  OF  PHENOL  BY CHLORINE  (7)
     100
S
Q.
0.
z
z
bJ
CC

V)

o
z
UJ
X
0.
                 1000
2000      3000      4000

  CHLORINE DOSAGE , PPM
                                                        500O
6000
                                                    FIGURE 21
                                    78

-------
OXIDATION  OF PHENOL BY  CHLORINE DIOXIDE   (7)
  100
   80
2
a.
Q.

o"
Z  60
2
Ixl
(T
UJ
I
a.
   40
   20
\
                 148 PPM
                        1000                2000

                       CHLORINE DIOXIDE DOSAGE, PPM
                                                   3OOO
                                                     FIGURE 22
                                      79

-------
     Chlorine dioxide, though expensive, has  two major advantages over chlor-
ine. (7)  First, the ammonia  content  of the waste does not affect the chemical
dose required  for  phenol destruction.   This  is  significant  because approxi-
mately 10 mg/1 of  chlorine  is  required to  oxidize  1  mg/1 of ammonia and that
all ammonia must be oxidized prior to phenol removal  with chlorine.  Also, a
chlorine dioxide  residual is not required.   Underdosing reduces phenol concen-
trations, but according  to  the  colorimetric tests used by Cleary,(7) does not
form chlorophenols.

     Hydrogen Peroxide  - The  FMC  Corporation,  a manufacturer of  hydrogen
peroxide, has  done extensive work  with phenol oxidation.(12,  13)   They have
reported that  in the presence  of  trace amounts of iron  salts,  hydrogen per-
oxide will rapidly oxidize  phenol  first to catechol and hydroquinone, then to
dibasic acids, and ultimately to carbon dioxide and water.   Eisenhauer, how-
ever, has disagreed and  stated  that the reaction  stops  with the formation of
muconic acid.(14)

     As part  of  their work  with phenols,  FMC  has studied the  effect  of the
H?0 /phenol ratio on the extent  of  phenol oxidation.  The results of this work
are  shown  on Figure  23. (12)   Similar work also led to the  development of a
relationship between H 0?/COD and phenol reduction as  is shown on Table 33 and
Figure 24.(12)

          TABLE 33.  HYDROGEN PEROXIDE TREATMENT OF PHENOL WASTES* (12)

H 02/COD
wt ratio
0.3
0.4
0.6
0.7
0.8
1.0
Reduction in phenol
(%)
86
94
99
99.8
99.9
100
Reduction in COD
(%)
28
32
40
44
52
69

*Initial phenol concentration 2,000 mg/1

     Eisenhauer (14) developed a  similar  relationship using Fenton's reagent;
or  in other  words  a  combination  of  hydrogen peroxide  and a  ferrous  salt.
Table 34 shows the results of this work.
                                     -80-

-------
                                      EFFECT   OF   INITIAL  HYDROGEN   PEROXIDE/

                                                        PHENOL   RATIO  ON THE

                                                 OXIDATION  OF  PURE  PHENOL (12)
00
5

_i
O
                            I
                            Q.
                            in
                            LU
                            OL
                                 \50-t
                                 125 -
                                 100-
                                  75-
                                  50-
                                  25-
INITIAL PHENOL  CONC =

500  Mg/Lifer

70° F  5 Minutes
0.1% ,Fe Calolyst

Initial pH = 5.5
                                             0.5     1.0      1.5     2.0     2.5     3.0     3.5    4.0

                                                      HYDROGEN PEROXIDE/ PHENOL WEIGHT RATIO
                              4.5
                                                                                                           FIGURE  23

-------
                          EFFECT  OF  HYDROGEN  PEROXIDE   /   PHENOL  RATIO

                           ON  CHEMICAL   OXYGEN  DEMAND REDUCTION  OF

                                                     .PURE PHENOL (12)
CO
N3
                         Q
                         Z
                         <
                         5
                         ui
                         Q
                         X
                         o
                         o
                         2
                               1200
                               1000-1
                               '800-
60O-
400-
200-
                                        INITIAL PHENOL CONC :
                                        500 Mg /Liter
                                        70'F  5 Minutes
                                        . 01% Fe Catalyst

                                        Initial pH =5.5
                                           0.5      1.0     1.5     2.0     2.5     3.0    3.5     4.O

                                                    -   HYDROGEN PEROXIDE/PHENOL WEIGHT RATIO
                                                                                                   FIGURE 24

-------
            TABLE 34.   REACTION OF PHENOL WITH FENTON'S REAGENT* (14)

                                                  Phenol concentration
a.- c v,rc.n , ) 2 *-OUA J 2
(moles/mole
of phenol)
0.1
0.25
0.5
1
1
1
1
1
O f
(moles/mole
of phenol)
9
9
9
9
3
2
1
0.5
2
min
33.6
21.2
3.0
0
2.0
13.2
16.0
22.4
5
min
27.8
11.6
0
0
0
8.8
14.2
19.0
15
min
8.5
0
0
0
0
8.0
13.6
16.0

     *Solutions containing 50 mg/1 of phenol were reacted at pH 3 and 10°C
      with air agitation.

Eisenhauer concluded that  optimum phenol oxidation takes place when the reac-
tion is carried out at a 3:1 mole ratio of peroxide to phenol.   Unfortunately,
when  this  reagent  was  applied  to phenolic  industrial wastes,  the  peroxide
requirement increased  three  to  six times.  This increase was believed to be a
result of other  organics in the waste.  Eisenhauer also reported that cost of
phenol removal to be $2.40/lb of phenol (1964 prices), which was too expensive
to compete with other removal processes-

     In addition to the  work described  above,  the FMC  Corporation  has also
investigated the  effect  of temperature and various catalysts on phenol oxida-
tion. (13)   Temperatures  between  70°  and 120°F were found to have no practical
effect on the  reaction.   Catalysts such as ferrous sulfate, iron wool, nickel
salts and aluminum  salts were also evaluated.  As  shown in Table 35, ferrous
sulfate appeared  to  be the most practical of all the catalysts studied.  This
finding was also supported by Ross (15)  and  Keating  et al.(l6)  Keating also
compared the effectiveness of ferrous and ferric iron and found no differences
with  the  testing conditions  used.   He  did  note,  however,  that the reaction
seemed to begin  sooner when ferrous iron was  used as evidenced by an earlier
color change.   FMC similarly tested  potassium persulfate  both with hydrogen
peroxide  and  alone  with  an iron  catalyst.   Even  though good  oxidaton was
obtained,  the  results were  the  same  as  if hydrogen peroxide  alone  had been
used.(13)

     A  1961  Russian article, as reported by- the FMC  Corporation,  "gave the
following order  of catalytic activity for the peroxide oxidation of phenol;
                                     -83-

-------
      Al' '  ,  Cr'''.   The catalytic Fe'    and Cu''  activity was also said to be
enhanced by the presence of Al   ." (13)  The previously described work of FMC
does  not  verify  this   conclusion.   FMC  has  pointed  out,  however,  that the
Russian work was performed on a much more dilute  solution.

     Merour (17) took another approach to the problem  and  found that oxidiz-
able material may be removed from wastes as follows:

     1.   saturate the  waste  streams  with oxygen  then  add  hydrogen per.oxide;
            TABLE 35.   EFFECT OF CATALYSTS ON PHENOL OXIDATION (13)
     H2°2/
     phenol
     ratio
Catalyst
Phenol
0*8/1)
 TOC
(mg/1)
                                                    COD
     none      none                      500.0        383

     0.5       none                      499.0        365
     1.0       none                  -   493.0        375
     0.5       0.01% Fe                   88.0        380
     1.0       0.01% Fe                    7.0        360
     2.0       0.01% Fe                    0.3        310
     3.0       0.01% Fe                    0.2        150
     4.0       0.01% Fe                   3ppb        138
     0.5       0.03% Fe                  118.0        365
     1.0       0.03% Fe                   22.0        345
     2.0       0.03% Fe                    0.2        315
     3.0       0.03% Fe                    4.04       210
     0.5       0.01% Fe, 0.01% Al         79.0        380
     1.0       0.01% Fe, 0.01% Al          4.0        310
     2.0       0.01% Fe, 0.01% Al          0.1        300
     3.0       0.01% Fe, 0.01% Al          0.05       200
     4.0       0.01% Fe, 0.01% Al         6ppb        196
     1.0       0.03% Fe, 0.03% Al          5.7        310
     2.0       0.03% Fe, 0.03% Al          0.3        320
     3.0       0.03% Fe, 0.03% Al          0.2        225
     1.0       Steel wool (room           16.0
               temp., 10-min.
               retention)
     0.5       0.01% Ni (as NiCl2    •    496.0        370
               6H..O) (room temp.
               test)
     Initial phenol concentration - 500 mg/1. 120°F, 30-min. retention
     time; initial pH = 5.5. Fe added as ferrous sulfate; Al added as
     aluminum sulfate.
                                                   1,105
                                                (theory 1,19-0)
                                                     974
                                                     679
                                                     392
                                                     106
                                                      66
                                                   1,020
                                                     620
                                                     418
                                                     150
                                                     971
                                                     625
                                                     495
                                                     172
                                                     105
                                                     510
                                                     405
                                                     210
                                     -84-

-------
     2.    adjust the pH to 6-7;  and

     3.    activate the mixture with light having a wavelength of 200 to 500 nm
          and an energy of 0.5 w/atom of carbon eliminated.

     This process was  found  to require 0.1 to  1.5  atoms of active oxygen per
atom of carbon removed.

     Keating  et al,  (16) reported  on  three  commercial  installations  where
phenolic wastes  are being  successfully treated with  hydrogen peroxide.   One
installation used a batch treatment system.  During a typical treatment opera-
tion, approximately  10,000  gallons of wastewater were  received  and tested to
determine the phenolic  concentration,  the HO /phenol rates, and the required
treatment time.   The waste was then adjusted to pH 5-6, iron added to equal 15
|Jg/g Fe,  and the required  amount of peroxide is added.   After  the necessary
treatment time,  the  waste was retested for phenol;  and if found satisfactory,
discharged to the receiving stream.

     Hydrogen peroxide  has  also  been used as a  standby system for biological
treatment.  At  one  of the facilities studied, inadequately treated wastewater
was diverted to a holding pond where it was stabilized with hydrogen peroxide.
It  is  interesting to  note  that  the amount of hydrogen peroxide required for
treatment was well in excess of that required for phenolics (H^O^/phenol ratio
of  8:1).   This  was  primarily attributed to the presence  of otner oxidizable
compounds.  This same treatment system also used hydrogen peroxide to counter-
act the  effects  of  biological shock loadings.  If upstream testing detected a
heavy load  of phenols,  hydrogen peroxide would  be added  to  an equalization
basin  until phenols  were  reduced to  a level  tolerable to  the  bio-system.

     The  final  commercial application  reported  by Keating  (16) was effluent
polishing  to remove  the  last traces  by phenol after.biological treatment.
This type  application, however,  is necessary only  when there  are stringent
effluent limitations.

     Reductive Degradation  -  Sweeny  (9)  reported  catalyzed iron  to  be  a low
cost, effective treatment, replacing substituted chlorine atoms with hydrogen.
This removal  of chlorine  is  stated to generally lead  to  products of greatly
reduced  toxicity,  for  example,   the  degradation of  chlorobenzene  to cyclo-
hexanol.  Use of a  column to provide the required contact was investigated in
two variations  -  diluted and fluidized beds.  The  fluidized bed was found to
be most  efficient,  allowing flow rates as high  as  23 gpm/ft  .  A degradation
trial with  an influent  concentration of 2,000 mug/1  p-nitrophenol at a flow
rate of  22.8 gpm/ft   produced an effluent containing  only  1.5  mug/1,  a 99.8
percent reduction.

     An  operating pH  range  of  5.5  to 8.0 was  reported to  avoid excessive
reluctant usage or precipitation in the bed.
                                     -85-

-------
                                 REFERENCES
1.    Pulp and Paper Research Institute  of  Canada "Review of  Catalytic  Oxida-
     tion of Pulp and Paper Mill Effluents",  CPAR Project Report 147-1, March
     1973.  48 pp.

2.    Wilhelmi,  A.R.  and  R.B. Ely.   "A  Two-Step Process  for Toxic. Wastewaters."
     Chemical Engineering,  83(4): 105-109,  1976.

3.    Chen,  J.W.,  J.A.  Chang, and G.V.  Smith,  Sonocatalytic Oxidation in Aqueous
     Solutions."   Vol. 67  (item 109):  18-26,  1971.

4.    Chen,  J.W. & G.V.  Smith.  "Feasibility  Studies of Applications  of Cata-
     lytic   Oxidation  in  Wastewater." EPA Water  Pollution  Control  Research
     Series 7020-ECI-11/71.

5.    Eisenhauer,  H.R.   "The Ozonization  of  Phenolic  Wastes."  Journal Water
     Pollution Control Federation 40(11):  1887-1899,  1968.

6.    Dence, C.W., R.  Hartenstein,  C.J.K.  Wang.  "The Elimination  of  Phenolic
     and Chlorophenolic  Materials in Spent  Chlorination and  Caustic Extraction
     Bleaching Liquors  using   Chemical  and  Biological Approaches."  Progress
     Report to February  1978, SUNY  Syracuse,  New York.

7-    Cleary, E.J.  and  J.E.  Kinney.   "Findings from  a  Cooperative  Study  of
     Phenol Waste Treatment."  In:  Proceedings  of the Sixth  Industrial Waste
     Conference Purdue University,  1951.

8.    Gould, J.P.  and W.J. Weber Jr.  "Oxidation of Phenols by Ozone." Journal
     Water  Pollution Control Federation,  49(1): 47-60,  1976.

9.    Sweeny, K.H.   "Reductive  Degradation:  Versatile,  Low  Cost."  Water and
     Sewage Works.  126(1):  40-42, 1979.

10.  Thompson,  W.S.   "Wood  Preservatives and  the Environment:   The  Treating
     Plant." Presented at Conference  of  American Wood  Preservers Association,
     1974.

11.  Mulligan,  T.J.  and  R.D.  Fox.   "Treatment  of  Industrial  Wastewaters."
     Chemical Engineering.  83(22):  49-65,  1976.

12.   FMC Corporation bulletin  entitled "Industrial Waste Treatment."

13.  FMC Pollution  Control  Release No.  84.  "Phenols in  Refinery  Waste Water
     Can Be Oxidized with Hydrogen  Peroxide."
                                     -86-

-------
14.   Eisenhauer, H.R.  "Oxidization  of  Phenolic  Wastes." Journal Water Pollu-
     tion Control Federation 36(9): 1116-1128, 1964.

15.   Ross, L.W., A.K. Chowdhury.   "Catalytic Oxidation of Strong Waste Waters."
     Presented at Conference of American Society of Mechanical Engineers,  July
     1977.

16.   Keating, E.J.,  R.A.  Brown E.S.  Greenberg E.S.  "Phenolic Problems Solved
     with Hydrogen Peroxide Oxidation."  Industrial Water Engineering,  15(7):
     22-27, 1978.

17.   Merour, Yves.   "Elimination of Oxidizable Materials from Polluted Waters."
     Chemical Abstracts, Vol. 87 (28610), 1977.
                                      -87-

-------
5.0  CARBON ADSORPTION

     Carbon adsorption is basically a surface phenomenon (1) and is influenced
by such  factors  as  the hydrophobic nature of  dissolved  organics in the waste
stream, and the affinity of the organic for the sorbent.

     Adsorption  is  related directly  to  a compound's  hydrophobic  behavior in
that organics of low solubility are more readily adsorbed.(1)  With respect to
phenols,  Strier  (2) described  the  order of increasing  adsorptivity  on acti-
vated carbon as:

                                                  Solubility in water
          Compound	Molecular weight	mg/1  at 25°C
Phenol
2,4,6 Trichlorophenol
Penta chlo ropheno 1
94.1
197.5
266.3
8.2 x 104
9.0 x 102
3.0 at 50°C
     This  is  also  the order  of increasing molecular  weight  and  decreasing
solubility,  which  supports  the  above  statement  concerning  solubility  and
adsorptivity.    The  solubilities  reported  above  by Strier,  however,  do  not
agree with other  published  data.   The affinity of the  organic for the sorbent
is due to  a  combination of electrostatic attraction, physical adsorption,  and
chemical adsorption.(1)  Since  activated carbon is negatively  charged,  it is
"more amenable  to complexing with  increasingly more positively  charged aro-
matic  (benzene)  compounds  such as  the more  highly  chlorinated phenol ico,
thereby facilitating adsorption".(2)  Strier  concluded  that:

     "...pentachlorophenol  could  best  be  removed  from  industrial waste  ef-
     fluents by oil-water separation and/or by  precipitation and filtration at
     low pH,  as  pretreatment,  followed  as  necessary by  carbon  adsorption
     and/or biological  oxidation,  to remove  other phenolics  and  organics."

     There  are  basically  two techniques  for using activated carbon  in waste
treatment:

     1.   The waste stream can  contact  granular carbon in a treatment column;
         or

     2.   Powdered carbon can be added directly to the  wastewater.

     Numerous  researchers  have reported on the  effectiveness of  contacting
type carbon columns for phenol  removal.   Zogorski and Faust  (3)  conducted 25
experiments using laboratory  prepared  phenol solutions  and  fixed  beds  of
                                     -88-

-------
granular activated  carbon.   From  their work, they  concluded  that  between 20
seconds and  two minutes  of  contact time is required to  reduce  phenols below
analytical sensitivity.  They also found that granular activated carbon rapid-
ly removes phenols,  therefore  filters with a bed depth of 3-to-5 ft can be as
efficient as the much larger units currently being  designed.   Hager (4) con-
ducted  similar  laboratory adsorption  studies  using actual  industrial waste-
waters  and found a  99 percent phenol  reduction  in all  five samples analyzed.
He also reported on five full scale adsorption systems installed between March
1973  and  March 1975.   A brief  description of  these  systems  is included  in
Table 36.  Bernardin  (5)  also  reported on  two  full  scale adsorption systems.
Plant "A" was  a chemical plant manufacturing, among other  compounds,  phenol.
A basic process flow  diagram  is  shown on Figure 25.  The  carbon  adsorption
system  consisted  of two moving bed  steel  adsorbers, each of  which  was 12  ft
diameter by 36 ft sidewall height.  Each contained approximately 124,000 Ib of
Calgon  Filtrasorb granular  activated carbon and was designed  for  a flow rate
of 175  gpm.  A summary of typical  plant  operation data is shown on Table 37.
Plant "B" manufactured phenolic  resins and produced a waste containing float-
ing  oils,  suspended  solids, phenolics, and other organics.   A  basic  process
flow diagram is shown  on Figure 26.  The carbon adsorption system consisted of
two fixed bed adsorbers operated in series.  Each column was 10 ft diameter by
11 ft  sidewall  height and contained approximately 20,000 Ib of granular acti-
vated carbon.  A summary of typical plant operating data is shown on Table 38.

	TABLE 36.  PHENOL ADSORPTION SYSTEMS (4)	

                                                               Phenol concentration
Plant         Flow                                Contact time        (mg/1)
No.	(1,000 gpd)	Pretreatment	(min)	Influent  Effluent

  1            50        Settling, equalization,       165          200       0.1
                         mixed media filtration

  2           200        Equalization                   41          600     100

  3           350        Biological, mixed              24          800       0.05
                         media filtration

  4           150        Sand filtration, settling      55        1,200       1.0

  5           500        Biological, settling,          33           80       1.0
                         multi-media filtration
                                     -89-

-------
                                                     PROCESS FLOW  DIAGRAM (5)

                                                                PLANT "A"
                                                                                                                        TO STREAM
VO
o
                       pH ADJUSTMENT
                                            NON IONIC
                                            POLYMER

f '
EQUALIZATION


t
FLOCCULATION


                                                        WASTE -*•
                                                                                                         FIGURE 25

-------
               TABLE 37.  OPERATING DATA - PLANT "A"  (5)
Raw waste characteristics




     Flow




     BOD




     COD




     Phenol




Effluent characteristics




     BOD




     COD




     Suspended solids




     Phenol
400,000 gpd




 16,000 Ib/day




 27,000 Ib/day




  1,500 Ib/day









  1,450 Ib/day




  2,675 Ib/day




     22 Ib/day




      0.5 Ib/day
                                  -91-

-------
                                                    PROCESS  FLOW DIAGRAM (5)

                                                               PLANT  "B"
vo
NJ

OIL
SEPARATOR





FLOCCULATION
DLYMER

                                                 •H   SEDIMENTATION
                                               WASTE
                                                                                                        CARBON
                                                                                                      ADSORPTION



	 *•




MIXED MEDIA
FILTRATION
















































TO STREAM





                                                                                                        FIGURE 26

-------
                    TABLE 38.  OPERATING DATA - PLANT "B" (5)
                     Raw waste       Clarifier effluent     Carbon effluent
Suspended solids
Oil/grease
TOC
Phenol
220 mg/1
50 mg/1
1,200 mg/1
160 mg/1
45 mg/1
25 mg/1
650 mg/1
130 mg/1
10.0 mg/1
5.0 mg/1
25 mg/1
0.1 mg/1

*Plant flow is approximately 100,000 gpd


     It is interesting to note that when the raw wastewater toxicity was eval-
uated using bluegill  sunfish,  all specimens died within  the  first 15 minutes
of exposure.   In  a similar test with  the  final effluent, however, there were
no adverse reactions even after a 10-day exposure test period.

     Thompson  (6)  pointed  out  that even though activated carbon has been used
to treat various types of industrial wastewaters, it has not been commercially
applied to  wood preserving effluents.  Laboratory  tests  have,  however, shown
that activated carbon provides good phenolic removals from raw creosote waste-
water.  Unfortunately,  other  organics are also adsorbed  and  these reduce the
useful  life  of the  carbon to  the  point where  regeneration  is  essential.
Carbon regeneration equipment is generally too expensive for individual plants
to install.

     Carbon regeneration has been shown to significantly affect the adsorption
process.  DeJohn  and  Adams (7) found  that  the  phenol and COD adsorption per-
formance of bituminous  coal activated carbon was sharply reduced after regen-
eration.  Data supporting this finding are shown in Tables 39 and 40.
                                      -93-

-------
                  TABLE  39.  EFFECTS  OF  CARBON REGENERATION (7)

Iodine number
Molasses number
Phenol, % removal
COD % removal
Virgin
carbon
950-1,000
230
99.9
71
Regenerated
carbon
560-680
280
63
13
Percent
change
-27 to -44
+22
-37
-82

           TABLE 40.   IMPURITY  REMOVAL DECREASE AFTER REGENERATION (7)

Impurity removal
(Ib/lb carbon applied)
TOC
COD
Phenol
Virgin
carbon
0.17
0.73
0.01
Regenerated
carbon
0.096
0.35
0.03
Percent
change
-44
-52
-25

     As  stated, .powdered activated  carbon added directly  to the  wastewater
serves  as  a viable  carbon  adsorption treatment  technique.   This  technique,
however, is  normally  used  in  conjunction with  an  activated sludge  system.
DeJohn and  Adams (7)  have reported on four full  scale activated  sludge systems
which have  tried  this treatment  alternative.  At plant No.  1,  a maximum BOD
removal of 90-95 percent  was  achieved with carbon addition vs.  23 percent for
the post test  control.   Similarly,  the effluent COD  was reduced from an aver-
age  of  1,800 mg/1 without  activated  carbon  to 350  mg/1  with  addition.   At
plant No.  2,  the  raw waste  TOC varied  between 100  and  1,000 mg/1.   After
activated carbon addition, however,  the  effluent TOC was maintained  below 20
mg/1, even during shock  load  periods.   Plant No. 3  involved  a  refinery waste
in which effluent TSS and COD  removals improved 40 percent when activated
carbon  was  added  to  the  second  stage of a two  stage  conventional activated
sludge  system.   Similar  improvements  were  also noted  at plant  No.  4.   It
should be noted that  even though the above results  are  expressed  in terms of
TOC and COD removals,  there  was a correlation between such removals  and phenol
reduction.(8)
                                      -94-

-------
     Rizzo and Shepherd  (9)  have outlined the various factors which should be
considered in an  economic  evaluation of a  contacting  type  adsorption system.
These factors have been broken down into three general categories:

     1.   capital cost of the contacting system;

     2.   capital cost of the reactivation system; and

     3.   operation and maintenance expenses.

     Figure  27 presents  estimated equipment  costs  (installed)  for two stage,
fixed  bed,  downflow  contacting systems  of various  capacities.  Figure  28
presents  the equivalent  costs  for  onsite  regeneration  facilities.   Typical
examples  of operation  and  maintenance  expenses  are  presented  in Table  41.
Since these items are extremely variable and  difficult to estimate, no attempt
was  made  to  develop typical  costs.   Instead  the report itemized the  major
components  of each  cost  and provided guidelines to assist in estimating them.
                                       -95-

-------
        TOTAL  INSTALLED  COST  OF TWO  STAGE

        ADSORPTION   CONTACTING EQUIPMENT (9)
      6 -i
 —    5 -
 to
 K
 o

 ~    4 -
 o

 o
in
 O
 OT
 O
 O
      3 -
      2 -
TOTAL INSTALLED COST

WITH ONSITE STORAGE
                                TOTAL INSTALLED COST WITHOUT

                                ONSITE STORAGE
                     23456



                             FLOWRATE , I05gal/d
                                                        10
                                                                  FIGURE 27

-------
TOTAL  INSTALLED  COST  OF CARBON  REACTIVATION

                    AND HANDLING  SYSTEM (9)
     o
     •o
     O
     O
16




14 -



12



10




 8-



 6



 4




 2



 0
                                                       •TOTAL INSTALLED  COST
                       2000      6000       10,000     14,000


                                     CARBON USAGE, Ib/d
                                                     18,000
                                                                     FIGURE  28

-------
TABLE 41.  MAJOR ANNUAL COST CONSIDERATIONS
       FOR ADSORPTION OPTIONS (9)

Ons ite
reactivation
Carbon makeup
Depreciation
Electricity
Fuel
Interest
Labor
Maintenance
Monitoring
Steam
Offsite
reactivation
Carbon makeup
Depreciation
Electricity
Interest
Labor
Maintenance
Monitoring
Reactivation cost
Transportation
Throwaway
carbon
Carbon disposal
Carbon makeup
Depreciation
Electricity
Interest
Labor
Maintenance
Monitoring
'
Service
Electricity
Labor
Service fee






                   -98-

-------
                                  REFERENCES
1.    Mulligan,  T.J.   and  R.D.  Fox.   "Treatment  of Industrial  Wastewaters"
     Chemical Engineering, 83(22):  49-65,  1976.

2.    Memorandum, M.P. Strier  to  R.B.  Schaffer.   "Treatability of Pentachloro-
     phenol." April 24,  1978.

3.    Zogorski,  J.S.  and Faust,  S.D.   "Waste Recovery:  Removing  Phenols  via
     Activated  Carbon."  Chemical  Engineering Progress,  73(5):  65-66,  1977.

4.    Hager, D.G.   "Wastewater Treatment via Activated Carbon." Chemical Engi-
     neering Progress, 72(10): 57-60,  1976.

5.    Bernardin,  F.E.    "Selecting  and Specifying  Activated-Carbon-Adsorption
     Systems." Chemical Engineering, 83(22): 77-82, 1976.

6.    Thompson,  W.S.   "Wood  Preservatives   and  the  Environment:  The  Treating
     Plant."  Presented  at Conference  of  American Wood Preservers Association,
     1974.

7.    DeJohn,  P.B.  and  Adams,  A.D.   "Activated  Carbon  Improves  Wastewater
     Treatment." Hydrocarbon Processing,  54(10): 104-111,  1975.

8.    FMC Corporation Bulletin. "Industrial Waste Treatment."

9.    Rizzo, J.L. and A.R.  Shepherd.  "Treating Industrial  Waste with Activated
     Carbon." Chemical Engineering. 84(1): 95-100,  1977.
                                      -99-

-------
6.0  STRIPPING OPERATIONS

     Stripping operations involve passing a gas or vapor through a liquid with
sufficient contact that volatile components are transferred from the liquid to
the gas phase.  The  driving  force for such an  operation is the concentration
differential between the liquid and concentrated equilibrium point of the gas.
The transferred compound may  then  be recovered  by  condensing the  stripping
vapor.

     It may be noted  that stripping is  essentially the reverse of adsorption.
The equipment utilized  for stripping  operations generally consists of a tower
or mliimn containing multiple bubble-cap or perforated plates.   Packing of the
tower, with  a high surface  area  medium,  is also a reported practice. (1)   An
external condenser cools the  gas.

     As in  many  other cases  of  mass  transfer,  a more  complete  separation of
components may be obtained  through  refluxing of  one or both of  the process
streams.  In  this  case,  reflux of stripped condensates  is  a  common practice.

     The  principal processes  investigated  dealing with wastewater  treatment
have been air and  steam stripping.  A single reference was noted (2) reporting
favorable phenol  removal with  carbon dioxide stripping at  elevated tempera-
tures (65 to 90°C).

     As evidenced  by  the  lack of published data,  air stripping  has failed to
exhibit the  capability  to  remove phenolic  compounds from' wastewater  within
reasonable contact times.  Air stripping was, however, capable of removing 90
percent of  total  reduced sulphur compounds and  10 to 20 percent  of BOD from
contaminated  condensate  streams.(3)  Effective  odor removal  has  also  been
noted.(1)

     Steam  stripping  is  currently practiced at  a number  of  pulp  and paper
mills, but mainly  as  an odor control, water  reuse, or chemical recovery mea-
sure.    These  mills  include  Great  Northern  (Port  Edwards, Wisconsin),  Mead
(Escanaba,  Michigan), Weyerhaeuser  (Springfield, Oregon),  Scott  (Somerset,
Maine),  St.  Regis  (Rhinelander, Wisconsin),  Georgia  Pacific  (Bellingham,
Washington),  Flambeau  (Park  Falls,  Wisconsin),  and  Simpson  Lee  (Anderson,
California).  Very little data,  however,  were found  in the search relating to
phenol stripping.   Work is currently underway in this  area.(4)

     The relative  steam  volatility  of organic compounds is a complex property
involving water solubility, boiling point, vapor pressure and tendency to form
hydrogen bonding.    Organic  compounds that  have undergone  ionization are  not
volatile to steam.  For  example,  at alkaline pH,  the hydroxyl (OH) groups of
phenolic  compounds  would tend  to ionize to  the  negatively charged phenolate
ion and would not  be removed by steam stripping.   Thus the relative acidities
and process  pH should be major  considerations  .in  attempts  to  apply steam
stripping for phenol removal.
                                      •100--

-------
     A plot of  relative .vapor pressures is included as Figure 29 illustrating
one  aspect of  the  sequential  susceptability  to  steam stripping  of organic
compounds.  In  general,  steam volatility  increases  as boiling  point,  water
solubility and hydrogen bonding decrease.

     Steam stripping of pulp and paper mill wastewaters is capable of removing
60 to 85% of  the influent BOD while  reducing TRS (total reduced sulpher com-
pounds),  and  toxicity by  95 percent.(3)  Recovery of  the  column streams may
also  result  in  salable products  such as  methanol,  ethyl acetate,  and fur-
fural.(l)

     Operations  utilizing steam  stripping were  reported  (3)  to  maintain a
steam usage of  about 15 to  18 percent (by weight)  of contaminated condensate
flow.   Rates  as low as 6  percent have also  been reported. (5)   A two percent
rate was  calculated as the  optimal economic  rate by Maahs et al.  A 1967 cost
estimate  for  a 130,000  Ib/hr stripping system was  $84,000  with annual oper-
ating costs amounting to about $0.32/ADT.

     A  more  recent study  (1977)  by  Guttierrez  et  al (6)  quantified steam
stripping  efficiency  for  phenol.   Stripper  steam/condensate  ratios varying
from  5.6 to 14.4  percent were noted in  a  survey  of  five  installations,  as
shown in Table 42.   Table 43 indicates the  variability in  phenol  content of
the  mills' condensate  streams.    It  was  reported  that the  lower  steam/con-
densate  ratios  greatly affected  BOD5_ removal  which decreased from  the pre-
viously  noted 60  to 85 percent to less than  5 percent.  Phenol removal ranged
from  0.4 to 61  percent,  but this variability  was attributed  to the rectifi-
cation  section  rather  than  the  steam  ratio.  Table 43 presents the removal
data.   The best  phenol  removal,  61 percent,  still was not  considered high
enough  to qualify  steam  stripping as an  effective  removal  technology.  Con-
tinuing work in this area may refute this  conclusion.
                                      -101-

-------
                              VAPOR  PRESSURES OF VARIOUS


                              VOLATILE  ORGANIC COMPOUNDS
o
NJ
              CO
              LJ
              OL
              UJ
              co
              o
                 100


                  50







                  10
1.0
                  0.5
                 0.05
                 0.01
                   -50
                     x^X
/w*
        kHV
X*
                    50        100


                        TEMPERATURE °C
                                                          150
                  200
              250
                                                                      FIGURE 29

-------
                         TABLE 42.
                                    CHARACTERISTICS OF VARIOUS CONDENSATE TREATMENT SYSTEMS AT
                                   	FIVE BLEACHED KRAFT HIU.S (6)           	
    Mill
    code
Production
 (ADT/day)
                        Wood furnish
 Type of
digesters
                                                         Contaminated condensates being treated
                                                                                                           Stripping system
                                             Flow
                                                             Origin
% of total
condensate
                                                                                                             General characteristics
Steam
Ratio %
OJ
   970       Mixed fir,     Kamyr     Digester area: Steam vessels       547         64
             balsam,                  and #2 flash tank; turpentine
             hemlock,                 decanter underflow; digester
             spruce                   area vents.
                                      Evaporators area: Evaporator
                                      effect #5; evaporator surface
                                      condensor & vacuum system'.

   450       Mixed          Batch     Digester area: Digesters' blow     100         31
             hardwood                 condensate.
                                      Evaporators area: Evaporators'
                                      surface condensor and vacuum
                                      system.

   510       45% spruce     Kamyr     Digester area: Steam vessels       103         12
             45% pine                 and flash tank #2.
             10% balsam               Evaporators area: Secondary
             fir                      surface condensors and vacuum
                                      system.

  1100       86-80%         7-Batch   Digesters area: Digesters'         493         24
             Douglas fir              blow condensate.
             14-20% pine    1-Kamyr   Evaporators area: Surface
                                      condensor and vacuum system.

   280       69-70% pine    Batch     Digesters area: Digester's         235         46
             40-30%                   blow condensate and turpen-
             hardwood                 tine decanter underflow.
                                      Evaporators area: Evaporators'
                                      effect #5; surface condensors
                                      and vacuum system.
                                                                                                             Free standing stripper using       5.6
                                                                                                             live steam, no condensor,
                                                                                                             reflux or decanter.
                                                                                                             Free standing stripper using       6.4
                                                                                                             live steam. Has rectifying
                                                                                                             section.
                                                                                                             Free standing stripper using      11.3
                                                                                                             live steam and some digester
                                                                                                             relief steam, with condensor
                                                                                                             decanter and reflux.
                                                                                                             Free standing stripper using      J3.3
                                                                                                             live and relief steam. Column
                                                                                                             has rectifying section.
                                                                                                             Stripper built into the 14.4
                                                                                                             evaporator train between 1
                                                                                                             and 2 effect.  Uses steam
                                                                                                             from 1st effect and some
                                                                                                             live steam.

-------
          TABLE 43.   POLLUTANT LOAD BY CONTAMINATED COMPENSATES * (6)
Source
Literature
Mill A
Mill B
Mill C
Mill D
Mill E
PH
range 7.5 to 10.5
8.3
10.0
9.6
8.3
9.9
Phenol
fag/1)
4 to 40(6>13)
5.8
33.2
17.6
26.6
29.1

(Ib/ADT)
0.03 to 0.3**
0.04
0.09
0.04
0.14
0.29

       *Contaminated condensates analyzed are those segregated by the mills
        for steam stripping.

      **Calculated on the basis of foul condensates flow of 1,000 gal/ADT.
                 TABLE 44.   REMOVAL OF PHENOL FROM CONTAMINATED
                      CONDENSATES BY STEAM STRIPPING (6)

Phenol
Mill
source
A
B**
C
0**
E
Steam ratio
(%)
5.6
6.4
11.3
13.3
14.4
Feed
(mg/1)
5.8
33.2
17.6
26.6
29.1
Treated
(mg/1)
5.0
15.8
13.1
10.0
26.2
Removed
(%)*
9.0
49.0
17.0
61.0
0.4
Removed
(Ib/ADT)
0.0036
0.0441
0.0068
0.0854
0.0012
 *Corrections have been made for the dilution effects by
  condensed steam.

**Column equipped with rectifier.


                                      -104-

-------
              AIR   STRIPPER  SCHEMATIC (3)
CONT. COND.
        FEED TANK
                                                        SEPARATOR
                                                       TO INCINERATOR
                                                      FAN











                                        COLUMN





                                        t	AIR SUPPLY
                                        STRIPPED
                                        CONDENSATE
                                                FIGURE 30
                                105

-------
            STEAM  STRIPPER  SCHEMATIC (3)
           STEAM
     ;LEAN
    CONDENSATE
                       CONDENSATE
                       PRE-HEATER
                                               CONDENSERS
                                                         COLD WATER
                                                    HOT >•
                                                              GAS
                                                    WATER
FEED TANK
REFLUX TANK PRODUCT TANK
                                                 FIGURE  31
                                 106

-------
.FLOW DIAGRAM
STRIPPER COLUMN INTEGRAL W/ EVAPORATOR (3 )
TO LIME KILN CONDENSER
___
4T--TT 	 -*] * 	 ~+\ r~*\ { 	 1 CONDENSER
| j 	 * -*Q [_.
1 	 J 1 1 1 1
STRONG B.L.
STEAM
COND.

i ,

_f-p
DIG. CONDENSATE METH
V
t i LJ-
i
*) 1 *I '
t> 2 L 3 t_ 4 ^5
t r ...^ . i
-r r-r- -r- r-7- -^ '
STRIplpER 1 ' ' " ''
!
• i . , . (
ii i *
r-n 1 'I
_Tl-J T
r'T-r' L
i I ' l — »
»L
r I 1 	
U-| I 1 f1 	 1 FOUL CONDENSATE
— 1 j 1 — 1 I
ANOL REFLUX WEAK B.U
FIGURE 32
J

-------
             STRIPPING COLUMN  DESIGN DETAILS (7)
             VAPOR
   FEED
NOZZLES
              /  N
             I    )
                         REFLUX
                        •MANWAYS
                         ALTERNATE
                         PLATES
 STEAM
             s  \
                             "STEEL
                              SHELL-
            BOTTOMS
                                             .DOWNCOMER
                                                     FIGURE 33
                                                                  LIQUID
                                                                  FLOW
                                     108

-------
                                 REFERENCES
1.    Rowbottom, R.S.  and J.G. Wheeler.   "Condensate Stripping and  Waste  Gas
     Incineration as an Odor Control Measure at Domtar's Cornwall  Kraft Mill."

2.    Schutt, H.  "Device for Purification of Wastewater from an Organic Chemi-
     cal Preparation Process."  Chemical Abstracts,  Vol.  80  (112285),  1974.

3.    Hough, G.W.  and R.W.  Sallee.   "Treatment of  Contaminated Condensates."
     Tappi 60(21): 83-86, 1977.

4.    E.G.  Jordan  Co.,  Inc.  (Contractor), Portland,  Maine.   EPA Project  No.
     68-03-2605.

5.    Willard, H.K.  "Pulp Mill Wastewater Stripping by Air & Steam." Paper  and
     Forest Industries  Research,  National Environmental Research Center, EPA,
     Corvallis, Oregon.

6.    Gutierrez, L.A., J.C. Mueller, and C.C. Walden.  "Steam Stripping of Con-
     densates  for Removal of  Odor,  BOD 5 Toxicity &  Fish  Tainting Propensi-
     ties." Paper presented  at CPPA Environment Improvement Conference, Monc-
     ton, NB, November  1977.

7.    Foust, A.S.  L.A.  Wenzel  C.W.  Slump, L.  Mails, and  L.B.  Anderson, Prin-
     ciples of Unit Operations Wiley &  Sons,  Inc., New  York, 1960.  578  pp.
                                      -109-

-------
7.0  RESIN ADSORPTION

     Synthetic  polymeric  resins  are  known to  not  only effectively  remove
phenol but  also to permit recovery  of this valuable chemical.   Crook  et al,
(1) did extensive work with industrial wastewater effluents containing phenol,
bisphenol A and  p-nitrophenol.   One  of their  studies  found that  Amberlite
XAD-4  reduced  the wastewater phenol  concentration from 6,700 to  less  than 1
mg/1.  Acetone or methanol was used for regeneration.   Both acetone and phenol
(99  percent  purity)  were recovered  by subsequent distillation.   It  was' also
noted  that  bisphenol A could be  removed by adsorption onto  a  combination of
XAD-4  and XAD-7.   Adsorption capacities  of each  resin  for  bisphenol  A were
found  to  be 34  and  16 g/1,  respectively.   In this case, resins  were  regen-
erated with  ethanol.   Both Herve  (2) and Chamberlain  (3)  had similar success
using resin adsorption for phenol  removal.

     Rohm and  Haas  (4)  has  stated  that  the phenol  adsorption  capacity of
polymeric resins  depends  on  the  type and  concentration  of  phenolics  in the
wastewater,   as well   as  the pH,  temperature,  viscosity,  polarity,  surface
tension  and background  concentrations of  other  organics  and  salts.   Resin
adsorptive capacity changes  dramatically with  increasing  pH.   One explanation
for  this  is that  the  phenolic  molecule  changes from a  neutral,  poorly dis-
sociated form  at  low pH  to an anionic,  charged  form  at  high pH.  Kim  et al.
(5), based  on their work  with  p-nitrophenol  (PNP),  support this  statement.
They also found,  with  weak  acids, however,  that as pH  decreased below a cer-
tain value,  the adsorptive capacity  was  a "maximum in the pH region in which
the resin was predominantly in the free base form and  the  PNP was present as a
neutral species."  The  effect of pH on the adsorption  of PNP  to  a weakly basic
resin is shown on Figure  34.

     A high salt background has  been  noted to enhance  phenol  adsorption.  Rohm
and  Haas conducted  experiments  in  which  aqueous  solutions of phenol  and
m-chlorophenol,  both  with  and without  salt  addition,   were passed  through
columns  of  XAD-4.   The results,  as  shown in Table 45, clearly  indicate that
the adsorptive capacity for both solutes  was greatly increased by the presence
of salt.
                                      -110-

-------
EFFECT  OF  pH  ON  ADSORPTION  OF  P-NITROPHENOI_(4)
                   "0 P
                i  .0-
               X
               I
               I  .0
                  10
                                                 Pr*> On AclivoKd
                                                 Carbon (Rtfi Z9, 30)
                                       7      9
                                          pM
PTunol On
Activottd Carbon
(R«t. XI)
                                                    Phtnol On ES-J3
                                        I      I       I      I
                                                           13     19
                                                                 FIGURE  34
                                           111

-------
                 TABLE 45.   EFFECT OF SALT CONTENT UPON CAPACITY
         OF AMBERLITE XAD-4 FOR PHENOL & m-CHLOROPHENOL ADSORPTION (4)

                                            Solute in influent
                                         solute  adsorbed,  Ib/ft
Solute
Phenol
Phenol with 13% NaCl
m- chl o r opheno 1
m-chlorophenol with 13% NaCl
ppm
250
250
350
350
milli-
mol/1
2.7
2.7
2.7
2.7
Zero
leakage
0.78
0.98
2.40
3.10
10 ppm
leakage
0.83
1.09
2.53
3.43

     As with  carbon  adsorption, solute solubility  is  reported to serve  as  a
guide  to  the  adsorptive capabilities of  XAD-4.   "This is illustrated  by the
adsorption  of phenol,  monochlorophenol;  2,4  dichlorophenol  and  2,4,6  tri-
chlorophenol.   The solubility of these  compounds  in water decreases  with the
level  of   chlorine  substitution,  but the  adsorptive   capacity  exhibited  by
Amberlite   XAD-4   for  them  increases with  the  level   of  chlorine  substitu-
tion. "(4)

     As shown in  Table  46,  the  specific adsorption  of  phenols  and  chlorinated
phenolics  on Amberlite XAD-4  resin has been researched. Even  with this data,
however,  column  studies should be  performed to  select the  best resin and
establish operating parameters for  specific  industrial  wastewaters.

     Kim et al,  (5) in addition  to  their work with specific  resins,  summarized
various feasibility   studies  for  using  resins  for phenol  adsorption.   One
summary reported the  work of  Anderson and Hansen in which  the hydroxide form
of strong  base resins  were  found to have much greater adsorption  capacity than
the  chloride form  of  the  same resin, or  of a weak  base  resin.   They also found
that the  strong  base resin could-not be  regenerated as efficiently  using an
alkaline solution or  polar  solvent as a weak base  resin.   The work of Pollis
and Kunin was also included in  the summary.   These  researchers found that the
free base  form  of weak base  resins  had  an  appreciable phenol adsorption ca-
pacity  and it  could  be regenerated  with either polar solvents or  caustic.
Adsorption was also found  to be  independent  of the organic  salt concentration.
                                                                          •
     To regenerate a  resin  adsorbent,  the  attractive forces  between the phenol
molecules  and the resin must first be overcome.   Numerous researchers  have
stated that this  is normally  accomplished with either  a caustic  solution or a
polar  solvent for non-ionic  adsorbents. (1,  2,  4,  5)   Kim et al  stated that
acidic  or  salt  solutions  may also be used to regenerate ion exchange resin
systems following treatment  of phenolic  bearing wastewaters.
                                      -112-

-------
         TABLE 46.  ADSORPTION OF PHENOLIC COMPOUNDS ON AMBERLITE XAD-4
	AT 25°C AND FLOW RATE 0.5 gpm/ft	

Compound                Influent concentration     _ Resin capacity
	(ppm)	(Ib/ft  at 0.1 ppm leakage)

Phenol                          6,700                       5.4

Phenol                          3,000                       4.5

Phenol (in 13% sodium             250                       1.0
  chloride)
Phenol
m-Chlorophenol
2 , 4-Dichlorophenol
2,4, 6-Trichlorophenol
250
350
430
510
0.8
2.4
5.1
12.0

     As  previously  described in the  Solvent Extraction  section,  Lorton (6)
evaluated the economics  of  removing phenols from coke plant process condensate
using biological oxidation,  solvent extraction and adsorption.  Of these three
alternatives,  solvent extraction was  the  most  economical followed by adsorp-
tion  and finally,  biological oxidation.  The  results  of this evaluation are
shown  in Table  31.  The  adsorption  system used in the analysis was developed
by  the Rohm and Haas  Company and is  shown on  Figure  35.  This process, with
the exception  of the superloading step, has  been  proven at two chemical pro-
cessing  plants.  One  such plant has been in operation since August 1975 and is
reducing wastewater phenolic concentrations from 6,000 to 20,000 mg/1 to less
than 3 mg/1.
                                       -113-

-------
                                              ROHM  a  HAAS  ADSORPTION   PROCESS  FLOW  DIAGRAM (6)
o o

 . o
 m —
   o
                              MOCCU COMMUTE    KIIN ADIOWTKM  ««fMiuiiaifnui   nuciKMTMHfO  »lucTkHUTmmw   nucticttMai      MUH-mfML      FHACIIOMAIM    'MCIICH.TM      macnauTM      MfiOM
                                 «IHU>a(ll          MUM       CONMNSm        DNUH       K1IIM9I11IIUUH   BOTTM! CWA1H
                                 Nail.        *ut* vai.ri. rTh«a  MIMDEW.H.



                                                            ".^t-^
                                                                                                                                                                                 MCNOL ITOfUM
                                                                    :p~i
                                                                     t	1.«>_^  »«••_  J    j-o
                                   I !• 00*1 •< MCTCHM TNI Ifluioul *-*
                                   rtl  fltlfCn. UMNIIM TO TM M1IMM3CM
                                   '*IM TO M«»IMMO« fHI Mllh LMV1H
                                                                       I tHl tMIDDII) llllHilf MUfiM
                                                                                                                                                                   HimiM    JCITDC PUMP
                        HkllMUM no*
«ATI
MO
IIO
4M
4M
• M
CMMIWTI
I,OM
DOO
KM.I10
•04^1M
I.MO
«MH.Mri
IOO
I,*...
IliJH
•in
«KU
"»
NkMO
• M«0 .
iro
jL
•\000
«../.
»M»i
ita
1MW4JW
I
1 !•,«!•
;»,4tl

< Mtiaa
IliO
at
1 1.010
11.1 u
w
1LITUM
,,,.™
IW44O
rM.40U
'M
^
•l.ooa
«4,«40
IUO,I»
'•°
IV\ "
.*?'
nuinc
4O
>i«,i4r
11*400
*••
MOMbf
:
•.1*0
1,110
10
                                                                                                                              PRELIMINARY
                                                                                                                                                               FIGURE  35

-------
                                  REFERENCES


1.    Crook, E.H.  et  al.   "Removal  and Recovery  of Phenols  from  Industrial
     Waste Effluents  with Amberlite  XAD Polymeric  Adsorbents."  Industrial &
     Engineering Chemistry Product Research and Development,  14(2):  113,  1975.

2.    Herve, D.   "New Procedure  for  Treating Wastewaters  Containing  Phenol."
     Chemical Abstracts, Vol. 77, (143609), 1972.

3.    Chamberlin, T.A.   "Oxazoline and/or Oxazine  -  Modified  Polymers."  Chem-
     ical Abstracts, Vol. 86, (141260), 1977-

4.    Correspondence  with the  Rohm and  Haas Company, Philadelphia,  Pennsyl-
     vania .

5.    Kim,  B.R.  et  al.   "Adsorption of Organic Compounds  by Synthetic Resins."
     Journal Water Pollution Control Federation, 48(1): 120-133, 1976.

6.    Lorton, G.A.   "Removal  of Phenols from Process Condensate." U.S. Depart-
     ment  of Energy Contract No. EX-76-C-01-2240,  October 1977- 24 pp.
                                       -115-

-------
8.0  ELECTROCHEMICAL OXIDATION

     Electrochemical oxidation,  sometimes  referred  to  as electrolysis,  in-
volves the generation of  an electrical potential across a  cell through which
wastewater is  flowing.    As a  wastewater  treatment  method,  this  technology
dates back at  least as far as  1900. (1)   Available data indicate that  no  or-
ganic decomposition  occurred  in these  early installations, but it  was noted
that  significant amounts  of  sodium hypochlorite  were generated   from  lead
dioxide,  graphite and platinum  electrodes.   The sodium hypochlorite may have
caused interferences with the analytical  techniques  employed.   It  was further
noted that low conductivity of  wastewaters  led to  uneconomical  power consump-
tion.

     More recently,  Westinghouse Electric  Corporation (2)  reported  100 percent
removal of phenols  by electrolysis  in  the  presence of an  oxidative catalyst
(MnO  or  CrO»)  suspended as  agglomerated  particles  in  the   reaction  cell.
Graphite electrodes  were  used  with  a  0.5 to  800 Hz  current  being passed through
the cell.  The wastewater,  a  coking  effluent containing 200 mg/1  phenol,  was
treated in this  manner for  one  hour  and  resulted  in  the  complete  elimination
of detectable phenols.

     A Russian effort (3)  investigated electrochemical oxidation on biological
treatment plant  effluents,  black liquor,  first stage  alkali extraction efflu-
ents  and  evaporator condensates.  Additionally,  a variety of  electrode  ma-
terials  and   electroconductive   additives  were  studied.   Phenolic  removals
ranged from  78 to 100  percent.   Current  densities and  pH  were  noted as major
operating parameters; 2 to  4  amp/dm   showed marked improvement in  COD reduc-
tion while a slightly alkaline  pH, 7 to 9,  was reported as suitable with lead
electrodes.   Sacrificial  aluminum anodes were  reported  to be useful  for  a
greater range of both current  densities and pH.  Less power was required with
this arrangement also.   A major drawback of  the  process still  appears  to be
excessive power consumption.
                                      -116-

-------
                                  REFERENCES
1.    Miller, H.C. and W.  Knipe "Electrochemical Treatment of Municipal  Waste-
     water." U.S. Department  of Health,  Education and Welfare,  Public  Health
     Service, Report AWTR-13,  March 1965.

2.    Westinghouse Electric Corp.   "Apparatus  and  Method  for  Removing  Oxi-
     dizable Contaminants  from an  Aqueous  Medium."  Chemical Abstracts Vol.  87
     (188952), 1977.

3.    Serbodol Skii,  E.N., M.I.  Anisimova,  V.A. Babkin,  and  G.N.  Permyakova.
     "Purification  of  Kraft Pulping Effluents  by Electrochemical  Oxidation."
     Abstract Bulletin of the Institute of Paper Chemistry,  49(5):  entry  No.
     4099, 1978.
                                      -117-

-------
9.0  IONIZING RADIATION

     Sufficient excitation of organic compounds in wastewater through exposure
to ionizing radiation can lead to their oxidation.  Lenz et al (1) studied the
effect of gamma  irradiation  on pulp mill effluent  constituents  while varying
aeration parameters.   The  studies  indicated  (via COD  reduction  data)  that
oxidation of  organic compounds  is  accelerated by gamma irradiation.   It was
noted that oxygen transfer  rates were the limiting reaction rate factor since
turbulent air  flow produced  better  results than  static high  pressure oxygen.
The authors expressed doubts about the economic feasibility of the process due
to  the  high  cost  of  radioactive  source  materials.    Table  47  contains  the
results of their investigations.

     Sunada  (2)  also reported favorably on the phenol  reductions experienced
when a chemical plant effluent was exposed to  0.01 Mrad.

     Touhill et al,  (3) studied the  effect of  a 175,000  rad/hr dosage of gamma
radiation from a cobalt-60  source on oxygenated phenol solutions.   The solu-
tions varied  in  phenol  concentration from 10  to  1,000  mg/1.   This-study also
indicated that oxidation  was  limited by oxygen transfer rates  since the rate
of phenol removal was  relatively independent  of initial concentration.   Solu-
tions containing 1,000  mg/1  phenol  required 20 hr  for  99  percent phenol des-
truction.  When oxygen was added to  the solution 75 percent phenol destruction
of  a  100 mg/1  solution occurred in  60 minutes.   At a lower  initial concen-
tration  of  10  mg/1, an  oxygenated  solution  experienced  95  percent  phenol
destruction within 30 minutes.   No  significant difference between  the  use of
air  or  oxygen was noted, although  the rate  of  degradation  was  greatly  ac-
celerated by  the addition of either gas at the  lower  initial concentrations.

     Higher  dosages  of  radiation  resulted in higher  rates  of  degradation.
                •
     Nenodovic (4)  prepared  an article setting forth  the  principles and prac-
tical applications  of  ionizing radiation for  wastewater treatment.  Reduction
of phenols is referred to as a viable application.

     Most  researchers  concluded  that  the  treatment  was  effective, but  too
costly.   An apparent  reason  for this was that the chain reactions expected in
pulp and paper effluents failed to occur.
                                      -118-

-------
             TABLE 47.  EFFECT OF GAMMA IRRADIATION AND OXYGEN PRESSURE ON pH,
                SUSPENDED SOLIDS AND COD OF VARIOUS PULP MILL EFFLUENTS (1)
          Irradiation
Sample     intensity
  no.  (million rad/hr)
                   Pressure     Time
             Gas   (p.s.i.)   (hr:min)	pH
                              Suspended
                               solids
                               (mg/1)
                                COD
                               (mg/1)
                                                                         COD
                                                                       removal
A.  Kraft strong effluent
 1-1
 1-2
 1-3
 1-4
 1-5
 1-6
Control

   8
   8
   8
   8
   8
 1,900
 2,100
 2,100
 2,000
   500
 2,100
4:00
0:30
1:00
1:40
2:00
2:00
9.08
8.19
7.52
7.30
7.33
7.05
                                                      7.30
 48
264
 34
 60
 74
104
 59
3,218
2,846
2,461
1,920
1,379
1,092
1,152
                                                    11.6
23.
40
57.
66
                                                                                    64.2
B.  Kraft strong effluent
 II-l
 II-2
 II-3
 II-4
 II-5
 II-6
Control
6
8
4
8
8
—

-
-
air
bubbled
air
C12
Cl
bubbled
-
-
2,100
-
2,100
2,100
_

                                1:30
                                0:30
                                  00
                                  00
                                  30
                      9.08
                      8.40
                      7.58
                      8.64
                      6.06
                      2.98
                      3.73
                      48
                      44
                     140
                     152
                      78
                     642
                     680
                     3,218
                     2,759
                     2,410
                     1,260
                     1,841
                     2,442
                     2,875
                   14.5
                   25.1
                   60.8
                   42.8
                   24.1
                   10.6
C.  Kraft weak effluent
 III-l
 III-2

 III-3
 III-4
 III-5
Control
   6
   4
          bubbled

            1
1,900
2,000
2,000
1:30
0:20

4:00
0:20
2:00
                      6.84
                      6.61
                      7.28
  02
  25
6.82
166
128
152

160
120
174
500
387
260

421
348
245
22.6
48.0

15.8
30.4
51.0
D.  NSSC main mill effluent
 IV-1
 IV-2
 IV-3
            Control
                     1,500
                     1,500
            2:00
            2:00
          6.02
          6.18
          6.78
           112
           148
           104
          862
          715
          130
         17.1
         84.9
E.  NSSC Woodroom Effluent
 V-l
 V-2
 V-3
Control
                     1,500
                     1,500
            2:00
            1:00
          5.31
          7.12
          6.51
           680
           236
           116
          2,911
          1,075
            482
         63.1
         83.4
                                           -119-

-------
                                  REFERENCES
1.    Lenz,  B.L. et al.   "The  Effect of  Gamma  Irradiation on Kraft and Neutral
     Sulfite Pulp  and Paper  Mill  Aqueous Effluents."  Pulp and Paper Canada,
     72(2):   75-80,  1971.

2.    Sunada, T.  "Wastewater Treatment  by Radiation."  Chemical  Abstracts  Vol.
     80 (112211),  1974.

3.    Touhill et al.  "The  Effects  of Radiation  on  Chicago Metropolitan Sanitary
     District Municipal  and  Industrial  Wastewaters."  Journal Water Pollution
     Control Federation.  41(2), Part 2): R44-R60,  1969.

4.    Nenadovic, M.O.  Micic, 0. Gal.   "Wastewater Treatment  by Use of Ionizing
     Radiation." Hem.  Ind.  28  (5):  217-220,  1974  (Russian).
                                     -120-

-------
10.0	ELIMINATION OF POLLUTANT DISCHARGE

     To  date,  the  only published  technology  for the  total  elimination  of
phenolic discharges  is  effective indirectly in that  it  involves  the elimina-
tion of any wastewater effluent.  A demonstrated technology to accomplish this
feat involves the Rapson-Reeve salt recovery process, substitution of chlorine
dioxide  for  a majority  of the  chlorine  used in  the bleach plant  and major
reductions  in water  usage by  complete  counter-current bleach plant washing
followed by  brown stock washing with the  same water.  This process is cur-
rently being  pioneered at  Great Lakes Paper  Company's  Thunder Bay installa-
tion - an 800 ton/day kraft pulp mill.

     In  the  D/CEDED bleach plant,  70 percent of  the  chlorine  demand is sup-
plied  by chlorine dioxide  rather  than chlorine.   This  results not  only in a
reduction in  chloride  ion  within the system (an approximate 80-percent reduc-
tion)  but also  maintains  pulp  quality  at the  higher temperatures  counter
current washing imposes.  Water use in the bleach plant  is anticipated to drop
to 4,000  gal/ton  of pulp.   This 4,000 gal/ton will be recycled as brown stock
wash water.

     The major innovation is the Rapson-Reeve salt recovery process.  Complete
reuse  of the  bleach  plant  white  liquor would  induce  the recycle  of large
quantities of  salt.   Recycling of the salt would  result in a build-up of the
salt to saturation levels within the system, ultimately  causing problems.  The
recovery  process  begins  by  concentrating  the bleach plant white  liquor  to
26-30 percent NaOH + Na2S.   This allows for precipitation of Na2CO., and Na SO,,
but not NaCl.  Following removal of these crystals, the  liquor is further con-
centrated  to  36  to 42 percent  NaOH  + Na S which  causes the  NaCl to precipi-
tate.   The supernatant  liquor  is  then  returned  to  the digesters  while  the
precipitate  is purified  and  stocked for  production of NaOH  and chlorine/-
chlorine dioxide.

     Although  this  method  is  not specific for the subject pollutants, its use
would eliminate their discharge to the environment.
                                      -121-

-------
PROCESS  FLOW  DIAGRAM
    CLOSED   PULP  MILL( 1,2,3)
                  UNBLEACHED
                                        BLEACHED PULP
	 --fc 	 — ^-~
•—-^ PULPING • 	
1 	
^EVAPORATORS) ^
A
H20,C02 0
/^L~ "NTO ATMOS. g

/FURNACEV5^" ^ 1
^\ s H f
> — ^ i f
If * I
• f PULPING CHEM ^
1 L_ .,. REfiFNERATION
1 SMELT AND NoCI
| RECOVERY
1
WATER
WOOD
CHIPS

BLEACHING ~~

Na2S04
TO FURNACE
\
H2S04 \
\ )
Cl BLEACHING
^ CHEMICAL
i
o
o

°

u
i-
<
%
0 1
o m

REGENER.









                                    FIGURE 36
                   122

-------
                                 PROCESS  FLOW  DIAGRAM
                   RAPSON/REEVE  SALT   RECOVERY PROCESS ( 1,2,3)
STAGE  I
                                                                     STEAM
   LIQUOR FROM
   STAGE
                                                                                             ^.LIQUOR TO
                                                                                               STAGE 2
                                                                                                   WATER
                                                         CONCENTRATED
                                                         WHITE LIQUOR
                                                         TO DIGESTERS
 PC  PROCESS CONDENSATE

 SC  STEAM CONDENSATE

 H E  HEAT EXCHANGE
                                                                         LEACH LIQUOR
                                                                         TO STAGE I
FIGURE  37

-------
                                  REFERENCES
1.    Stevens,  F.   "First Pollution-Free Bleached Kraft Mill Gets Green Light."
     Pulp and  Paper Canada,  76(10):  27-28,  1975.

2.    Stevens,  F.  "The  No.  2  Kraft  Mill,  and  What's New About  It."  Pulp and
     Paper Canada 77(11): 28-32,  1976.

3.    Stevens F.  "Effluent-Free Bleached Kraft Mill  is Pioneered at Great Lakes
     Paper." Pulp and Paper  Canada,.  78(3):  94-98,  1977.
                                      -124-

-------
     APPENDIX B
ANALYTICAL PROCEDURES
        -125-

-------
                                       B-l

                  ANALYTICAL PROCEDURES FOR TOTAL PHENOLICS

1.0  TOTAL PHENOLICS

     Modified Lowry procedure (Folin-Ciocalteau)  of Markwell,  Mass, Bieber and
Tolbert, Anal. Biochem.  87,  206-210,  (1978).   Details:

     Reagent A:

          2.0 percent sodium carbonate

          0.4 percent sodium hydroxide

          0.16 percent sodium tartrate

          1 percent sodium dodecylsulfate

     Reagent B:

          4 percent copper sulfate pentahydrate

     IN Folin phenol reagent:

          2N Folin-Ciocalteau reagent (Fisher Scientific) 1:1  with
          deionized water

     In the procedure 100 parts  of reagent A  are  mixed with 1 part reagent B
to form reagent C.   Standard curve prepared by  incubating a  sample volume o'f
1.0  ml  containing  10 to  50 micrograms  of  2,4-dichlorophenol  (a  surrogate
standard) with 3.0 ml reagent C  for 10 to 60 minutes at room temperature.  0.3
ml of IN  Folin reagent  is then  added, mixed, and incubated 45 minutes at room
temperature.  Absorbancy is  read  at  660 nm.   Samples are diluted in deionized
water to provide readings within the  standard range.  Note:  pentachlorophenol
does not react in this analysis.
                                      -126-

-------
                                  B-2

       ANALYTICAL PROCEDURES FOR PENTACHLOROPpNOL - MILL A

The  phenol  extraction procedure  for  FID/ECD  analysis  is as  follows:

1.   Place a 25 ml to 100 ml aliquot of each sample in a 125 ml or 250 ml
     separatory funnel and adjust to pH 12 with ION NaOH.

2.   Extract once with 50 ml hexane, drain the water layer into a 250 ml
     beaker and cover with a watch glass.

3.   Back  extract the  hexane  once with  25  ml of  IN NaOH, combine  the
     aqueous layers and discard the hexane.

4.   Bring the  solution to pH <3 with  concentrated  HC1  (use a hood)  and
     transfer to a 125 ml or 250 ml separatory funnel.

5.   Rinse a beaker  with 50 ml of nanograde dichloromethane* then add to
     the separatory funnel.  Extract the aqueous layer and dry the chlor-
     omethane* layer through anhydrous sodium sulfate washed with chloro-
     methane*.   Drain  the aqueous  portion into a 500 ml  flask equipped
     with an ampule.

6.   Perform three  extractions  with 50  ml  of dichloromethane*  as  des-
     cribed in step 5.

7.   Concentrate the organic portion on a steam bath.

8.   Adjust  the  volume to  6 ml and  store in a 2 dram  vial with Teflon
     seal cap.

9.   Analyze via G.C. using the analytical conditions shown below:

          1% SP1240DA
          on 100/120 Supelcoport, 3 ft x 2mm I.D.  glass
          Helium carrier gas @ 30 ml/min  (FID), Argon/methane
          95:5  @ 30 ml/min (ECD)
          Column temperature 80°C 2 min to 180°C/@ 8°/min.,
            Det @ 250°C, injector 250°C
                                 -127-

-------
          All glassware  acid  washed, rinsed  with distilled  deionized  water,
          acetone,  and CH^Cl^.

*Petroleum ether is used instead of CH^Cl^  for ECD analyses.   Standard
 curves prepared using authentic PCP.
                                      -128-

-------
                                       B-3

              ANALYTICAL PROCEDURES FOR PENTACHLOROPHENOL - MILL B
1.   Place 50  ml sample  in 250 ml  separately funnel.   Adjust to pH  <2  with
     concentrated HC1.

2.   Extract with 50  ml  nanograde petroleum  ether.   Collect organic  layer.

3.   Repeat  step  2 twice  with 25  ml nanograde petroleum ether.   Combine or-
     ganic extracts and dry over anhydrous sodium sulfate.

4.   Concentrate  on  a steam bath using a  Kuduna-Danish  concentrator  composed
     of a 500  ml flask,  a  3-state Snyder  Column  and  a graduated 10  ml ampule
     when necessary for FID analysis.

5.   ECD  analysis performed  on extracts  directly  or diluted with  petroleum
     ether as necessary.

6.   Analyze via GC using the conditions shown below:

     1  percent SP-1240  DA on 100/120 Supelcoport,  3  ft.  x  2  mm ID  glass.

     A.   ECD:

          Argon/methane carrier gas (95/5) at 30 ml/min.
          Injector temperature - 250°C.
          Detector temperature - 350°C.
          Temperature program - Isothermal @170°C.
          Volumes injected - 2-10 microliters.

     B.   FID:

          Helium carrier gas at 30 ml/min.
          Injector temperature - 250°C.
          Detector temperature - 350°C.
          Temperature program - 85°C for 2 minutes to 190°C
                                at 10°C/minute increments.
                                      -129-

-------
7.   Chromatographic performance of underivatized pentachlorophenol was accom-
     plished by the following passification procedure.


A.   Glass Column and Injector Sleeve Cleanup

     Aspirate  20  to 25  ml  of  the  following slowly  through in  order given:

     1.   Chloroform or dichloromethane

     2.   Acetone

     3.   Water

     4.   50 percent KOH in water

     5.   Water

     6.   Concentrated nitric acid

     7.   Water

     8.   Concentrated hydrochloric acid

     9.   Water

    10.   85 percent phosphoric acid

    11.   200 to 250 ml water

    12.   Acetone

    13.   Chloroform .

    14.   Dichloromethane

    15.   Dry at 100°C in oven

B. Treatment of Glass Wool with Phosphoric Acid

     Place. quantity  of silanized  glass  wool in  85 percent  phosphoric acid.
Let stand  10  or 15 minutes.  Rinse well with water.  Soak in acetone, chloro-
form, and finally dichloromethane.  Dry in 100° C oven.
                                      -130-

-------
                                 REFERENCES
1.    Unpublished information, I.E.  Acree and B.  K.  Wallin,
                                       -131-

-------
                                       B-4

                  ANALYTICAL PROCEDURES FOR PENTACHLOROPHENOL,
                  GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS)
EXTRACTION CONDITIONS

1.   Adjust pH of 1 liter sample to <2 with concentrated HC1.

2.   Spike with deuterated internal standards* (200 to 400 ppb).

3.   Serial extraction with nanograde dichloromethane (125x50x50 ml).

4.   Break emulsions by glass wool filtration or solvent addition.

5.   Dry  combined  dichloromethane  extracts  with  anhydrous  sodium sulfate

6.   Concentrate sample by Kuduna-Danish evaporation to 1.0 ml.

7.   Add 100 micrograms d.. anthracene.

MASS SPECTROMETER - HP 5985

     amu range:  35450
     Scan speed:  300 amu/sec
     A/D per 0.1 amu:  3

GAS CHROMATOGRAPH - HP 5840

     Column:  SE-30 SCOT, (SGE D grade 40,000 Ngff)

     Flow rate:  22 on/sec at 200°C

     Injection volume:  1 microliter, splitless injection
*phenol - d,, octadecanoic acid - d3_, naphthalene -
                                      -132-

-------
     Temperature program:  30 to 260 at 6°C/min.

DERIVATIZATION CONDITIONS

     Place  100  microliters sample  plus  50 microliters  N-methyl N-trimethyl-
     silyl  trifluoroacetamide in  conical  vial,  heat for 15 min. at 70°C, cool
     to room temperature prior to injection.
                                      -133-

-------
                                       B-5

                       QUALITY CONTROL/QUALITY ASSURANCE -
                           PENTACHLOROPHENOL ANALYSIS

A.   Recovery  pentachlorophenol,   Gas  Chromatography  Extraction  Procedures:
     Trial No.

         1
         2
         3
         4
       mg/1              mg/1             Percent
pentachlorophenol  pentachlorophenol  pentachlorophenol
	added	     extracted	      recovery	
       1.0
       1.0
       1.0
       1.0
      1.07
      0.89
      0.99
      1.00
         107
          89
          99
         100
B.  GC/FID, GC/ECD, GC /MS comparison:
                                         mg/1
                                   pentachlorophenol
     Sample No.          GC/FID         GC/ECD
     5/21/1
     5/22/4
     5/22/2
     5/22/15
     5/23/1
     5/23/6
     5/23/7
     5/23/15
     5/23/23B
     6/1/15
     5/31/19
     6/6/13
     6/7/4
     4/26/79
     5/23/24A
     5/23/27A
     5/24/6
     74
     85
     26
     71
    164
    259
     35
     31
    187
   1.35
   0.18
     48
     61
     45
     36
   76
   98
   26
   82
  128
  293
   40
   37
  140
 .1.55
 0.42
0.112
 2.88
   50
   57
   59
   31
                                  GC2/MS*
0.170
 2.60
 36.5
*Corrected for phenol - d_ recovery;
                                      -134-

-------