PROCEEDINGS OF
THE OPEN FORUM
ON MANAGEMENT OF
PETROLEUM REFINERY
WASTEWATERS
  PRESENTED BY

    THE ENVIRONMENTAL
     PROTECTION AGENCY

    THE AMERICAN PETROLEUM
     INSTITUTE

    THE NATIONAL PETROLEUM
     REFINERS ASSOCIATION

    THE UNIVERSITY OF TULSA

-------
                         PROCEEDINGS

                             OF

                         OPEN FORUM

                             ON

    MANAGEMENT OF PETROLEUM REFINERY WASTEWATERS



                          Presented by

            The U.  S. Environmental Protection Agency

                The American Petroleum Institute

            The National Petroleum  Refiners Assdeiation

                     The University of Tulsa
                       Francis S. Manning
                    Editor and  Project Director
                Professor of Chemical Engineering
              University of  Tulsa,  Tulsa, Oklahoma
                        Fred M. Pfeffer
                        Project Officer
                        Research Chemist
Robert S. Kerr Environmental Research Laboratories, Ada, Oklahoma

-------
                              ACKNOWLEDGEMENTS

       An Open Forum on Management of Petroleum  Refinery Wastewaters was held at the
Camelot  Inn, Tulsa,  Oklahoma.   This four-day (January 26-29,  1976) symposium was
sponsored by the U. S. EPA in the form of Grant No. R803957-01-0 to the University of
Tulsa.   Fred M. Pfeffer and Francis S. Manning served as  Projector Officer and Prefect
Director  respectively.   In turn the University of Tulsa contributed partial matching funds;
while the American Petroleum Institute and the National Petroleum Refiners Association
helped defray the costs of publishing the written proceedings of the Open Forum.

       The cooperation of the API,  the NPRA, and many petroleum industries in  suggesting
and engaging speakers and publicizing the Open Forum  is gratefully acknowledged.

       Success of any project is aided immeasurably by support "from the top".   Such
support was provided enthusiastically by Nick Gammelgard and Arne E. Gubrud, API; Herb
Bruck, NPRA; Bill  Galegar, EPA, Ada/Okla.; and J.  Paschal Twyman, University of Tulsa.

       Of course the speakers' contributions cannot be  overestimated.   Not only did they
present most knowledgeable and timely papers but they also reviewed the resulting discussions.

       The Project Director and Project Officer regret that it is impossible to identify all
of the numerous colleagues and friends who contributed  so much to this symposium.  How-
ever special thanks are due Katie  Whisenhunt, Shirley  Clymer,  and Cathy Whisenhunt for
managing the registration and typing the proceedings; Ed Andrews for tape recording the
proceedings; Nelda Whipple and Dawn Buss for typing the proceedings; and Tom Redmond
for redrafting many figures.
                                             11

-------
                                 CONTENTS

Acknowledgements                                                        (ii)

List of Participants                                                        (v)

Welcome                                                                 1

Keynote Address                                                          3

Session 1              Petroleum  Refining Guidelines                       17
Allen Cywin  "Establishment of Petroleum  Effluent Guidelines"               19
Martha Sager  "The Role of the Effluent Standards and Water Quality         29
             Advisory Committee"
Joe G. Moore, Jr.  "The Role of the  National Commission of Water          39
             Quality (NCWQ)
Robert T,  Denbo  "Economic Impact of Wastewater Effluent Guidelines        51
             for Petroleum Refining"

Session II              Permits for the Point-Source Category"                57
Carl J. Schafer "The Environmental Protection Agency's Permit System"      59
Robert F.  Silvus  "Current Approach to Refinery Permits"                     63
William K. Lorenz "Petroleum Industry Experience With The  NPDES Permit    69
                   System "
Harless R. Benthul  "The Enforcement  of Permits"                            77
F. T. Weiss       "Measuring the Parameters  Specified in Permits"           91

Banquet Proceedings and Address                                          105

Session III             Biological Treatment                               113
W. Wesley Eckenfelder  "Activated Sludge Treatment of Petroleum          115
                        Refinery Wastewaters an Overview"
James F. Grutsch  "A New Perspective on the Role of the Activated         127
                  Sludge Process and Ancillary Facilities"
Charles E. Ganze "Case History on Biological Treatment of  a Petroleum     153
                  Refinery Wastewater"
Mohammed A. Zeitoun  "Optimization of the Activated Sludge Process      159
                   Through Automation"
FredM. Pfeffer "The National Petroleum Refinery Wastewater Characterization
                 Study"                                                183

Session IV             Sludge Management                                197
Carl E. Adams, Jr.  "Sludge  Handling Methodology for Refinery  Sludges"    199
Jacoby A.  Scher  "Processing of Waste Oily Sludges"                      233
                                     • • •
                                     in

-------
C.  B. Kincannon  "Oily Waste Disposal by Soil Cultivation"                 257
R. L. Huddleston  "The Disposal of Oily Wastes by Land Farming"            273

Session V             Activated Carbon Treatment                          293
Davis L. Ford  "Current State of the Art of Activated Carbon Treatment"      295
C.  T. Lawson  "Cautions and Limitations on  the Applications of Activated     345
               Carbon Adsorption to Organic Chemical  Wastewaters"
Joyce A. Rizzo "Case History: Use of Powdered Activated Carbon in an      359
               Activated Sludge System"
Leon H.  Myers  "Pilot Plant Activated Carbon Treatment of Petroleum         375
                Refinery Wastewater"
M.  A. Prosche  "Activated Carbon Treatment of Combined Storm and          399
                Process Waters"

Session VI             Miscellaneous  Topics                                 411
Lial F .  Tischler  "Inherent Variability in Wastewater Treatment"             413
R. T. Milligan   "Reuse of Refinery Wastewater"                            433
Sterling  L. Burks "Biological Monitoring of  Petroleum Refinery Effluents"     445
Ronald G.  Gantz "API - Sour Water Stripper Studies"                        459

Session VII            Future Research                                      483
Wilson K.  Talley  "The EPA's Role in  Future Research"                       485
Arne E. Gubrud "The Industry's Role"                                      497
Joseph F. Molina  "The University's Role in  Future Research"                 501

Summary                                                                  509
                                          IV

-------
                            PARTICIPANTS
                            OPEN FORUM
                         January 26-29, 1976
Arthur J. Abington
Applications Engineer
Mapco
1437 South Boulder
Tulsa,  OK  74103

Ray W. Amstutz
Industry Manager
Williams  Bros. Waste Control
6600 South Yale
Tulsa,  OK   74130

Mike Anderson
Pollution Chemist
Amerada  Hess
Box 425
Purvis, MS 39475

Raymond  V. Anderson
Independent
217 Gadman Drive
Williamsville, NY 14221

Gary Bartlett
Sales Engineer
Chase  Incorporated
P.O.  Box 42546
 Tulsa, OK 74145

R.N.  Beals
Manager,  Engr. & Admin.  Services
Chemplex Company
P.O.  Box 819
Clinton,  |A 52732

Harless R. Benthul
Chief,  Legal Branch
EPA
1600 Patterson St., Suite 1100
Dallas, TX 75201
Carl E. Adams
President
AWARE, Inc.
P.O. Box 40284
Nashville, TN  37204

Jack W. Anderson
Coordinator, Ref. & Gas Plant Operations
CRA,lnc.
P.O. Box 7305
Kansas City, MO 64116

R.C. Anderson
Superintendent
CRA,|nc.
P.O. Box 311
Scottsbluff, NE  69361

W.L. Banks
EPA
1735 Baltimore Avenue
Kansas City, MO 64108

Glendon W. Bassett
Environmental Control  Engineer
Lion Oil Co.
Lion Oil Refinery
El Dorado, AR 71730

Norman L. Benecke
Manager,  Environmental  Engineer
The Lummus Company
1515 Broad Street
Bloomfield, NJ  71730

John H. Bentz
Senior Engineer
Exxon USA
P.O. Box  3950
Baytown, TX  77520

-------
Robert  E. Bergman
Sales Engineer
Johnson Div., UOP, Inc.
Box 111, LA-PA Twoer
Conroe, TX 77301

Edgar H. Bienhoff
Plant Superintendent
CRA,  Inc.
P.O. Box 608
Phillipsburg,  KS 67661

John C. Blake
Bechtel Corporation
P.O. Box 3965
San Francisco, CA  94119
Edmund  D. Blum
Coordinator-Environmental Programs
Union Oil Co. of CA
Union Oil Center, Box 7600
Los Angeles,  CA 90017

Desmond H. Bond
Process Engineer
Ford, Bacon & Davis Texas
P.O. Box 38209
Dallas,  TX 75238

Richard  Brown
Dept. Staff
Mitre Corp.
1820 Dolley Madison Blvd.
McLean, VA 22101

Robert H. Bruggnik
Director of Environmental Control
Clark Oil & Refining Corp.
Box 297
Blue Island, IL  60406

Marion Buercklin
Coordinator Environmental Affairs
Sun Oil Co.
Box 2037
Tulsa, OK 74102
Milton R. Beychok
Consulting Engineer
1 7709 Oak Tree Lane
Irvine, CA 92664
Richard J.  Bigda
President
Richard J.  Bigda and Associates
5577 South Lewis Avenue
Tulsa, OK 74105

Robert Blick
Systems Manager
Ameron Process Systems Division
1000 South Grand Avenue
Santa Ana, CA  92701

G.C. Blytas
Sr. Staff Research Chemist
Shell  Development
14323 Apple Tree
Houston, TX  77024

Clifford W. Bowers
Director, Technical Operations
Metcalf & Eddy Industrial
50 Staniford St.
Boston,  MA 02114

Herb Bruch
Technical Director
National Petroleum Refine rs Association
1725 DeSales St. NW
Washington, D.C. 20036

B .B. Buchanan
R &D
Phillips Petroleum
Batlesville, OK 74003

Sterling L. Burks
Assistant Director, Reservior Center
Oklahoma State University
Stillwater, OK  74074
                                       VI

-------
Robert D. Burns
Process Engineer
Great Canadian Oilsands Ltd.
Box 4001
Ft. McMurray, Alta, Canada T9H-E33

Janis Butler
Lab Director
Wilson & Co  .
Box 28
Salina, KS 67401

R.D. Cameron
Assist. Air & Water Conservation Coordinator
Texaco Canada Limited
90 Wynfbrd Drive]
Toronto, Ontario,  M3C 1 K5, Canada

John  Carpenter
Mobil Oil  Corporation
635 Elk Street
Buffalo,  N.Y. 14210
A.W. Catanach, Manager
Env. Control & Laboratories
Neches Butane  Products Co.
P.O. Box 817
Pt. Neches, TX 77651

Frank J. Cebula
Lead Process Engineer
Sun Oil Co.
Box 426
Marcus Hook, PA  19061

J.J. Chavez
Process Technologist
Caltex Petroleum Corp.
380 Madison Avenue
New York, N.Y.  10017

B.E. Clark
Tech. Manager
Williams Bros. Waste Control
6606 South Yale
Tulsa, OK   74136
                                      VII
Lee C. Burton
Environmental Engineer
OK State Health Dept.
NE 10th and Stonewall
OK City,  OK 73118

John Byeseda
Graduate Student
University of Tulsa
600 South  College
Tulsa,  OK  74105

J.N. Cardall
Environmental Staff Engineer
Mobil Oil Corporation
3700 West 190th Street
Torrance,  CA  90509

Rosemary Carpenter
Engineer
Ameron Process Systems Division
1000 South Grand Avenue
Santa Ana, CA 92705

Jack L. Caufield
Environmental Engineer
Toscopetro Corp.
P.O. Box 2860
Bakersfield, CA  93303

Samuel Z. Chamberlain
Prod. Eng.
Monsanto
P.O. Box 711
Alvin,  TX 77511

Cho  K. Ching
Environmental Engineer
U.S. EPA
26 Federal Plaza
New York, N.Y.  10007

Edward L.  Clark
Ind.  Sales Manager
Union Carbide
2 Greenway Plaza  E.
Houston, TX  77046

-------
William R. Clarke
Reject- Eng. Air & Waier Consv.
Texaco,  Inc.
P.O. Box 2389
Tulsa, Ok 74101

Larry T.  Clere
Environmental Engineer
Sun Oil Co.
P.O. Box 920
Toledo, OH  43693

Harold H. Coffman
Manager of Quality Control
Petrolite Corp. Bareco Division
Barnsdall, OK  74002
R.'G.  Coker
Texaco Refinery
902 W. 25,  Box 2389
Tulsa,  OK  74101
Gary Congram
Proc. Eng. Editor
Oil  & Gas Journal
P.O. Box 1260
Tulsa, OK 74127

Leon  R. Corpuz, Jr.
Mgr.  of Utilities
Pasco, Inc.
P.O. Box 277
Sinclair, WY  82334

Dennis Creamer
Analytical Chemist
Continental Oil Co.
5801  Brighton Blvd.
Commerce City,  CO 80022

Ben Crocker
Process Engineer
Williams Bros. Waste Control
 6600 South Yale
 Tulsa, OK  74136
Dr. L. Davis Clements
Assistant Professor
Texas Tech University
School of Chemical Engineering
Lubbock,  TX  79409

Michael J. Coffey
Manufacturing Representative
Dow Chemical Co.
P.O. Box BB
Free port,  TX 77541

Bruce W.  Cogswell
Process Engineer
Champlin  Petroleum Co.
Box 552 -Ref. Office
Enid, OK 73701

H .H. Comstock
Director of Environmental  Control
Phillips Petroleum Co.
2029 Firfax Road
Kansas City,  KS  66115

Jack B. Cornett
Exec. Vice President
W.R. Holway & Associates
4111 South Darlington
Tulsa, OK 74135

Allan Craig
Partner
Craig & Keithline
1129 East  15th Street
Tulsa, OK 74120

L.W. Cresswell
Conoco
Ponca City,  OK 74601
Ernest F. Cross
Manager of Engineering
Brown & Root, Inc.
 P.O. Box 3
 Houston, TX 77001
                                       VII I

-------
Joseph B.  Crowley
Technical  Coordinator
Marathon Oil Co.
539 South  Main
Findlay, OH 45840

Leonard J. Daniels
Director of Marketing
Air Resources, Inc.
P.O. Box  87
The Woodlands, TX 77373

Mike Davis
Supervisor of Environmental Control
Southwestern Refining Company
P.O. Box  9217
Corpus Christ!,  TX  78408

Wi 11 ia m Dea n
Associate Analytical Chemist
Sun Oil Co.
17th and Union Avenue
Tulsa, OK  74102

James F. Dehnert
Pilot  Plant Supervisor
Phillips Petroleum Co.
Avon Refinery
Martinez,  CA 94553

Robert T. Denbo
Environmental Control Coordinator
Exxon Co.
Baton Rouge, LA  70801
F.A. Devine
Senior  Project Engineer
Exxon Research and Engineering G>,
P.O. Box 101
Florham Park,  N.J.  70801

John H. Dobson
Technical Services Manager
Champlin  Petroleum Co.
Corpus Christie,  TX  78408
Allen Cywin (WH 552)
Director, Effluent Guidelines Div.
EPA
401 M St.,S.W. East Tower,  Rm 913
Washington, D.C. 20460

Carter B. Davis
Ch. Chemist
OKC  Refining  Co.
Box 918
Okmulgee, OK  74447

Irv Deaver
W. Div. Mgr.
Ford,  Bacon & Davis TX.
7966 E. 41st
Tulsa,  OK   74145

Wallace R. Decker
Special Projects  Coordinator
Derby Refining Co.
P.O.  Box 1030
Wichita, KS 67201

Paschal B.  Dejohn
Project Leader
1C I United  States
Route 202 & Murphy Road
Wilmington, DE  19897

Albert P. Dennis
Senior Associate Engineer
Mobil Research & Development Co.
P.O.  Box 1026
Princeton,  NJ 08540

Harold L.  Dinsmore
Process Engineer
Williams Bros.  Waste Control
6600 S. Yale
Tulsa,  OK  74136

V.O. Dodd
Texaco
P.O.  Box 2389
Tulsa,  OK 74101
                                       IX

-------
Jay R. Dungan
Project Manager
Ford^acon & Davis Texas
P.O. Box  38209
Dallas,  TX  75238

M.L. Dunton
Amoco Production Co.
P.O. Box 591
Tulsa, OK 74102
Larry Echelberger
Environmental Coordinator
Marathon Oil  Co.
P.O. Box 1191
Texas City,  TX 77590

Charles D.  Edwards
Product Manager
ARCO Chemical Co.
P.O. Box 370
Sand Springs,  OK   74063

Kenneth Eisdorfer
Project Engineer
Coalcon
1 Penn Plaza
New York,  N.Y. 10001

S. David Ellison
Manager,  Industrial Processes
CH 2 M Hill
1930 Newton Square,  Room 201
Reston, VA  22090

R.E. Etter
Environmental Coordinator

Sun Oil Co.,-Tulsa Refinery
P.O. Box  2039
Tulsa,  OK  74102

R.T. Evans
Environmental  Coordinator
Mobil Oil  Co.
Box 546
Augusta, KS  67010
R.D. Dunn
Phillips Petroleum
Bartlesville, OK  74003
John H. Dyer
Process Engineer
Union Texas Petroleum
P.O.Box 2120
Houston, TX  77001

W.Wesley Eckenfelder
Professor
Varderbilt University
Box 6222-Station B
Nashville, TN  37203

Robert Ehrlich
Environmental Coordinator
Hess Oil  Virgin  Islands Corp.
Engineering Dept.  Box 127
St. Croix, US Virgin Islands 00850

J.P. Ellis
Operations Supervisor
Standard Oil  Co. of California
324 W.  ElSegundo  Blvd.
El Segundo,  CA  90245

Richard Elton
Project Engineer
Engineering Science
3109 N.  Interregional
Austin,  TX 78740

Edward D. Evans
Chief Chemist

Skelly Oil Co.
P.O. Box 1 121
El Dorardo, KSK  67043

R.A.  Farnham
Chevron  Research
Richmond, CA   94802

-------
B.J. Flemming
Project Engineer
Texaco, Inc.
P.O. Box 2389
Tulsa,  OK 74101

Davis L. Ford
Regional V.P.
Engineering Science,  Inc.
3109 N. Interregional
Austin, TX 78722

Arthur E. Franzen
Supt. Reclamation & Conservation
Amoco  Oil Company
Whiting Refinery
Whiting,  IN  46394

Milton G. Freiberger
Process Engineer
Williams Bros. Waste Control
6600 South Yale
Tulsa,  OK  74136

Neale Fugate
Project Manager
Ford, Bacon & Davis Texas
P.O. Box 38209
Dallas, TX 75238

William C. Galegar
Director
R.S. Kerr Research Laboratories
EPA
Ada, OK  74820

Ronald G. Gantz
Supervising Process Engineer
Continental Oil Co.
P.O. Box 1267
Ponca City, OK 74601

N.E. Garland
Environmental Engineer
Ethyl Corp.
P.O. Box 341
Baton Rouge,  LA  70821
Peter J.  Foley
Environmental Engineer
Mobil Oil Co.
150 East  42nd Street
New York , N.Y. 10017

Orval L. Fouse
Supervisor, Utilities & Environmental Eng.
Gulf Oil Co.—U.S.
P.O. Box 701
Port Arthur, TX 77640

Allden R. Frederickson
Supervisor Environmental Control
Murphy Oil Corp.   Meraux Reg.
P.O. Box 100
Meraux,  LA 70075

George Frondorf
Engr. Suprv.
Amerada  Hess Corp.
Purvis Refinery, P.O. Box 425
Purvis, MS 39475

R.E. Funk
Environmental Coordinator
Cities Service Co.
P.O. Box 300
Tulsa, OK 74102

P. N. Gammelgard
Senior Vice President, Environmental Affairs
American Petroleum Institute
2101 L St., N.W.
Washington,  D.C.  20037

Charles Ganze
Facilities Supervisor
Gulf Coast Waste Disposal Authority
910 Bay Area Blvd.
Houston,  TX  77058

John E. Garner
Manager, Environmental Affairs
TOTAL Leonard, Inc.
P.O. Box 231
Alma, Ml 48801
                                        XI

-------
 Refer C. Gaskin
 Supervisor,  Environmental & Quality Control
 Kerr-McGee Chemical Corp.
 P.O. Box 25861
 OK  City, OK  73125

 William  G. Geubelle
 Analytical  Chemist
 Conoco
 P.O. Box 1267
 Ponca City, OK  74601

 Dan  H . Gillum
 Environmental Engineer
 Phillips Petroleum
 Box  271
 Borger,  TX  79007

 E.F. Gloyna
 Dean of Engineering
 University of Texas
Austin,  TX  78712

 James W. Godlove
 Environmental Engineer
 Phillips Petroleum
 10 Bl Phillips Bldg.
 Bartlesville, OK  74004

 H,M. Gomaa
 Environmental Engineer
 Lummus Co. Canada
 255  Consumers Road
 Willowdale, Ontario, Canada

 Fred  W. Gowdy
Senior Staff Engineer
 Exxon
 P.O. Box 551
Baton Rouge, LA   70821

Robert W. Griffin
Manager, Special Projects
NUS Corp.
1910 Cochran Road
Pittsburgh,  PA 15220
Rene Gaudet
Environmental Engineer
Tenneco Oil
P.O. Box 1007
Chalmette, LA  70043

G .H. Gaumer
Refinery  Representative
Phillips Petroleum
Avon Refinery
Martinex, CA 94553

John A. Ghser
Director Lab and Environmental Affairs
Gulf Oil Co. US
Box 2487
Santa Fe Springs, CA 90670

H.W. Goard
Phillips Petroleum
Barflesville, OK   74003
Robert Goldsworthy
Project Engineer
Standard Oil Co. of CA
324 W. El Segundo  Blvd.
El Segundo, CA 90505

W,.J. Gossom
Vice  President
Resource Sciences Corp.
6600  South Yale
Tulsa, OK   74136

John  H. Gray
Environmental Engineer
Ashland Petroleum Co.
1409  Winchester Ave.
Ashland, KY 41101

Jim Grutsch
Coordinator of Environmental Projects
Standard Oil (Indiana)
200 East Randolph Drive
Chicago,  IL  60601
                                       XII

-------
Arne E. Gubrud
Director, Environmental Affairs
American Petroleum Institute
2101 L St.,  N.W.
Washington, D. C.   20037

John D. Hallett
Senior Engineer
Shell  Oil Co.
P.O. Box 2463
Houston, TX 77001

Mike Henebry
Mkt.  Mgr. of  Petroleum  & Chemicals
Environmental  Research and Technology
Concord, MA  01742
Thomas Hickey
Process Design Engineer
CE Lummus
1515 Broad Street
Bbomfield,  NJ 07003

Guy Hillard
Mgr. of Quality Control
Witco  Chemical  Corp.
77 N.  Kendall Avenue
Bradford,  PA  16701

Larry C.  Holland
Environmental Coordinator
Cities  Service Co.
P.O. Box 300
Tulsa,  OK 74102

W.F. Hoost
Process Manager
Pullman - Kellogg
Three Greenway Plaza  East
Houston, TX  77046

R.L. Huddleston
R &D
Continental Oil  Co.
Ponca  City,  OK 74601
Jack A. Guthrie
Assistant Refinery Manager
Sun Oil Co.
Box 426
Marcus Hook,  PA 19061

Pat Havener
Mechanical Engineer
Navajo  Refining Co.
P.O. Box  159
Artesia, NM  88210

S.F. Heneghan
Engineer
Amoco Oil Co.
P.O. Box  182
Wood River, IL 62095

L.E. Hilgers
Lead Process Engineer
Sun Oil Co.—Tulsa  Refinery
P.O. Box  2039
Tulsa, OK 74102

C.J. (Pete) Hoffman
Manager,  Environmental Services
Champlin  Petroleum  Co.
P.O. Box  9365
Fort Worth, TX 76107

Harold Holt
Process Engineer
Great Canadian Oil Sands Ltd.
Box 4001
Ft. Me Murray, Alta ., T9H33E3T Canada

Jan F. Horwath
Environmental  Coordinator
Bukeye Pipe Line Co.
P.O. Box  368
Emmanus,  PA  18049

Darrell W. Hughes
Plant Supt.
OKC Refining  Inc.
P.O. Box 918
Okmulgee, OK 74447
                                      XIII

-------
 E.E.  "Gene" Humes
 Design Engineer
 Crest Engineering
 P.O. Box 1859
 Tulsa, OK 74101

 M.K. (Don) Hut-ton
 Mgr. , Mec & Environ.  Eng.
 Kerr-McGee Refining Co.
 P.O. Box 25861
 Oklahoma City, OK   73125

 Raymond L.  Jackson
 General  Manager
 Environanental Design Eng.  Co.
 Box 1026
 Findlay,  OH 45840

 John  O .  Johnson
 Utilities  Superintendent
 Sun Oil  Company
 Tulsa, Refinery, P.O. Box 2389
 Tulsa, OK 74102

 Ted Johnson
 Staff Assistant
 Hunt Oil Co.
 P.O. Box 1850
 Tuscaloosa, AL   35401

 Robert S. Jones
 Sr.  Environmental Technologist
 Continental Oil  Co.
 P.O. Box 1267
 Ponca City,  OK 74601

 Jack A.  Kamps
 Mgr.  of Engineering
 Toscopetro Corp.
 P.O. Box 2860
 Bakersfield,  CA  93303

 S.M. Khan
 Mgr.  Refinery and Utility Sales
The Ducon Co.
 147 E Second Street
 Minneola, NY  11501
                                     XIV
James (Jim) P. Hutchinson
Regional Administrator
Champlin Petroleum Co.
P.O. Box 9365
Fort Worth,  TX 76107

George  F. Jackson
Crest Engineering, Inc.
P.O. Box 1859
Tulsa, OK 74135
Jere M. Jo hnson
Supervisor, Environmental  Engineering
Exxon Co., U.S.A.
P.O. Box 3950
Bay town,  TX  77520

K .R. Johnson
Process Engineer
Sun Oil-Tulsa  Refinery
P.O. Box 2039
Tulsa, OK 74102

L.W. Jones
Research Asso .
Amoco Production Co.
Box 591
Tulsa, OK 74102

John R.  Kampfhenkel
Lead Refinery Engineer
Suntide Refining Company
P.O. Box 2608
Corpus Christ!,  TX 78403

W.G. Kelly
Manager of Environmental  Operations
Atlantic Richfield  Co.
Box 2679 --T.Z.
Los Angeles, CA 90051

R. Kilpert
Senior Engineering Associate
Exxon Research and Engineering Co.
P.O. Box 101
hlorham Park,  N.J.  07932

-------
C.B. Kincannon
Texas Wafer Qualify Board
P.O. Box 1.3246, Capitol Station
Austin,  TX  78711
Dr. A..T. Knecht
Supervisor
Atlantic Richfield Co.
400 E. Sibley Bldg.
Harvey,  IL  60426

Charles E.  Knipping
General  Foreman,  Wastewater Comtrol
Clark Oil & Refining Corp.
P.O. Box 7
Hartford,  IL 62048

R.D.  Kuerston
Chem. Eng.
Phillips Petro. Co.
1523  Hampden Rd.
Bartlesville, OK  74003

Gerald  Lamb
Process  Manager, Dravo Corp.
1777 Borel  Place
San Mateo, CA  94404

Juanifa G. Lairmore
Environmental Control Technologist
Standard Oil of California
P.O. Box 1300
Pascagoula, MS  39567

Cyron T. Lawson
Project Scientist
Unbn Carbide
Tech Center
So. Charleston, WV   25303

Eddie Lee
Public Information  Officer
U.S.  EPA
Rt. 5,  Box  411
Ada,  OK 74320
Conrad W. Kleinholz
Zoology Graduate Student
Oklahoma State University
Life Science West 03
Stillwater, OK 74074

Eva Knettig
Chem. Engr.
Imperial Oil Enterprise,  Ltd.
Samia, Canada
Ray Knutson
Environmental  Eng.
Koch  Refining  Co.
Box 3596
St. Paul , MN  55165

Walter C. Lackemann
Chief Chemist
The Refinery Corp.
5800 Brighton Blvd.
Commerce City, CO  80022

Neale V.  Lamb
Engineer
Phillips Petroleum Co.
Bartlesville, OK, 74003

Les Lash
Marketing Specialis!1
Envirofech
Eox 300
Salt Lake, City, UT 84110

Peter  Bu Lederman
Director R & D
EPA
401 M St., S.W.
Washington, D.C.

Norman Lehnhardt
Tesoso Petroleum
P.O. Box 17536
San Antonio, TX 78286
                                        xv

-------
 Carl Lev!
 Pac. Div.  Mech. Eng.
 Champlin Petroleum
 P.O. Box  125
 Wilmington,  CA  90748

 R.I. Loeffler
 Chief Chemist
 Rock Island Refining  Corp.
 500 W. 86th  St.
 Indianapolis, IN   46268

 WM. K. Lorenz
 Sr. Eng. Engineer
 Sun Oil  Co.
 Box 1135
 Marcus Hook ,  PA 19061

 Lester  M. McCright
 Project Manager
 Williams Bros. Waste  Control
 6600 South Yale
 Tulsa,  OK 74136

 Wayne McLaury
 Senior Staff Engineer
 Crest Engineering, Inc.
 P.O. Box 1859
 Tulsa,  OK 74101

 Joe F- Molina
 Cicil Eng.  Dept.
 University of Texas
 ECJ 3.6
Austin, TX 78712

 Kris D. Manchanda
 Mkt. Res.  Supervisor
 Union Carbide Corp.
Old Saw Mill River Road
 Tarryfown, NY 10591

 Francis S. Manning
 Professor/ ChE
 University of  Tulsa
600 S.  College
 Tulsa, OK  74104
Henry E. Lloyd, Jr.
Technology Services  Engineer
Getty Oil Co.  (Eastern  Operations) Inc,,
Delaware  Refinery
Delaware , DE  19706

Douglas G.  Lofgren
Environmental Engineer
Phillips Petro.
10 Bl  Phillips Bldg.
Bartlesville,  OK  74003

W.C. McCarthy
Phillips Petro. Go.
Bartlesville, OK  74003
Michael D.  McGee
Process Engineer
Continental  Oil  Co.
21 0 So.'  9 th
Ponca City,  OK  74601

Patrick McMahon
Process  Engineer
Brown & Root
Box 3
Houston, TX 77001

Russ  C.  Mallatt
Standard Oil
200 E. Randolph Drive
Chicago,  IL  60601
Jim Mann
Burmah Oil & Gas
909 S. Meridan Suite 603
Ok City, OK 73128
Ralph E.  Maple
Manager Refinery Services
Merichem Company
P.O. Box 61529
Houston, TX  77208
                                      XVI

-------
Richard Marcum
Sanitary Engineer
Alaska Department of Environ. Conserve,
Pbuch O
Juneau, AK

Charles D. Martin
Senior Engineer
Crest Engineering
4310 S. Lakewood
Tulsa, OK  74135

F.J. Martin
Supt.
Union Oil of California
Box 237
Nederland,  TX 77627

Anil Mathur
Student
2602 E. 10th Street,  Apt.  2
Tulsa,  OK 74104
Umesh Mathur
Environmental & 208 Project Engineer
Indian Nations Council of Governments
630 W. 7th Street
Tulsa, OK 74104

Krishna Merchant
Tertiarary Management
University of Tulsa, ChE Dept.
600 S. College
Tulsa, OK 74104

WM.  R. Miles
Environmental Engineer
Charter Inter Oil Co.
Box 5008
Houston, TX  77012

Otis Miller
Process Engineer
CRA,  Inc.
P.O.  Box 570
Coffeyville, KS    67337
James A. Maricle
Air & Water Conservation Coordinator
Mobil Oil Corp.
P.O. Box 8
Ferndale, WA 98248

D.P. Martin
Director, Environmental Affairs
Gulf Oil Company, U.S.A.
P.O. Box 2100
Houston, TX  77001

J.L. Marzak
Vice President—General  Manager
Metcalf & Eddy Industrial Div.
50 Staniford Street
Boston,  MA  02114

Jimmy Mathur
Process  Engineer
Brown & Root
4100 Clinton  Drive
Houston, TX  77001

R.V. Mattern
Supt. Environment Cons.
Shell Oil/Chem.
P.O. Box 2633
Deer Park,  TX 77586

John Migiliavacca
Process  Engineer
Brown & Root
P.O. Box 3
Houston, TX  77001

Gilbert  Millar
Manager-Ufilities
Brown & Root
4100 Clinton  Drive
Houston, TX 77001

Richard  B. Miller
Supt.,  Chemical Operations
Skelly Oil Company
P.O. Box 1121
El Dorado,  KS 67042
                                       XVII

-------
 Robert Milligan
 Pollution Cbntrol Technologist,  Mgr.
 Bechtel Corp.
 P.O. Box 3965
 San Francisco, CA 94119

 Joe. G .  Moore, Jr.
 University of Texas,  Dallas
 P.O. Box 688
 Richardson, Texas  75080
 Gary W, Munson
 Engineer
 LaGloria Oil & Gas Co.
 P.O. Box 840
 Tyler, TX  75701

 R.E.  Myers
 General Marketing  Manager
 Nalco Chemical Co.
 2901  Butterfield  Road
 Oak Brook, IL  60521

 Alee  J. Nash
 Environmental  Engineer
 Conoco
 P.O. Box 37
 Westlake, LA  70669

 W.H.  Nichols
 Chief Project Engineer
 Champlin  Petroleum Co.
 Box 552
 Enid,  OK   73701

 William R. Oberman
 District Manager
 Envirotech  Corp.
9235 Katy Road, Suite 202
 Houston, TX 77024

W.P.  Olivent
Senior Mechanical Engineer
GulfOil Co. U.S.A.
 P.O.  Box  Drawer G
V,
'enice, LA  70091
                                      XVII I
                                          G .T. Minnick
                                          Senior Environment Engineer
                                          Sun Oil Co.
                                          P.  O. Box 2039
                                          Tulsa OK  74102

                                          Ronald F. Morgan
                                          Advanced Scientist
                                          Marathon Oil Co. -Denver Research  Center
                                          P.O. Box 269 (7400 S.  Broadway)
                                           Littleton,  CO  80120

                                          Leon  H.  Myers
                                          Research Chemist
                                          R.S.  Kerr Environmental Research Labs, EPA
                                          Ada,  OK 74820
Ted Nairn, Jr.
Coordinator Air & Water Conservation
Cosden Oil &  Chemical
Box 1311
Big Spring, TX 79720

John C. Nemeth
Manager-Environmental  Sciences Dept,
Law Engineering Testing Co.
2749 Delk Road, SE
Marietta,  GA  30062

Howard  W. Nickle
Operating Technical Specialist
Sun Oil Co.
Box 426
Marcus Hook , PA 19061

G .F.  Oldman
Effluents Adviser
British Petroleum
Britannic House, Moor Lane
London EC2, England

Ken Orr
Project  Manager
Great Norfhern Electric Co.
2016 W. 43rd  Street
Kansas City, KS  66103

-------
Leonard Orwin
Engineer
Metcalf & Eddy  Engineering
Box 10046
Palo Alto, CA 94303

David  Parrish
Research Assistant
Oklahoma State University
LSW 402
Stillwater, OK 74074

Howard Patterson
Chemist
Mapco Inc.
11391  E.  Tecumseh
Tulsa,  OK   74116

Fred Pfeffer
Research Chemist
R.S. Kerr Environmental  Research Lab
EPA —P.O. Box 1198
Ada, OK 74820

B. Vail Prather
Staff Specialist
WM. Bros. Waste Control
6600 South Yale
Tulsa,  OK 74136

Jean B. Reeder
Technical Editor
L.R. Reeder & Associates
5200 South Yale
Tulsa,  OK 74135

George Reid
Professor
University of Oklahoma
202  West Boyd Street, Room 301
Norman, OK  73069

A. Kim Reyburn
Manager-Pollution Control
Crest Engineering, Inc.
P.O. Box 1859
Tulsa,  OK 74101
                                       XIX
R.E. Palmer
Senior Process Engineer
The Litwin Corp.
P.O. Box 282
Wichita, KS  67201

Don  Patterson
Chief Chemist
Apco Oil  Corp.
 P.O. Box 857
Arkansas City, KS 67005

Wes  Perkins
 Coordinator Environmental Affairs
ARCO
Box 1346
Houston, TX  77001

M.L. Fbrter
Project Administrator
Phillips Petro. Co.
14 D 1 Phillips Bldg.
Bartlesville, OK   74004

Marvin A.  Prosche, Mgr.
Refinery Technology
Atlantic Richfield
1801 E. Sepulveda BUM.
Carson, CA 90745

Pedro Regnault
University of Tulsa
819 S. Harvard
Tulsa, OK 74112
John C. Reidel
Sr. Process Engineer
Williams Bros. Process Services
6600 South Yale
Tulsa, OK 74136

William R. Reynolds
Mechanical Engineer
Navajo  Refining Co.
P.O. Box 159
Artesia, NM  88210

-------
Mrs. Myria M.  Rivera
Assoicate  Chemist
Yabucoa Sun Oil
Box 186
Yabucoa,  Puerto Rico 00767

Gary Roehr
Chemist
Indiana Farm Bureau  Coop, Ass'n Inc.
1200 Refinery Road
Mt. Vernon, IN 47620

G.A. Rohlich
Civil  Eng. Dept.
ECJ8.6.
University of Texas
Austin,  TX  78712

Jim E. Russell
President
Aquaculture Industries, I nc.
2116 West 24 th Avenue
Still water, OK 74074

William D. Rutz
Mgr.,  Lab Sciences & Chemicals
Mapco
1437 South Boulder
Tulsa,  OK 74114

Martha Sager
EPA
(WH 551)  Crystal Mall, Bldg. 2
Arlington, VA  20460
J.K .  Sargent
Manager Marketing
Dravo Corp.
One Oliver Plaza
Pittsburgh, PA  15222

H.F.  Scarsdale
Environmental  '.Coordinator
Skelly Oil  Co.
P.O.  Box 1650
Tulsa, OK  74102
                                        xx
Joyce A. Rizzo
Sun Oil Company
Box 1135
Marcus Hook,  PA 19061
Arthur Rogers
Mobil Oil Corp.
635 Elk Street
Buffalo,  N.Y. 14210
W.L. Ruggles
Project Administrator
Phillips Petroleum
14 D 1 Phillips Bldg.
Bartlesville, OK  74004

Simon Russo
Student
University of Tulsa
801 North Gary Place
Tulsa,  OK  74110

Edward R. Sager
Project Manager
Williams Bros. Waste Control
6600 South Yale
Tulsa,  OK  74136

Max Samfield
Chemical Engineering
EPA
Room N-123, Mail  Drop 62
Research Triangle Park,  NC 27711

James R. Satcher
Assistant Manager Refineries
Southland Oil Co.
P.O. Box 128
Sandersville, MS  39477

Carl Schaffer (En  336)
Acting Director, Permits Division
EPA,  Crystal Mall  Bldg.  2
1921 Jefferson Davis Highway
Arlington, VA  20460

-------
Jacoby A. Scher
Environmental  Engineer
Fluor  Engineering and Construction
4620 North Brae wood
Houston, TX

K.F.  Schmidt
Amoco Production Co.
Box 591
Tulsa,  OK  74102
Edward C. Sebesta
Project Engineer
Brown & Root, Inc.
P.O. Box 3
Houston, TX  77001

George Sexton
Gooridnator-Environmental Controls
Beta Laboratories, Inc.
Somerton Road
Trevose,  PA 19047

Robert Silvus
Chief, Industrial Branch
Texas Water Quality Board
P.O. Box 13246, Capitol Station 78711
Ausfin,  TX 78711

John Single/
WBWC
1433 W.  Loop S #634
Houston, TX  77027
Dave Skamenca
Sales Engineer
Eimco BSP-Suite 202
9234 Katy Freeway
Houston, TX  77024

John H. Smith, III
Sales Engineer
Jacobs
837 South Fair Oaks Avenue
Pasadena, CA 91105
Marc Schillinger
Environmental  Engineer
Atlantic Richfield Co.
400 E.  Sibley Blvd.
Harvey, IL 60426

J.W.  Scrivner
Environmental  Engineer
Amoco Oil Co.
P.O. Box 8507
Sugar Creek, MO  64054

Robert  P.  Selm
Wilson & Co.
Box 28
Salina, KS 67401
E.M. Sheets
Laboratory Section Supervisor
Sun Oil Co.
Box 2039
Tulsa,  OK  74102

Gary W. Simms
Chemist
Marion Corp.
Rangeline Road
Mobile, AL

Suzette Sivyer
Environmental   Technician
EPA
100 California St.
San Francisco, CA 94111

John Skinner
Chief Chemist
Kerr-McGee Refining Corp.
Wynnewood, OK  73098
Phil J. Smith
Process Engineer
Ford,  Bacon & Davis Texas
P.O.  Box 38209
Dallas,  TX 75238
                                       XXI

-------
 S.R. Spears
 Texaco Refinery
 P.O.  Box 2389
 Tulsa, OK  74101
 David O. Story
 Director of  Process Engineering
 Gulf Oil Corp.
 2935 Front Street
 Toledo,  OH 43697

 E.J. Sullivan
 Environmental  Specialist
 Amoco Oil  Co.
 P.O. Box 182
 Wood River, IL 62095

 W.L. Swander
 Manager, Process  Engineering
 Sun Oil  - Tulsa Refinery
 P.O. Box 2039
 Tulsa, OK  74102

 Kenneth J.  Tacchi
 Chemist
 Tretolite
 369 Marshall Avenue
 St. Louis, MO   63119

 J. Anthony  Tall is
 Environmental Systems  Consultant
 NUS Corp.
 71 1  Louisiana Street
 Houston, TX 77022

 Richard Thorstenberg
 Engineering  Assistant
 Continental  Oil Co.
 P.O. &ox 1267
 Ponca City,  OK  74601

Jul io Ortiz  Torres
Director-Technical Planning
Caribbean Gulf Refining Corp.
G.P.O.  Box 1988
Sun Juan , Puerto Rico  00936
Larry W. Stinnett
Senior Process Engineer
Kerr- McGee Corp.
Box 305
Wynnewood, OK  73098

Ronald D.  Stover
Assistant Refinery  Superintendent
Indiana Farm Bureau Coop Ass'n.,  Inc.
1200 Ref inery Road
Mt.  Vernon, IN  47620

John C.  Suplicki
Application Engineer
Ecodyne  -  IWTD
2720 U.S.  22
Union, NJ  07083

Nicholas D.  Sylvester
Professor of Chemical Engineering
University of Tulsa
600 South College
Tulsa,  OK  74104

Wilson K.  Talley
Assistant Administrator - R & D
EPA
40 M St.  S,W.
Washington,  D.C. 20460

Gerald Thornhill
EPA
1600 Patterson Street
Suite  1 100
Dallas, TX  75201

Lial F. Tischler
Manager, Austin Office
Engineering-Science, Inc.
3109 N.  Interregional
Austin, TX 78722

John W.  To we
Engineer
Sun Oil Co.
1600 Walnut
Philadelphia, PA 19103
                                      XXII

-------
Paul C. Tranquil I
Environemental Engineer
The Standard Oil Co. (Ohio)
HSOSouthMetcalf St.
Lima, OH  45804

T.J. Tseng
Environmental Engineer
Ralph M. Parsons Co.
Pasadena, CS 91124
Michael  Ray Twitchell
Process Engineer
Charter Internation Oil Co.
Box 5008
Houston, TX  77012

Herbert Uhlig
Manager Petroleum Technology
Stone &  Webster Engineering Corp.
1 Pen Plaza
New York, N.Y.  10001

R.E.Van Ingen
Manager, Mfg. Environmental Conservation
Shell Oil Co.
P.O. Box  2463
Houston, TX  77024

W.A.  Wadsack
Supervisor, Environmental  Control
Cities  Service Oil Go.
P.O. box 1562
Lake Charles,  LA 70601

Joe D. Walk
Project Director
Standard  Oil (Indiana)
200 East Randolph Dr.
Chicago,  IL 60601

Gordon J. Wanless
Senior  Technologist
BP North America, Inc.
620 Fifth Avenue
New York, N.Y.  10020
                                    xxiii
D.V. Trew
Environmental Coordinator
Cities Service Co.
P.O. Box 300
Tulsa,  OK 74102

W.N.  Turtle
Process Engineer
Williams Bros. Waste  Control
6600 South Yale
Tulsa, OK 74136

J. Paschal Twyman
President
University of Tulsa
600 South College
Tulsa,  OK 74104

Russ Vandenberg
Manager- Industrial Sales
Resources  Conservation  Co.
P.O. Box 936
Renton, WA 98124

Gary R. Van Stone
Adsorption Systems Specialist
Calgon Corp.
4800 West 34th Street
Houston, TX 77043

J.N. Wakefield
Process Engineer
The Litwin Corp.
P.O. Box 282
Wichita, KS 67201

Henry F. Walter
US ERDA
Mail Stop-D-227
Washington, D.C.  20545
David Ward
Manager Engineering and Adminstration
Chemplex Co.
P.O. Box 819
Clinton,  IA  53732

-------
 William L. Warnement
 Manager - Environmental Engineering
 Crown Central Petroleum Corp.
 P.O. Box 1759
 Houston, TX 77001

 L.  Duvall Webster, Jr.
 Professional Engineer
 Turner, Mason,  & Solomon
 1950 Mercantile Dallas Bldg .
 Dallas, TX  75201

 Fred Weiss
 Shell Development
 Box 2099
 Houston, TX 77001
 R.B.  Wells
 Sales Supervisor
 Amoco Chemicals
 3815  Dacoma
 Houston, TX 77018

 Morris Wiley
 Coordinator-Environmental  Protection Dept.
 Texaco,  Inc.
 P.O. Box 509
 Beacon,  N.Y.  12509

 Edward J. Wolf
 Senior Process Engineer
 Gulf Oil Co. ~ U.S.
 13539 E.  foster Road
 Santa Fe Springs,  CA  90670

 Marvin L. Wood
 Deputy Director
 R.S. Kerr Environmental Research Lab
 Box 1198
Ada,  OK  74820

 Thomas Zale
 Process Engineer
 Sun Oil Co.
 1608 Walnut Street
 Philadelphia, PA 19103
James  E. Watkins, Jr.
Utilities Chemist
Sun Oil Co.
Tulsa Refinery, 17 th  & Union
Tulsa,  OK  74102

Irving  Wei
Senior  Technical Consultant
Process Res. Div.
Environmental Research & Technology
Concord, MA 01742

R.A. Weldon
Research Engineer
Sun OilCo.
Box 1135
Marcus Hook  . PA   19061

John White
Marketing Manager
Mapco
1437 South Boulder
Tulsa,  OK 74103

Jim Wilkerson
Tank Farm Foreman
Marion Corp.
P.O. Box 526
Theodore, AL 36582

Joe F.  Wood
Technical Advisor
Skelly  Oil Co.
Box 1 650
Tulsa,  OK  74102

Robert  L. Wortman
Acting  Director, Water Quality Service
OK State Dept. of Health
NE 10th &  Stonewall
OK City, OK

Eugenio Zavala
Student
University of Tulsa
2423 East 5th Place, Apt. 33
Tulsa,  OK  74104
                                      XXIV

-------
M.A. Zeitoun
Senior Research Specialist
Dow Chemical
Freeport, TX 77541
                                       XXV

-------
XXVI

-------
                            WELCOME ADDRESS
                      William C. Galegar,  Director
              Robert S. Kerr Environmental Research Laboratory
                         E.P.A., Ada, Oklahoma
        Honored Guests, Ladies, and Gentlemen,  It is a great pleasure  to welcome
you to the first 'Open Forum on Management of Petroleum Refinery Wastewaters. "
My hope  is that this will be only the first of a continuing series of technical
information exchanges between  the industry, academia, governmental agencies at
all levels, and the interested public.  This conference, as you will note,  is a
series of sessions dealing with problems and solutions to protect the environment.
In each session the program representatives have attempted to offer a balanced
presentation of view points.  The list of speakers is most impressive, and they are
all quite knowledgeable in their fields.
        Success or failure of any conference, however, can be linked to two
action words-"participation" and "communication" by attendees.  While not being
an active part of any of the committees responsible for the symposium,  I was
close enough to experience the  enthusiasm of your respresentatives from the
American Petroleum Institute, the National Petroleum Refiners Association, the
University of Tulsa, and the Environmental Protection Agency.
        Last  evening, upon entering the hotel,  the undercurrent of expectation  was
very evident.  Meeting old and new friends convinced me that you came to
"participate" and "communicate".  You have already, by your past and present actions,
indicated the true welcome of this meeting; I have the priviledge of expressing that
feeling.  On behalf of the organization sponsoring this symposium, the working
committees  who planned it, the speakers for the various sessions,  and our host,
Tulsa University,  I bid you welcome.  May the next three days be as helpful to
you as I expect them  to be to me.
BIOGRAPHY

        William C. Galegar holds a  BS in Chemical
Engineering from Oklahoma State University and an
MS in Chemical Engineering  from the University of
Oklahoma.  His career has included 1 1 years with
Oklahoma Department of Health and 16 years
with the Environmental Protection Agency and its
predecessor agencies.  Mr. Galegar is currently
Director of the EPA's Robert S. Kerr Environmental
Research Laboratory in Ada,  Oklahoma •

-------

-------
                                 KEYNOTE ADDRESS

 "WASTEWATER CONTROL — WHERE ARE WE, WHERE ARE WE GOING, AND WHY?

                                 Milton R. Beychok
                         Consulting Engineer,  Irvine, California

     I am sure that all of us must be impressed by the broad and diverse range of view-
points represented by the speakers assembled for this conference.  Very briefly,  we have:
     — The EPA Divisional Directors for Effluent Guidelines and for Permits
     — The Chairman of the EPA's Effluent Standards and Water Quality Information
       Advisory Committee
     — The Program  Director of the National Commission on Water Quality
     — A speaker from the Texas Water Quality Board
     — The EPA's Assistant Administrator for Research and Development
     — Speakers from a number of the major oil  refining companies
     — Speakers from several well-known engineering and consulting firms
     — Eminent university professors
     — Key administrators from the  EPA's Robert S.  Kerr Research Laboratory
Aside from being impressive,  these diverse viewpoints give us a  unique opportunity to
share perspectives and to explore important policy and technological  questions related to
industrial wastewater control.  Toward that end, I would  like to explore a few key issues
in this 'keynote1 address	and one of those is the problem of the communication of
viewpoints between industry and the federal  EPA.

EFFECTIVE COMMUNICATION REQUIRES A TWO-WAY STREET

     Effective communication works best on a two-way basis.  That may be a cliche, but
it is nevertheless true.  Mr. Joe Moore of the NCWQ recently stated that (1), "We  must
not only  listen, but we must also hear what is said." And I would add that we can accom-
plish much more by listening to each other than by shouting at each other.

     In the three years since enactment of PL 92-500 (the Federal Water  Pollution Control
Act Amendments of 1972), at least 634 lawsuits have been instituted  involving implemen-
tation of the  Act (2).  I submit that there would be far less court litigation between  in-
dustry and the EPA if industry could effectively participate  in developing effluent and
emission  control regulations	and if industry  viewpoints were truly given serious con-
sideration.  I fully realize that some industry viewpoints are non-productive and perhaps
overly opposed to meaningful regulations. But the large majority of industry is willing to
accept equitable regulations based upon realistic data and commercially-proven techno-
logy. The EPA should solicit industry viewpoints very early in their process of developing
regulations, and provide industry a  real opportunity to participate in their decision-
making.   It is not enough to merely invite formal written comments within a legalistic
30-60 days after publication of a criteria document or proposed  promulgations	espe-
cially not when those criteria  or proposed promulgations represent months or years of work
and many irreversible decision steps already taken by the EPA and their outside contrac-
tors.  What industry wants is early involvement, face-to-face dialogues and a meaningful
opportunity to help formulate proposed regulations through all the stages of decision-
making.   Formal comments made after the fact,  or during public hearings conducted under

-------
the rigid rules of law,  are poor substitutes for a meeting of minds around a conference table.
Formal comments and rigid public hearings simply invite adversary  debate merely to esta-
blish a legal record	what we need  instead is the chance to voice, to hear and discuss
valid viewpoints.  I  submit again that a working partnership between industry and the EPA
would drastically reduce the court litigation involving PL  92-500.   Obviously, the same
participatory partnership between the EPA and the State regulatory agencies would be
equally helpful.

     It is probably too  late to institute  any working partnership of industry and the EPA
involving the 1977 and 1983 effluent limitations mandated by PL 92-500.  But it is not too
late to learn some lessons from the past three years	and apply them to the tasks and
decisions that lie ahead of us.  For example,  what of PL 92-500's objective regarding the
elimination of the discharge of pollutants (EOD) by 1985	the so-called 'zero discharge1
goal?  And what of the yet to be issued toxic pollutant standards called for in Section 307
of PL 92-500?  And  what of the recent decisions (3) by the Third and Seventh  Circuit
Courts of Appeal	which made a clear-cut distinction between Section 301  effluent
limitations as being 'minimum' levels of control and Section 304 guidelines as  being "factors
or ranges' to be used by the permit issuers in deciding whether or by how much any indivi-
dual permit is to be more stringent than the 'minimum' level Section 301 limitations? Do
these Circuit Court decisions  lend more validity to the matrix approach for  Section 304
guidelines  as advocated by the EPA's ES&WQIAC committee?  Certainly,  these are ques-
tions which must be of urgent concern to all of industry and there is still an opportunity
to tackle these questions by a broad-based working partnership between  industry, the EPA,
and the State regulatory agencies.

WHERE ARE WE NOW?

     I think most of us would agree that the tasks to be accomplished in order to achieve
the mandated 1977 and 1983 goals of PL 92-500 are now defined in terms of the EPA's 1977
and 1983 Effluent Limitations and their New Source Performance Standards.  Obviously,
resolution of the recent Circuit Court decisions and of the pending  backlog of  litigation
may very probably add new dimensions to the  1977 and 1983 regulations.  But  at least we
have a framework upon which to assess where  we are now, what we may accomplish by 1983,
and what it may cost.

    We are indeed fortunate at this  conference that the National Commission  on  Water
Quality has very recently released its Staff Draft Report.  That Commission was created by
Section 315 of  PL 92-500 and charged  with making a complete study of:

               "...  all aspects of the  total economic, social and environmental effects of
               achieving or not achieving the goals set forth for 1983. "

The NCWQ staff draft  summarizes PL 92-500 in terms of the Act's primary objective, two
key goals and a number of specific policy statements.  The primary objective of the Act is:

               "to restore and maintain the chemical, physical and biological Integrity of
               the Nation's waters."

-------
The two key  goals of the Act are given by the NCWQ as:  (4)
    — To reach, "wherever attainable," a water quality that "provides for the protection
       and propagation of fish, shellfish, and wildlife" and  "for recreation in  or on the
       water" by July 1983.
    — To eliminate the discharge of pollutants (EOD) into navigable waters by 1985.
Focusing on the  1983 goal,  the NCWQ staff draft states that "much of the nation's waters
are already of sufficient quality to meet the interim (1983) goal" as the result of pollution
control efforts prior to PL 92-500 or  as a natural condition (5).  In order to be more de-
finitive,  the Commission selected certain  minimum water quality criteria that would con-
stitute meeting the  1983 goal	one of which was a dissolved oxygen level of at least
4 ppm during 'worst case1 conditions of seasonal  low flow  (5).  Based on an assessment of
41 sites,  the NCWQ developed the data in Table 1.  These data indicate that 7-8% of
the nation's waterways do not presently meet the  'worst case1 dissolved oxygen  criteria
for satisfying the 1983 goal. Thus, in  effect, we might define 'where we are now' by
saying that 92-93% of our waters already satisfy a key criteria for meeting  the  1983 goal
of PL  92-500.

WHERE ARE WE GOING NEXT AND AT WHAT COST?

    Clearly, where we are going next is to achieve the 1977 and  1983 goals.   Table 1
tells us that achieving those goals will upgrade the 7-8% of our waters which do not pre-
sently meet those goals.  But at what cost?

    Table 2, developed from data in the NCWQ staff draft,  sheds some light on the mag-
nitude of  the costs involved. The total capital expenditure required by industry and by
municipally owned treatment systems will  be about 233 billion dollars (1975 basis), and
the annual operating costs will  be about 16 billion dollars.  I must stress that these figures
do not include interest charges  for borrowing the  capital to finance this vast program, and
that these costs will  be spread over a period of at least 10 years.  On the basis  of about
65 million families  in the nation, these costs amount to $3,600 of capital expenditure
plus 245 $/year for each family	or about 600 $/year per family for at least  the next
10 years.

    However, the costs will in fact  be  paid by each of us in  the form of taxes and price
increases for manufactured  goods.  Since tax burdens  and buying power are graduated in
terms  of income  level, Table 2  estimates that most of us in the middle and upper middle
income brackets might expect to finance the 1983 goals by paying from 50-100  $/month
for at least the next 10 years.   But remember that these figures do not include interest
charges of inflation.  A simple  7% per annum interest charge for 10 years could easily
increase that monthly payment range to 70-140 $/month per family.

    The 1983 benefit will be an upgrading of 7-8% of our nation's waters.  The cost
might be 70-140 $/month per family  for 10 years.  Each of us will probably  assess that
balance from a different perspective  in deciding whether it is an equitable  balance.

-------
WHY GO BEYOND 1983?
     As mentioned earlier,  the second goal of PL 92-500 is the elimination of the discharge
of pollutants (EOD) by 1985.  For simplicity, I will  refer to that goal as EOD in my further
remarks.  What do we mean by EOD?  The NCWQ staff draft uses the following definition:
                 "The elimination of the discharge of pollutants shall apply to the
                 removal of those  constituents which are added during use of the water.
                 The resultant discharge  must be of equal or lower concentration than
                 that of the original supply."

     I cannot accept that as a satisfactory definition, since it does not define what con-
stitutes a pollutant. Do pollutants include simple  inorganic salts such as sodium and magne-
sium chlorides or calcium and magnesium carbonates and sulfates?  If so, on what basis
or criteria have  these simple salts  been defined as  being 'pollutants'?

     Nor can I accept the fact that the second part of the definition is not consistent with
the first part. If we simply boil some water to produce  steam, the resultant boiler blow-
down will have a higher inorganic salt concentration than the original supply despite the
fact that no salts were added during use of the water.  Or if we simply evaporate some
water in a closed loop cooling tower, the resultant cooling tower blowdown will have a
higher salt content than the original supply again with no salts being added during use.
Are these  salts truly undesirable or toxic pollutants?

     I am reminded of the following philosophical parable:
                 First philosopher:     "Why did the ancients consider the world to be flat
                                     rather than  round?"
                 Second philosopher:   "I suppose simply because it looked flat."
                 First philosopher:     "Then how would it have looked to them  if it had
                                     looked round?"
The point  of my  parable is  that  we had better take another look at this definition of EOD.
Dissolved  inorganic salts may look like pollutants to some people, but they don't look like
pollutants to me.  I would  most certainly agree with defining pollutants to include phenols,
cyanides,  BOD,  heavy metals and so forth	but I cannot agree that simple inorganic
salts are pollutants.

     If we achieve the 1983 goal,  presumably all of our nation's waters, "wherever attain-
able" will have  a quality that "provides  for the protection and propagation of fish, shell-
fish and wildlife  . .. and for recreation in or on the  water."  Then why go beyond that?
What do we gain from EOD? Table 3 presents the answer to  that question  as given in the
NCWQ report	and the  answer would appear to be that the benefits of EOD  will  be  very
minimal.  The NCWQ report is very explicit on this point (7):
                 "the predictable  environmental effects (of EOD) would appear to be
                minimal,  particularly in the absence of adequate control  of nonpoint
                sources (urban and agricultural runoffs)."

-------
    On the question of costs for achieving EOD, the NCWQ report does not quantify
total costs but only estimates unit costs in terms of $/1000 gallons for  various innovative
technologies that might possibly be used.  However, the report does say that the costs
would be (7):

               "prohibitively expensive,  and  the economic and social effect would be too
               severe to be absorbed within the foreseeable future."

    The NCWQ report  also points out that EOD technologies are considerably more energy
intensive than those required to achieve the 1983 goals (8). Although energy require-
ments for the innovative EOD technologies are estimated in terms of KWH/1000 gallons,
the report does not quantify the total energy requirements.  However,  I make my own esti-
mate (in Tables 4, 5, and 6) based  largely on  the data within the NCWQ report.   As
shown on Table 4, I have estimated the total energy requirements to achieve EOD  in terms
of barrels  per day  of fuel oil	and then expressed the fuel oil requirement as a percent-
age of our current oil imports of some 7 million barrels per day.  If  we assume a post-1983
industrial  effluent flow  of only 10% of the current nationwide  industrial water intake, and
if we assume we must produce at least a highly concentrated slurry for disposal by landfill
or by land spreading, then the energy requirements to achieve  EOD in the industrial sec-
tor will range from about 5-20% of our current oil imports.

     Based on the NCWQ's projection of post-1983 municipal effluent flows, Table 5 pre-
sents an estimate of the EOD energy requirements for the municipal  sector.  Assuming that
half of the municipal effluent can  be disposed of as a 2 wt% brine (via ocean disposal,
solar evaporation ponds, injection wells,  etc.), and assuming  that the other half must be
concentrated to a  slurry for land disposal, then the municipal sector energy requirements
to achieve EOD will amount to about 10% of our current oil imports.

     On Table 6, the total industrial and  municipal energy requirements to achieve the
1977,  1983 and EOD goals are presented.  The total will  range from 1,430,000 to 2,660,
000 barrels per day of oil, or about 20-38% of our current oil  imports.  The  EOD require-
ments relative to the 1977 and 1983 requirements are very high.  One must certainly agree
with the NCWQ's statement that the EOD technologies are considerably more energy inten-
sive than those required for the 1983 goals.

    On the basis of costs versus benefits, the  case against EOD appears overwhelming:
    — The environmental benefits would be very minimal.
    — The economic and social costs would be prohibitive and too severe to absorb in the
        foreseeable future.
     — The energy consumption would defeat any hopes we have  of significantly reducing
        our dependence on imported oil.
    Finally, in my opinion, the removal  of simple water-soluble inorganic salts from our
industrial  and municipal effluents is a complete absurdity.  As  an intellectual engineering
exercise,  we very probably could remove those salts.  But what a waste of the nation's
wealth and genius! And after the  salts are removed  as a brine  ...  or  a slurry ... or eva-
porated to dryness, how then do we dispose of them?  In fact,  how  do we prevent their
ultimate return to  Mother Nature by dissolution in rainfall or by leaching into underground

-------
aquifers? The NCWQ draft would seem to agree with me on this point:

               "The dominant consideration for achieving EOD relates to the removal
               of soluble salts.  Removal of these salts creates large quantities of brine
               which must  then be disposed of ... at very high costs. "(8)

THE TOXIC POLLUTANT STANDARDS

     "Toxic pollutant standards:  paralysis or progress?"	that was the title of a recent
article in the Journal of the Water Pollution Control Federation  (9).  The purpose of the
article was to analyze the delays in developing the toxic pollutant standards mandated by
Section 307 of PL 92-500.  Let us examine the chronological history of EPA's actions to
date on these standards:
     Published a list of 9 toxic pollutants                                July 1973
     Revised their criteria for selecting toxic pollutants,
     re-issued original list of 9 pollutants plus  25 other potential
     pollutants                                                         Sept 1973
     Proposed effluent standards for the 9 toxic pollutants                 Dec  1973
     Held public hearings on proposed standards                    Apr-May   1974
     Officially withdrew the proposed standards, and
     released a  "Draft Advance Notice of Proposed  Rule-
     making for Toxic Pollutant Standards"  in which new
     proposed standards were presented on an informa-
     tional basis                                                       Nov 1974
In the 14 months since releasing  that "Draft Advance Notice ...," no further official pro-
posals have been issued. According  to the schedule mandated in PL 92-500, the toxic
pollutant list was to be published in Jan.  1973,  the final standards were to be promul-
gated by Jan. 1974, and source  compliance was to  be achieved  by Jan. 1975	some
12 months ago.

     Much of the delay can be attributed to two  factors:  (9)
     — The lack of adequate scientific information  on the effect of pollutants at extremely
       low concentrations  in water.
     -- The counterproductive provisions in PL 92-500 which required virtually impossible
       time schedules that were subsequently forced upon the EPA by court orders, and
       which required trial-type adjudicatory public hearings.

     I can readily sympathize with the EPA on  these points.   But whatever the past rea-
sons for delay,  I would urge again that it is not too late to work with industry and with the
State regulatory agencies in developing a practical set of toxic pollutant standards upon
a firm base of scientific information.

EFFLUENT LIMITATIONS VS. EFFLUENT GUIDELINES

     As I  said earlier, recent court decisions (3)  have made  a clear-cut distinction between
PL 92-500's  Section  301 effluent  limitations and Section 304 effluent guidelines.  In the
case of the American Iron and Steel Institute versus the EPA (10),  the Third Circuit Court

-------
of Appeals ruled:

               "Section 304 requires the EPA to develop effluent guidelines to assist per-
               mit issuers in setting individual point source discharge limits.  Section  304
               is not satisfied unless EPA specifies 'factors' and 'permissible ranges' of
               effluent limitations  to be taken into account by the permit issuers in setting
               individual permit limits.  These permissible ranges  must be within the
               nationwide parameters established by Section 301 effluent limitations."

A few weeks after that ruling, the Seventh Circuit Court of Appeals in ruling upon a
similar case issued a decision (10) which agreed with the Third Circuit's ruling on this
point.

     It would appear that these rulings (unless reversed by further appeal) mean that all of
the EPA's effluent limitations promulgated to date are Section 301  nationwide parameter
limitations	and now the EPA must establish Section 304 effluent guideline 'ranges'  and
'factors'  within those nationwide limits.  I  hope that the Director of the EPA's Effluent
Guidelines Division  (who is our next speaker) will address this point	because it should
be of vital concern to industry.

     It would also appear that these rulings should lead to a re-consideration by the EPA
of the matrix approach for Section 304 guidelines which has long been advocated by the
EPA's ES&WQIAC advisory committee.  Mr.  Leon Myers of the EPA's Robert S.  Kerr
Research Laboratory  has been quoted as saying (1 1)  that his group plans to study the appli-
cation of the ES&WQIAC matrix approach to the petroleum refining industry.  I  hope that
we may hear more about these plans from Mr. Myers when he speaks to us later in this
conference.

     Personally, I have  very little knowledge of the ES&WQIAC matrix approach, and  I
have no opinions concerning the merits of that approach.  But it was certainly disappoint-
ing that the work of the ES&WQIAC was not  discussed or evaluated in depth anywhere  with-
in the 832 pages of the  NCWQ staff draft report.  That is an oversight which should be
corrected in the NCWQ's final report.

THE  PETROLEUM REFINING EFFLUENT LIMITATIONS

     Leaving the area of broad  issues,  I would now like to focus upon a specific technical
point which concerns the petroleum refining  effluent limitations as applied to treatment
plant design.

     First, it would be useful to review how the refining limitations were developed.  In
1972, the API and the  EPA conducted an extensive wastewater survey of 167 refineries.
150 of those refineries submitted comprehensive data, and  136 were considered acceptable
after screening.  Upon eliminating  those which used once-through cooling water for more
than 3%  of their process cooling, 94 were left in the final  data base.  That data base  is
probably the most realistic information ever collected for the  wastewaters from  the petro-
leum refining industry.  However, the data were based on one-time samples and did not

-------
represent attainable long-term performance.  The EPA analyzed ^the data from the 94 re-
fineries by statistical regression techniques to determine  how refinery classification, size,
and processing configuration affected the raw waste  load.  The regression analysis led to
the five  refinery subcategories, and  the size  and process factors, promulgated in the efflu-
ent limitations.

     The next step by the EPA was to determine the attainable levels of long-term (annual)
pollutant concentrations in the treated wastewater discharge from each of the five refinery
subcategories, as well as the attainable long-term average wastewater flows. The data
for these determinations were obtained by visiting  refineries, reviewing monitoring records,
and from various other sources.

     Finally, data were gathered  on  the day-to-day  and  month-to-month variations in
ref'nery  wastewaters discharges.  These data  were  analyzed by statistical  techniques to
arrive at variability factors which related long-term averages to short-term values.  Thus,
the attainable annual averages were multiplied by the applicable variability factors to arrive
at the daily or 30-day averages in the final promulgated effluent limitations.

     Unfortunately, the attainable long-term averages (flow and concentrations) and the
variability factors are not included in the actual published  EPA regulations (40  CFR 419).
Due  to the various amendments and revisions  that have taken place, the unclear presentation
in the original Development Document, and to the fact that the preambles to the various
Federal Register promulgations are not included in the final Code of Federal Regulations —
it is  virtually impossible at this point in time to determine the applicable  long-term averages
and variability factors with any degree of certainty.  The waste treatment  plant designer is
left only with the amended short-term effluent  limitations as published in  the Code of
Federal Regulations.

     However, it is most important that the waste treatment plant be designed to achieve
the more stringent long-term levels if the plant is not to  exceed the allowable peak levels
for daily and 30-day averages.  For  that reason, I have attempted to reconstruct (as best
I  could) the applicable variability factors and equivalent annual averages.  Tables 7 and 8
present some of those reconstructions for:
              1977    limitations for existing sources (30-day averages)
              NSPS    limitations (30-day averages)
These reconstructions reveal a number of anomalies.   These anomalies may be due to my
inability to correctly reconstruct  the currently applicable information, or they may be due
to inconsistencies which the EPA  should correct. For example,  in Table 7 for the 1977
existing source limitations:
     —The annual  average  values were obtained from dividing the promulgated  30-day
      averages by the annual average wastewater flow and by the  variability factor,  and
      applying a conversion factor to arrive at ppm.
     —We can see a number of trivial anomalies in values,  most probably due to incon-
      sistent round-off procedures.
    --But in the COD values, the anomalies amount to as much as 20% between the
      extremes.  I am fairly sure that the four COD values, shown  enclosed in a box,
      were not meant to exhibit  a 20% difference.

-------
                                                                                 11
     —Looking at the four NhL-N values, shown enclosed in a box, I am at a loss as to
       why the more complex refinery subcategories (Lube Oil and Integrated) are expected
       to achieve 38% lower NHL-N values than the less complex subcategories.
 Table 8  illustrates the same anomalies for the New Source Performance Standards, and here
 the four NHL-N values vary by a factor of 2-to-l.

     Since the waste treatment plant design  should be based on attaining the annual averages,
 I would  strongly urge the EPA to:
     —Publish a clear and explicit listing of the latest applicable values  for long-term
       flows and concentrations in the format of Tables 7 and 8 —  for daily averages as
       well as 30-day averages, and  for the  1983 limitations as well as 1977 and the NSPS.
     —Publish a clear explanation of the COD and NhL-N anomalies that appear to exist.

 SUMMARY

     In closing,  I wish to express my deep appreciation for the opportunity to address this
 conference	and for your patience with my lengthy address.  I  can only hope that it will
 stimulate some open  discussion of  the issues  we face, and  I look forward to listening to and
 hearing  what the other speakers have to say.

 REFERENCES

 (1) Environmental  Industry Conference, Council on Environmental  Quality,  Washington,
     D.  C.,  December 10,  1975
 (2) Staff Draft Report, National Commission on Water Quality, November 1975, page V-6
 (3) American Iron  and Steel Institute vs. EPA, Third Circuit Court of Appeals, November
     1975
     American Meat Institute vs.  EPA, Seventh Circuit Court of Appeals,  December 1975
 (4) Staff Draft Report, National Commission on Water Quality, November 1975, pages
     1-1  and  1-2
 (5) Ibid, pages  1-23 and 1-24
 (6) Ibid, page IV-10
 (7) Ibid, page 1-74
 (8) Ibid, page 76
 (9) "Toxic Pollutant Standards:  Paralysis or Progress?", JWPCF, December 1975
(10) 8 ERC 1321, Environment Reporter,  Decisions No. 10, November 28, 1975
     8 ERC 1369, Environment Reporter,  Decisions No. 11, December 12,  1975
(11) Environment Reporter, Current Developments, page 1252, November 14, 1975

-------
12
DISCUSSION
Umesh Mathur:  I  found  your comments extremely interesting.
                                                             work for the Indian Nations
 ,	^TGovernments'and we are a designated Section 208 planning agency.  All of our
 guidelines (for Section 208 long-term regional planning) tell us to assume that  industry will
 meet the EPA's effluent limitation standards. We can then evaluate the long-term effect
 of industrial discharges within our region on that basis.

     However, if our evaluations on that basis indicate  that long-term water quality goals
 will not be attained,  what are we to do then?  Can  industrial NPDES  permits call for  more
 stringent limitations because Section 208 long-range planning indicates that more stringent
 controls may be needed?

     Is there any real  purpose in long-term  planning  based on receiving water quality when
 NPDES permits are issued  on technology-based effluent standards?

 Milton Beychok: I am not sure that I fully  understand your questions.   Each individual state
 already has the  option of setting NPDES effluent limits  which are more stringent than the
 established EPA guidelines.  And the court decisions (which I discussed in regard to effluent
 limitations versus guidelines) seem to mandate a  range of values to assist the permit issuers
 in deciding how stringent  to make the individual permit.  But I don't know how a  Section
 208 agency interfaces with the state NPDES permit issuers.

     As for technology-based standards versus long-term receiving water quality, I guess
 that Congress  crossed  that bridge when they wrote  PL 92-500 and decided upon technology-
 based standards.

 Umesh Mathur:  Perhaps we can  have the EPA speakers address this point later on, because
 I think it is of great importance.

 BIOGRAPHY

     Milton R. Beychok  is a consulting Chemical
 Engineer in the  field of environmental  technology.
 He has a B.S. degree in Chemical Engineering
 from Texas A &  M University, and he is a regis-
 tered professional engineer in California and Texas.
 He is a Diplomate of the American Academy of
 Environmental Engineers and  is a member of the
 AlChE, Air Pollution  Control Association and the
 Water Pollution Control Federation. He has
served on  the  California Water Quality Control
Board and consulted for  the EPA, the National
 Science Foundation and the National  Commission
on Water Quality.  Prior to entering private
practice,  he was with Fluor Engineers &
 Constructors for 20  years.

-------
                        TABLE I  "WHAT DO WE ACHIEVE BY 1983 COALS?'
                                                                                              13
            0.
            Q.
            Ill
            O
            0
            Ul
            >

            o
            VI
                                              (4 PPM DISSOLVED OXYGEN)
                MINIMUM GOAL DURING THE
                     LOW FLOW SEASON
7-8% OF WATERWAYS DO NOT PRESENTLY
MEET THE 1983 MINIMUM  GOAL
                                                                 I
                            10%
           20%
30%
40%
50%
        % OF WATERWAYS WITH DISSOLVED OXYGEN EQUAL TO OR LESS THAN LEVEL SHOWN
                  Reference:  Page 1-24, NCWQ Staff Draft Report, November 1975


Involving 9, 000 river and stream miles, 10, 500 miles of coastline, 6,000 square  miles of estuaries, and  6,000
square miles of lakes and reservoirs.  The total  surface of the watersheds where each of the sites is located
encompasses 20% of the nation's land surface and a population of 80 million people (6) .
                             TABLE 2  "WHAT WILL 1983 GOALS COST?"
                                            CAPITAL
                                             COST ($)
 ALL INDUSTRY (1)
 MUNICIPAL SYSTEMS (2)

      TOTAL
          75,000,000,000
         158,000,000,000

         233,000,000,000
           ANNUAL
          OPERATING COSTS ($)

         13,900,000,000
          2,000,000,000

         15,900,000,000
                   ESTIMATED MONTHLY COST PER FAMILY FOR NEXT 10 YEARS (3)
                                 (IN TAXES AND PRICE INCREASES)
 INCOME LEVEL:

 29,000 - 37,500 $AEAR

 20,700- 29,000 $AEAR

 17,700- 20,700 $AEAR

 13,100- 17,700 $/YEAR
                          MONTHLY COST

                         $80 - 100/MONTH

                         $50 - 65/MONTH

                         $40- 50/MONTH

                         $25- 30/MONTH
 (1) Page II - 81, NCWQ Staff Draft     (2) Page II - 28, Ibid
                                         (INTEREST COSTS AND
                                          INFLATION ARE NOT
                                          INCLUDED )
                                 (3)  Page II I- 273, Ibid

-------
14
BOD REDUCTION - -
                 TABLE 3  "BEYOND 1983 . . . WHAT DO WE GAIN FROM EOD?"

                              (By achieving EOD for point sources
                              after achieving 1977 and 1983 goals)

                                  INSIGNIFICANT COMPARED TO UNCONTROLLED NON-POINT
                                  URBAN RUNOFF AND AGRICULTURAL RUNOFF
TOXIC SUBSTANCES REDUCTION - -
EUTROPHICATION REDUCTION	
SUSPENDED SOLIDS REDUCTION	
OVERALL BIOLOGICAL IMPACT -
                                  MINOR COMPARED TO UNCONTROLLED URGAN AND
                                  AGRICULTURAL RUNOFFS
                                  VERY LITTLE SINCE NON-POINT SOURCES CURRENTLY ACCOUNT
                                  FOR 80% OF TOTAL NITROGEN AND 50% OF TOTAL PHOSPHORUS
                                  ENTERING OUR WATERWAYS
                                  VERY LITTLE SINCE 90% OF THE LOAD ENTERING OUR WATERWAYS
                                  IS FROM NON-POINT SOURCES
                                  ONLY 6-10% OF WATERS PRESENTLY UNSUITABLE WILL BE UPGRADED.
                                  THUS, 5-10% of 7-8% OR PERHAPS ABOUT 0.4 - 0.8% OF NATION'S
                                  WATERWAYS WOULD BE UPGRADED.
              Reference:  NCWQ Staff Draft Report ( Nov.  1975)  Pages 1-77 and 1-78
                TABLE 4  "HOW MUCH ENERGY TO ACHIEVE EOD FOR INDUSTRY?"
                       GROSS
                        USE
 NET INTAKE (1)

15 x 10  gals/yr
                            ALL
                          INDUSTRY
                                         EVAPORATION,
                                         CONSUMPTION,
                                        '  LOSSES
                                            EFFLUENT
                                       ca. 10%/of INTAKE
      EVAPORATION
      TO A SLURRY
                                               50-200 	
                                               KWH/1000 gals

                                             ENERGY CONSUMED TO ACHIEVE EOD
                                                DISPOSAL
                                                                                OF SLURRY
EFFLUENT (AFTER 1983)	

% OF/INTAKE  (1)  GALSAR   KWH/1000 GALS (2)   KWH/YR  EQUIV./BPD OIL (3) % OF CURRENT/IMPORTS (4)
     10
     10
               1.5x 10'
               1.5 x 10
200
 50
                                               3x 10
                                                   11
                                             0.75x 10
                                                     11
1,371,000
  342,750
19.6
 4.9
(1) CURRENT INDUSTRIAL WATER NET INTAKE  = ca. 15 x 10 GALS/YR  (p.  11-53, NCWQ Staff Draft).
    EFFLUENT (AFTER 1983) ASSUMED TO BE 10% OF INTAKE.

(2) 50-200 KWH/1000 GALS (60-240 BTU/POUND) TO ACHIEVE A CONCENTRATED SLURRY (p. 11-144, Ibid)

(3) EQUIVALENT BPD OF OIL =  (KWHAR) ( 10,000 BTU/KWH)/(365 DAYS/YR) (6 x 106 BTU/BBL OIL)
                         = (KWHAR) (4.57x 10" )
                     BPD  =  BARRELS OF OIL PER DAY, WITH HEATING VALUE OF 6 x 106 BTU PER BARREL
         10,000 BTU/KWH  =  FUEL CONSUMED IN GENERATING ELECTRIC POWER AT ca. 33% EFFICIENCY
(4) CURRENT OIL IMPORTS TAKEN TO BE ca. 7 MILLION BARRELS PER DAY

-------
                                                                                                 15
             TABLES  "HOW MUCH ENERGY TO ACHIEVE EOD FOR MUNICIPAL EFFLUENTS?1
                                                                               OCEAN DISPOSAL,
                                                              	      SOLAR EVAPORATION,
                                                               50%          INJECTION WELLS, ETC.
  EFFLUENT FLOW      5(Mo-1     0.7x 1Q9 GALS/DAY (1)
(AFTER 1983 GOALS) CONCENTRA  CONTAINING 2 WT%
        9               TION       DISSOLVED SOLIDS                        _
  35x10 GALS/DAY                                                        EVAPORATION  LANDFILL
  CONTAINING 0.04                                           J£g _   TO A SLURRY   _
  WT % DISSOLVED                                                         _
                  10 KWH/1°°°                                            50-200 KWH/1000 GALS(2)

  CONCENTRATION = (35 x 109 GALS/DAY)(365 DAYSAR)(10 KWH/1000 GALS)(4.57x 10"6)
                   = 584,000 BPD OF OIL

   EVAPORATION   = (0.35 x 109 GALS/DA Y)(365 DAYSAR)(50 to 200 KWH/1000 GALS)(4.57 x 10"6)
                   = 29,000 to 117,000 BPD OF OIL

  TOTAL ENERGY REQUIRED  =  613,000 to 700,000 BPD OF OIL
                               8.8 % to 10.0 % OF CURRENT IMPORTS

  (1) Page 11-155, NCWQ Staff Draft
  (2) Page 11-144, IBID
        TABLE 6  "HOW MUCH ENERGY (NATIONWIDE) TO ACHIEVE 1977, 1983 AND EOD GOALS?"
                                                                          % OF OUR
                                              10  BPD OF OIL           CURRENT IMPORTS
 Manufacturing Industry:
     1977 goal  (1)                                0.17                         2.4
     1983 goal  (2)                                0.20                         2.9

 Steam-Electric Industry:
     1983 thermal  limitations (3)                 0.02-0.13                    0.3-1.9

 Municipal Effluents:
     Treatment, collection, sewer control (4)     	0.09	            	1.3

 TOTAL TO ACHIEVE 1983 GOALS               0.48-0.59                    6.9-8.5

 EOD for industry  (5)                          0.34-1.37                    4.9-19.6

 EOD for municipal effluents (5)                  0.61  0.70                    8.8- 10.0

 TOTAL BEYOND 1983 (w/EOD)                  1.43 - 2.66                   20.6 - 38.1

 (1)  376 trillion Btu/year (p.  11-100, NCWQ Staff Draft)
 (2)  446 trillion Btu/year (p.  11-100, Ibid)
 (3)  45 to 294 trillion Btu/year (p. 11-106, Ibid)
 (4)  Page 11-41, Ibid
 (5)  See Tables 4 and 5

-------
16
                  TABLE 7 "1977 EFFLUENT GUIDELINES FOR PETROLEUM REFINING (1)"
                   EFFLUENT LIMITATIONS IN POUNDS/1000 BBLS OF CRUDE OIL (2,3)
SUBCATEGORY
TOPPING
CRACKING
PETROCHEMICAL
LUBE OIL
INTEGRATED
BOD-5
4.25
5.5
6.5
9.1
10.2
TSS
3.6
4.4
5.25
8.0
8.4
COD
21.3
38.4
38.4
66
70
OIL
1.3
1.6
2.1
3.0
3.2
EQUIVALENT ANNUAL

TOPPING
CRACKING
PETROCHEMICAL
LUBE OIL
INTEGRATED
VF(4)
TABLE

15.0
15.5
15.3
14.3
15.0
1.7
8 "NEW

10.2
10.1
10.0
10.2
10.2
2.1

79.9
115.2
96.0
110.0
109.4
1.6

4.9
4.8
5.3
5.0
5.0
1.6
PHENOL
0.027
0.036
0.0425
0.065
0.068
AVERAGE p

0.10
0.10
0.10
0.10
0.10
1.7
NH3-N SULFIDE
0.45
3.0
3.8
3.8
3.8
.p.m.

1.8
9.6
10.1
6.8
6.3
1.5
0.024
0.029
0.035
0.053
0.056
Cr
0.071
0.088
0.107
0.160
0.17
CONCENTRATIONS

0.10
0.10
0.10
0.10
0.10
1.4

0.25
0.25
0.25
0.25
0.25
1.7
Cr
0.
0.
0.
0.
0.
(4)

0.
0.
0.
0.
0.
1.
\J 1
0044
0056
0072
on
on


02
02
02
02
02
4







(5)
20.0
25.0
30.0
45.0
48.0

SOURCE PERFORMANCE STANDARDS FOR PETROLEUM REFINING (1)"
EFFLUENT LIMITATIONS IN POUNDS/1000
SUBCATEGORY
TOPPING
CRACKING
PETROCHEMICAL
LUBE OIL
INTEGRATED

TOPPING
CRACKING
PETROCHEMICAL
LUBE OIL
INTEGRATED
VF(4)
BOD-5
2.2
3.1
4.1
6.5
7.8
TSS
1.9
2.5
3.3
5.3
6.3
COD
11.2
21
24
45
54
OIL
0.70
0.93
1.3
2.0
2.4
PHENOL
0.016
0.020
0.027
0.043
0.051
EQUIVALENT ANNUAL AVERAGE p.

14.8
15.6
15.2
15.0
15.1
1.7

10,3
10.2
9.9
9.9
9.9
2.1

80.0
112.5
94.7
110.7
110.9
1.6

5.0
5.0
5.1
4.9
4.9
1.6

0.11
0.10
0.10
0.10
0.10
1.7
NH3
0.45
3.0
3.8
3.8
3.8
p.m .

3.4
17.1
16.0
10.0
8.3
1.5
BBLS OF CRUDE OIL
-H SULFIDE
0.012
0.017
0.022
0.035
0.042
Cr
0.037
0.049
0.068
0.105
0.13
CONCENTRATIONS

0.10
0.10
0.10
0.10
0.10
1.4

0.25
0.25
0.25
0.24
0.25
1.7
(2,3)


Cr +
0.
0.
0.
0.
0.
(4)

0.
0.
0.
0.
0.
1.
0025
0032
0044
0072
0084

02
02
02
02
02
4





(6)
10.5
14.0
19.0
30.5
36.5

 NOTES FOR TABLES 7 & 8
 (1) AS AMENDED ON MAY 20, 1975
 (2) BEFORE APPLYING SIZE AND PROCESS FACTOR MULTIPLIERS.  BASED ON 30-DAY AVERAGES.
 (3) EXCLUDING ALLOCATIONS FOR RUNOFF AND BALLAST WATER
 (4) EQUIVALENT ANNUAL AVERAGE p.p.m.
                                  =    (30-DAY AVERAGE,  LBS/1000 BBLS CRUDE) (120)	
                                      (ANNUAL AVG. WASTEWTR. FLOW, GALS/BBL CRUDE) (VF)
   VARIABILITY FACTOR   (VF)
                                  =   RATIO OF 30-DAY AVERAGE TO ANNUAL (LONG-TERM) AVERAGE
 (5) ANNUAL AVERAGE WASTEWATER FLOWS USED AS BASIS FOR 1977 GUIDELINES  PER EPA DEVELOPMENT
    DOCUMENT, EPA-440/l-74-014-a, APRIL 1974
 (6) ANNUAL AVERAGE WASTEWATER FLOWS USED AS BASIS FOR NSPS GUIDELINES  PER EPA DEVELOPMENT
    DOCUMENT, EPA-440/1-74-014-a, APRIL 1974

-------
             SESSION
"PETROLEUM REFINING GUIDELINES"
    Chairman
    William C. Galegar

    Director, Robert S. Kerr Environmental Research Laboratory
    U.S.EPA  Ada, Oklahoma
    Speakers

    Allen Cywin

    "Establishment of Petroleum  Effluent Guidelines"


    Martha Sager

    "The Role of the Effluent Standards and Water Quality  Information
    Advisory Committee  (ES&WQIAC) in  Furthering the Scientific
    and Technical Developments Towards  Clean Water"


    Joe G.  Moore, Jr.

    "The Role of the  National Commission of Water Quality(NCWQ)


    Robert  T. Denbo

    "Economic Impact of Wastewater Effluent Guidelines for
    Petroleum Refining"
                    17

-------
18

-------
        "ESTABLISHMENT OF PETROLEUM REFINING EFFLUENT GUIDELINES"

                                Al len Cywin
                    Director, Effluent Guidelines Division
                         E.P.A.,  Washington, D.C.
     Let me start by saying  I am always pleased to be  back in Oklahoma, having learned
how to  fly here in World War II at Norman; and I am  always pleased to talk and partici-
pate in conferences with the petroleum industry.  Mr. Brinegar, President of Union Oil
was  talking to me  recently and  said "You know you folks in the Environmenta1 Protection
Agency and the environmental movement ought to thank our industry as we got you started
as a result of such incidences as the Tory Canyon and Santa Barbara oil spills. "  Well,
of course,  there are other industries and  there are other reasons and we have had a whole
series of laws in which this nation tried  to tackle the  problem of water pollution control.

     Three years ago the Congress over Presidential veto enacted  Public Law 92-500.  Now
why did they do so - why did the public demand cleaner water  - why did the Congress
intensely debate the issue for two years  - why did they pass one of the most comprehensive
pieces of  legislation in order to establish new mechanisms and specific goals for  water
pollution  control departing in significant ways from a  whole series of  laws passed }ust two,
three, four, five  years earlier?   Now why does the law state in its  very first sentences,
in the very first page,  that the objective of this Act is to restore and  maintain the chemical,
physical and biological integrity of the  nation's waters?  In order to achieve this objec-
tive it is hereby declared that consistent with the provisions of  this  Act, I) it is the
national goal that the discharge of pollutants into navigable waters be eliminated by 1985;
now that is the goal.  Is the national goal that wherever obtainable an interim goal of
water quality which provides for  the protection and propagation of fish,  shell fish and
wild life and provides for recreation in and on the water be achieved  by July I,  1983; now
that's a goal also.  It is a national policy that the discharge of toxic  pollutants and toxic
amounts be prohibited; now that's a policy.   In order  to accomplish these goals the Act
created a new regulatory mechanism requiring uniform technology based effluent standards.
As you  know a  level of standards for existing  plants cal'ing for  Best Practical Control
Technology is  to be achieved by  July I,  1977 followed by a more  stringent  requirement
to be achieved by July I, 1983  which, and I  am quoting from the  Act, "shall require
application of the Best Available Technology Economically Achievable for such  category
or class which will result in reasonable further progress toward the national goal of
eliminating the discharge of all pollutants. "  The Congress places faith in technology,
practicable, available,and achievable by 1983.  Why?

     For one reason industrial wastes are  the principal point source of controllable  water-
borne wastes.   In terms of the generally quoted measurement of just BOD,-, fhe gross
wastes of industries are about three times greater than that of the domestic sewered waste.
Moreover the rate of U.S.  industrial production is increasing at a rate of over four  times
of that of the population growth rate.  But BOD,, is just one consideration, industry also
discharges the bulk of the metals, chemicals, toxicants, oils and  greases to our waterways.
Recent drinking water surveys illustrate  the great amounts of noxious waste discharges that
provide a  threat to our drinking -water supplies.  Therefore of paramount concern is  that

                                        19

-------
20

public health now and in the future is at issue.  Also at issue are other major beneficial
uses of water - for agricultural purposes, for recreational  purposes and  indeed for use by
industry itself.  Industry which presently uses three times  the volume of water required  for
domestic water supply is a major beneficiary of water pollution control.

     Now let us look at the mechanisms for achieving industrial water  pollution control in
an equitable fashion on a  national basis.  As 1 have said industry has a stake in meeting
the objectives of the Act since it not only is a water pollution control  act but a water
conservation act.  The  U.S.  needs for industry,  for agriculture and for domestic use, are
increasing all the  time, but as we all know, the volume of water available  in the United
States is finite. The  1972 amendments to the Federal Water Pol'ution Control Act, have
as a goal  the elimination  or  reduction of pollutant discharges through  effluent limitations
that encourage recycle and reuse and the  law is to be  implemented through  a series of
permits issued by the  Federal Government and the States.   At this date, we have more  than
half the states of the union implementing the program.

     The industry effluent limitations are to  be technology based; progress, not status quo,
is to be emphasized.  Two new terms are coined, best  practical control technology avail-
able based on the  average of the best practices within an  industry or class or category of
industries and best available technology economically achievable, which could be based
on a single application within an industry.  Technology transfer is permitted where there
is little or no technology being practiced and, of course,  we are interested in reducing
the amount of pollution not just the measurement of pollution so mass units are used wherever
possible for our limitations.

     Now for industry, and by the way industry is required to do more  than  municipalities,
best practical  is to be achieved by July I,  1977, best  available by July I,  1983 and also
new source performance standards are to be  established and the latter is to be based on best
available demonstrated technology or technology transfer. Now  these  are for point sources
that discharge directly  to  the waterways.

     In addition to that, we are to establish pretreatment  regulations and  these could be
and usually are somewhat different for existing plants that are tied into municipal systems
and for new sources.  By the way we have something like  40,000 industrial  point source
discharges and we have something like 300,000  commerical and industrial establishments
on municipalities, but basically ingoing through the program we have  had  to establish
some  five different regulations,  and  there they are.   Now the July I,  1977  regulation, as
I  said before is to  be based on the average of the best  practices,  and by the way these
terms  are  not something we conceived at EPA.  In establishing what a  law requires one has
to follow  the history of the law and it does get complicated in our democracy because
often  times as the  House works through its side it differs somewhat from the  Senate; then
there's a conference committee and a conference report and reports to the Senate and
congressional record.  These  ideas and these thoughts on the  effluent regulations are
therefore  based  not only on public discussions of the issue, intense discussions and debate
within EPA and  within the federal establishment, but also on briefing the Congress on a
very regular routine basis  on  these ideas and these thoughts throughout  the development
arid history of this program.   Everything that we have done from the very first definitions

-------
                                                                                  21
has been supplied  to fhe Congress.  There are a number of factors that we have to rake
into account.  By  fhe way in establishing these kinds of regulations, no one had ever
done this before and no one had token into account1 many of these  factors which became
quite important as fhe  country entered into inflation, recession and energy crises.  But
right from the beginning we did take into account as the law required the total cost
versus  the effluent reduction benefits,  the age of the plants,  where this was a factor,
the reduction processes, where these were factors, process changes,  engineering  aspects,
non-water quality environmental impacts, in other words what are the tradeoffs between
transferring  waste  from one media to another, etc.  There are only three places to put
waste  - air, land and  water.

     Now occasionally, as I said before, we  do practice technology transfer.  There are
industries that are doing very little or  nothing, yours is not one of these.  The 1977
standard emphasizes end of process  treatment but  it does include in-process treatment
where  this is common practice and in very many industries there are very good housekeep-
ing practices on-going and certainly should be used across the industry.  The objective
is to reduce the point source pollution discharge.  Mr.  Moore will be discussing the
aspects of the Act which go to  questioning whether or not this particular date or this
particular goal should  be achieved  but the 1983 goal in the Act  (the only way one could
study this is to establish what the standards are) is quite a difficult one.  It establishes
that really  you had better take  the best available technology and apply  this across the
industry.  The law which was passed on October 18,  1972 envisioned that there would be
a ten year effort to reach that objective and  in-process controls are suggested. Alter-
native water uses, water reuse  conservation,  by-product recovery,  reuse of waste -
water constituents, good housekeeping and quality control are all techniques in use in
your industry and other industries and which can be utilized in many other applications.
The objective in the Act, the ultimate objective  is no  pollutant discharge.  Now in fact
a few of our regulations do recognize that certain plants and  certain industries are indeed
achieving no pollutant discharge at the present time  but in most cases the regulations do
have limits and these limits are  not zero. Now to go through the thought processes we
take a major industry category,  and some 28  were published as a minimum in  the  basic
law, and within that industry we look  at like  families, or classes or categories (as the law
states) of plants, by type,  by size,  or  by age, or by  climate, or by treatability or finished
product and we call these sub-categories.  We then establish what the raw waste loads
are for these sub-categories, what the best or the average of the best in-planf controls
are and fhe  end of pipe treatment and  come out with our standards.  Once again  there
are five sets of standards.  We  look at the annual  costs, both infernal and external. This
was a departure from previous rulemaking in most other agencies in  that we actually took
our costs and tried to establish what the economic consequences of these regulations were
and these were considered and debated within and without the federal establishment before
fhe regulations were promulgated.  Now for publicly-owned  treatment works,  which are
referred to  in the law as POTWs, fhe law requires secondary  treatment by July I, 1977
and if can require  industry prefreatment and it does say in fhe law that industrial waste
which cannot be treated or will  interfere with the operation of the municipal treatment
plant or pass through untreated should  be prefreated.  The law established fhe technology
based limitations as a minimum;  if further stated that where wafer quality could not be
achieved by applying fhe technology based limitations something more stringent may be

-------
 22

required.  It also said that states may make their  limitation more stringent than the  Federal
limits for whatever reasons occurred to the states.  Now let me just explain some of the pro-
cedures we have gone through in order to  come up with this.   First of all the  law itself
expressed the view that for existing plants and at least for those 28  industries ( and  the
petroleum refining was one), that we should publish proposed regulations twelve months
after the law was passed from October 18, 1972 to October 18,  1973 and some four or five
months thereafter the new source standards.  We started the program building a little bit
from the 1899 permit program.  Recall that the Corp of Engineers suddenly discovered and
we discovered about 1970 that the 1899  Rivers  & Harbors Act called for a "permit" system
for discharges to waters.  For a variety of reasons no more than 20 permits were issued in
that two year period and by that  time Public Law 92-500 overtook the effort.  Now if you
recall the President vetoed the bill and the Congress overrode the veto on October  18,
1972.  Because of that we were inhibited somewhat in building up the resources necessary
to implement the bill as quickly as we would  have liked to, but in any event starting with
a secretary and myself about four years ago and still only with a real small handful  of
staff, we undertook to hire outside contractors to help  us.  We can  start with the fact that
good contractors were hard to  come by in  a hurry. With regard to previous comments by
Mr. Beychok, I  remember asking, if Fluor wanted  to participate and Mr. Beychok said "Oh
no1.", so we  had to pick and choose the best we could. We also set up certain other review
mechanisms.  The  law provides for us  to consult with other  federal agencies,  states  and
other interested  persons.  Now of course the people affected  by the regulations are  the
most interested persons, but you also have environmental groups and professional societies
and others.   Now EPA, in establishing what our program plan would be, where our  dead-
lines would be,  who our contractors and who our  project officers would be, what our
definitions would be, called in many groups, other federal  agencies, states,  industrial
organizations, environmental professional groups and explained what we were going to do
and solicited their cooperation.  Now a lot of people didn't think the whole  thing was for
real although, of course, many did.  Of course,  the cooperation  was spotty across
industries and even between companies within industries.  Within your industry,  an
environmental committee was established headed by Russ Mallett and we had a very good
dialogue with that group throughout the rulemaking process.  Now we also had to establish
our effluent standards and water quality information advisory committee and get  that set
up and for a  little while I was the executive/secretary/director of that committee and
carrying water on both shoulders.  We asked that committee to hold public hearings and
Dr. Sager will go  into  that aspect of their operations.  Faced with these deadlines, EPA
set some very stringent timetables on the contractual efforts and we told our contractors
that their very first draft report would be exposed to the public.  Now this is highly
unusual, I don't think anyone  here would  want to expose his first draft effort  to the  public,
but we did.  We did it for a purpose,  part of the  purpose went with a covering letter that
went to  the public.  We said,  that if you  don't like what you see in this document, please
let us know what you would  like to see but give us your background, your data,  your
rationale —  for any changes,  we just can't pluck things out of the air.  In some  cases this
was successful, in other cases  not so successful.   In addition to that external  review;
within EPA,  we  tried to establish throughout the agency regional office and lab  personnel
who might be knowledgeable in a particular industry and we set up  what we call working
groups of these people  for an internal  review.   These were  technical task  forces, and  we
ourselves --  (we just didn't wait for comments to  come in) we ourselves went  over these
materials and in some cases wrote and rewrote these documents several times.  We then

-------
                                                                                 23
as an agency came out with a proposed regulaMon.  This is the first time EPA's name
appeared on it and a development document which was a manual which explained how
we got from here to there.   Everyone of the  factors in the Act is a  chapter, purposely
written  in that guise.  We also  took all of our background  data, file  cabinet loads, and
made these available to the public in what we call a  "freedom of information center, "
and once again we asked the public to tell us what it is they would like to see in  there
and give us their data, their background and their rationale, for any changes.

     Now the timetable of 30 days or 60 days whatever they turn out  to be were very
short because during this period of time, not only were we faced with the congressional
deadline which we didn't quite make (we made it for a number of regulations but not quite
all) but the National Resource & Defense Council took us to court and the court set us on
specific deadlines.   Now what  the court said is you will do competent professional work
on this schedule and they told us what the schedule was.  We have tried and so far we
have stayed out of contempt, but in addition to that the court also  added another  30
industrial categories, because the first 28 or 30 covered about only half the permit
applications.  Now in addition  to all of that,  of course, as I say there are some five
regulations and we have  established some well over 500 sub-categories so we have
promulgated and proposed over  2500 regulations in less than four years.  As a  part of
the regulation writing  process !n the United  States in fact as a part of our very basic
system of democracy we have three parts of government.  We have  the executive, the
congressional and the judiciary and too many people  lose sight of the fact that the rule-
making  procedures under the administrative procedures act and also under this Act,
Public Law 92-500,  call  for public lawsuits if the public so desires.   Now a number of
our regulations were not  sued, and a number of the lawsuits that were entered against
us we have long since  settled out of court, some for, to me, very trivial reasons.  1 will
give you one.  The textile  industry sued us and before they did they came in and said we
want to take on a research  study to illustrate whether or not the projected numbers which
we have reasonably agreed to for the 1983 standards could  really be met through mi xed
media filtration. If we,  the textile industry, do that would you look at the results and
we said as public servants we are obligated to do  that and  the law  tells us to look at
this every few years.  They said, well our lawyers tell  us that we have to have this in
writing  for after all you might not be here by the  time our results come in so they  sued
us and we and they developed a letter.  We said if they come up with the results we  will
look at  it and  that was the  way that suit was settled.  Others were  a  little more compli-
cated than that and others have gone right to the  mat.

     Now with respect to the two law suits previously mentioned by Mr.  Beychok, I will
give you my interpretation. Neither Milton Beychok nor  I are  lawyers, and perhaps
this will all end up in the Supreme Court.  The Seventh Circuit did not  mention ranges
and  I think in my view I  could not have written a better treatise for the  position taken by
EPA, than they did.  The Third  Circuit, although it did mention range,  qualified  that
term by saying that we have to give very precise guidance to the permit writers as to how
to apply anything over and  above that minimum number.  We will be  wrestling with  this
issue in  the months ahead.  We  have still to hear  from several other Circuits.  I think of
great significance is the  fact that we are all citizens and we have all expressed our  views
and indeed the House of  Representatives had an oversight hearing on  our program a year

-------
  24


and a half ago; something like the Church Committee on  Intelligence.  I  would commend
the report resulting from that hearing to your reading.

     My summary of it is that the House said to itself; gentlemen this is a  very complex law,
people of goodwill are trying to work their way through it and let the process continue.
With that I think I am on time to answer questions.

DISCUSSION

W. C. Galegar:  We have talked primarily about history of the guidelines program, what do
you think the future holds in this area?

Allen Cywin: Well immediately ahead  if we did nothing  else, we still have a  number of
industry categories to get out still  in the court order deadlines, for example by January 30
(this Friday)  we  have to get out pulp and paper - phase II and we have a  series of others to
develop and  these are all in process.  Every time we promulgated a  new source standard as
law required we also promulgated a new source pretreatment standard.  We have a very
complex procedure in existence for pretreatment standards for existing sources and we have
undertaken a program to try to clarify that somewhat better.  So we still have a lot to do
in the field of pretreatment.  More than that last spring we proposed to the agency (I am
just spinning this off and I know it is being taped so take  this in the context of just discus-
sion), a conception whereby 307a, would be used really  as a last resort.   For toxics  in
general, once they are  established on a list,  should be integrated into  301 and 306 as we
go about reviewing and  upgrading  those regulations. Conceptually  this seems to be taking
hold in  the agency and  conceptually we are negotiating with NRDC and others on this
issue.  Here again we have been taken  to court by the  environmental organizations to get
on with the  job in both  pretreatment and in toxics.  So basically we still  have a number of
regulations to finish.  The court suits themselves will give us a lot of work to do for exampl e,
the Eighth Circuit court remanded  some of the standards in grain milling and just gave us an
opportunity  to go back and look at it again.  As public servants, if  our review had shown us
to be wrong  we would easily have admitted it, but in this case we felt  we were correct and
added 2500 more pages to the record to show that the standards were based on sound data.
The lawyers again in responding to that for the other side said all you did was prove you
were right in the first place and therefore you are arbitrary and capricious.  So some call
Public  Law 92-500 the lawyers employment act and  in some cases,  the  lawyers call it the
engineers employment act.   Perhaps it's both, but there is an awful  lot of work to do  to
finish what we have started.

Frank Manning:  Would you care to comment how you selected the parameters that were
Iisted in your guidelines and promulgations,  for example  BOD,  why did you choose that,
instead of COD or TOC?

Allen Cywin:  Basically, of course, it started with a look at the industry  and the common-
ly regarded prinicpal pollutant parameters.  Often times  as we developed our regulations
we dropped parameters because  we thought it might  cost too much to monitor and we knew
that a certain amount of reduction  would take place anyway if the  technology  was employed,
so we tried to keep our parameters to a minimum. In the  case of the petroleum refining

-------
                                                                                 25
industry we had been discussing the parameters even before the law was passed with your
technical  advisory committee for a long period of time and we are in general agreement
on the parameters that were selected.

Milton Beychok: Allen,  I'm not sure whether you have had a chance yet to read the
Commission's staff draft report but what is your personal opinion about the 1985 goals and
the elimination of discharge as they relate to the removal  of dissolved inorganic  salts?

Allen Cywin:  Well let me just say that the agency through testimony submitted by Mr.
Ruckelshaus before the House in 1972,  I guess  even in 1971, gave  its ideas about  universal
no-discharge.  As I say some plants and some industries are achieving this now.  Other
places like storm sewers and for certain dissolved solids, I personally as an engineer,
don't think we are going to achieve that for God knows when if ever in many industries.
This does  not mean that the goal is not a good goal you see.  This does mean because of
our population growth as the  Commission reports  and because of the industrial growth,
we have got to get as close to that as is economically and technically feasibly possible.
And for a  lot of industries and a lot of plants this means from going from virtually unlimited
discharge  to something that is quite within the realm of possibility and  is being practiced
today.  We have not suggested any technologies that those of us who have been working
in this field for a long time have not worked with for a long time.

Morris Wiley: I believe Mr.  Cywin that you stated that industry was discharging  about
three times as much waste as  municipalities if  I understood your statement.  The National
Commission on Water Quality indicated that the industrial water intake was a little more
than half of the municipal  waste discharges and the information that I would  have would
suggest that say in the oil and grease pollutant parameter  that municipalities are  larger
sources than industry and probably true in BOD as  well.  What  is  the basis for saying
that industry,  apart from agriculture, discharges more pollutants and where can one find
this  information so that it can be examined by industry.

Allen Cywin:  The basis of my information is the 1968 Department of Commerce report
(and my three-times statement was in terms of  BOD,-)-  If you will just  think  a large pulp
and paper plant introduces in one plant as much  BOD,- as  the population of a million
people. A cow introduces ten times more BO D^. than a person and so forth.   I am inc-
luding feed lots as they are very large  industrialoperations.  The  processing  of animals
(you can estimate in  terms of petroleum refining) can get 100,000 chickens in one place
you can get 10,000 cattle in one place and that's a large point source of discharge.  I
was looking at a film this morning just  before I came down here on the  Today Show for
just a second and  it was about California dairy operations. Everything was automated,
cows came into a  line, they got washed down, the machines were automatically  applied,
the machines were automatically taken off, they milk a cow in  eight minutes - they just
run them through this machine, there, wo-san awful lot of cows in  that place and  a lot
of residual material.

Anonymous:  Where can you find population equivalents for livestock?

-------
 26
 Allen Cywin:  For that one number look at the 1958 census Department of Commerce.
 As far as the cattle and what not we have published some papers which give population
 equivalents and so has The Economic Commission  for Europe of the U.N., but we are
 talking about concentrated animal feeding operations.  The environmental protection
 agency was taken to court once again to include  every  cow and  chicken not every farm,
 but we  have established population equivalent limits based upon research, you know
 minimum limits, size  limits and we have also established the fact that these have to be in
 concentrated areas.

 J. Anthony Tallis:  During your presentation you  mentioned toxicity analysis  is going to
 Fe factored into the guidelines in  terms of cost.   How is this a part of the policy of deter-
 mination as to  limitations?  Maybe I can quote an example:-

        as  to whether a dam should be built in a certain location is useful in comparison
        with alternative source of water supply, water supply being an absolute necessity
        for industry and municipalities.

 In something like this, what is the break-off point?

 Allen Cywin:  Let me answer in two ways, first of all  the term is cost versus benefits of
 effluent reduction.  In other words, do you double the cost by taking out 2% more or 5%
 more of BOD, or of suspended solids,  NUS was one of our contractors and they established
 a good series of graphs; for example, the iron and steel  industry which showed what the
 levels of technology would be, what their costs were and how much tonnage on a model
 plant basis which would apply across the industry would delete from discharge to the water
 environment. As a part of this analysis, we had a separate series of contractors  look at
 the economics of the Industry as a whole.  Now where are looking at the industry across
 the board with reference to  what technology can  achieve and in some parts of the industry
 you will find the smallest plants achieving let's say the  best technology.  Normally you
 will find in economics that the smallest  plants are the most marginal and in many cases
 you will find in our regulations size cutoffs for example, where there is no or a lesser
 regulation  because we don't feel that that segment of the industry should be regulated on
 a technology basis because the cost or the total economics across that spectrum of plants
 is  prohibitive.  Or you may find some differences by geography as for example in the
 Alaskan crab operations,we  have a different limitation than the contiguous states.  We
 have a number of variabilities throughout our regulations on this basis.  I don't know  the
 exact numbers for example and the history behind them but obviously that Mr. Beychok
 was showing and calling anomalies but I would guess that the average of the best practices
 in  those sub-categories created those anomalies,  since certain sub-categories especially
 more complex you perhaps have more sophisticated operations or designs and therefore are
 taking out  more pollutants and you should because you have more in there than just BOD,-
and suspended solids.   We have not identified the toxics at this stage but the more
sophisticated chemical operations, petrochemical operations, are eventually going to have
 to  address themselves to this point of toxics.

 W. L. Ruggles:  What is the future of toxicity in  the guidelines?

      '. w:n:  Amongst our own staff you will find different attitudes and different

-------
                                                                                27
statements.  Toxics obviously are something that can get quite emotional.  You see in
Kepone for example, I  think most of you have seen enough on TV recently of Kepone,
and the emotions that this has generated.  One of the reasons why EPA has not been
able to  come out under 307a with some real hard numbers as yet is because of the great
divergence of opinion as to I) what you can achieve and 2) the obvious economic costs
that are involved.  You run to two viewpoints,  if it's my health that is in danger I don't
care what the cost is, if it's your health I  might not care either,  but you most certainly
would and I am sure that this kind of issue is going to be debated not only for water
pollution but it is being debated for air pollution and OSHA and  other regulations.  It's
a very hard  thing to cope with because what is toxic to some people in toxic amounts or
to the aquatic life isn't something we  can  prove quite easily across the board.  There is
a point of view that let's play it safe until we find out some more about it.  Industry on
the other hand obviously says well prove it and  then we will do something about it.  You
get these two different points of view  which  EPA is trying to balance all  the time.  So
let me say I think we will bring forth a balanced prospective when we finally get the
thing out, I hope so.

-------
 28
BIOGRAPHY
       Allen Cywin was appointed Science Advisor
for the Environmental Protection Agency's Office of
Water and Hazardous Materials in February  1976.
Prior to this, Mr. Cywin organized and served as
director of the  Effluent Guidelines  Division in the
Office of Water and Hazardous Materials.
       Mr. Cywin also  served as Director,  Division
of Applied Science and  Technology in the Water
Pollution Control Research and  Development Office,
Assistant Commissioner for Operations and Engineering
of the Community Facilities Administration, and
Chief of the Saline  Water Conversion Demonstration
Plant  Program of the Department of the  Interior
where he was responsible for directing the design,  con-
struction and operation of the Nation's first large saline
water conversion plants.
       Mr. Cywin has also  worked for the Resarch
Division of the Navt Facilities  Engineering  Command and
is  retired Seabee Reserve Officer.

-------
  "THE ROLE OF THE EFFLUENT STANDARDS AND WATER QUALITY INFORMATION
   ADVISORY COMMITTEE (ES&WQIAC) IN FURTHERING THE SCIENTIFIC AND
               TECHNICAL DEVELOPMENTS TOWARD CLEAN WATER"

                                   Martha Sager
                Chairman, ES&WQIAC  EPA7 Washington, D.C.

STATUTORY BASIS

      The Effluent Standards and Water Quality Information Advisory Committee (ES&WQ-
IAC) is a statutory Committee established under Section 515 of PL 92-500.

      "Sec.  515 (a) (1) There is established on Effluent Standards and   Establishment
      Water Quality Information Advisory Committee,  which shall be
      composed of a Chairman and eight members who  shall be appoint-
      ed by the Administrator within sixty days after the date of enact-
      ment of this Act.
      "(2) All members of the Committee shall be selected from the    Membership
      scientific community, qualified by  education, training, and
      experience to provide, assess, and  evaluate scientific and
      technical information on  effluent standards and limitations.
      "(3) Members of the Committee shall serve for a term of four     Term
      years.

MEMBERS AND STAFF SUPPORT

      PL 92-500, the "Federal  Water Pollution Control Act Amendment", was enacted in
October  1972.   The ES&WQIAC  members were appointed on  February 3, 1973 for four
year terms -  which  may be renewed.  The committee is composed of a chairman and
eight members.   The staff consists of an Executive Director with administrative and
secretarial support.

ACTIVITIES

      The data bases and  information relative to industrial point source effluent discharges
are of such a particular technical nature that from its earliest meeting in March 1973,
members of ES&WQIAC decided to work closely with the technical engineers and project
officers  in the Effluent Guidelines of the  Office of Water Planning of EPA so that differing
engineering viewpoints might be settled at an early stage in guidelines development to
cut down the Administrators involvement in issues of such technical  nature at the time of
promulgation.

      This relationship was developed in concert with the director of the Effluent Guide-
lines Division who arranged the first joint meeting with project officers of  EPA early in
April, 1973.

      The Committee has  carried out its statutory functions through public  hearings; public
                                       29

-------
 30

meetings; industrial workshops; critique and review of documents; special studies; and
agency, industry, congressional; and public interest group liaison.

FIRST YEAR ACTIVITIES, MARCH 2, 1973 - MARCH 1, 1974

      In its first year of effort the ES&WQIAC conducted 60 Industrial Workshops, 7 Public
Hearings, and 15  Committee Meetings.  It provided detailed comments on an extensive
series of technical documents and proposed effluent limitations related to the initial 27
basic industries specified in PL 92-500, and proposed a new approach for developing effluent
limitations for industrial point source discharges  became  later known  as the Matrix Method.

                                 MATRIX APPROACH
                               (BNA News Article 1973)

      "This new matrix approach for  establishing effluent limitations under the Federal Water
Pollution Control was developed by  ES&WQIAC and presented to EPA Administrator  Russell
E. Train September 26, 1973.

      The new method entails a modification of the guidelines based on state-of-the-art waste
water treatment technology.  In addition to current subcategorization of industries, considera-
tions which affect an individual industrial operation are  included in mathematical models used
to establish as best practicable technology.   The current data base available from EPA and
industry serves as a baseline for establishing raw waste  load.   The baseline is modified as
additional data are generated and as predictive models of treatment technology are improved.

      The subcategorization of industries considers four constraints on effluent standards of
performance, these include treatment technology, operational characteristics, economic equity,
and  geographic and climatic considerations."

      The matrix provides a quantitative, technically sound basis for  subcategorization of
industrial categories which  is necessary to insure that effluent limitations are consistent with
the requirements of the act  and  realistically achievable by  classes and categories of point
sources.  This method for quantitatively determining effluent limitations could be interpreted
as a National Guidelines.

      Basic steps for development of  an effluent standards model by the matrix method  are
outlined, as follows:  (1) Develop statistical plots of raw waste load and flow  for an industrial
category.   Determine if size, age,  or type of process has a major effect and if so break the
statistical plots accordingly.   Eliminate plants which have extensive water reuse or product
recovery not normally practiced in the industry.   Base lineflow/unit production is one stand-
are deviation from the mean.  (2) Determine from waste water treatment operating data on
best practicable technology if the waste water characteristics have an effect on process per-
formance under optional design  operating conditions.   If so, categories should be established
with appropriate effluent qualities.   (3)  Define operation  characteristics of  the industry.
Establish effluent qualities for each grouping.  (4) Delineate geographical and climatic
effects.   Best practical technology  in some cases can be defined by area if spray irrigation
or evaporation ponds  are applicable.   However, in all cases an alternative for discharge

-------
                                                                                   31

using waste water treatment should be considered because land availability or cost could
mitigate against land disposal, waste water characteristics,  particularly temperature,
should be considered.   (5)  For each of the groupings defined in  (1), the cost of compli-
ance/unit production should be generated.   Best practicable technology is defined and
costs are computed by selecting the large plants with prevalent technology.   Processes are
selected and effluent quality defined for each grouping considering a reasonable cost equity.
(6) As additional data is complied,  it would be added to the data base for current best
practicable technology.   Treatment models would be updated as  the state-of-the-art ad-
vances.

      Industry support for this quantitative method was strong and was evidenced by several
hundred  letters and telegrams  to the  Administrator of EPA requesting its evaluation by the
Agency.

      In December,  1975, the matrix method was  discussed by the EPA's R&D division at
Ada, Oklahoma for review and revision,  if necessary.

      In addition the Committee developed data formats for information acquisition from
industry and  received and transmitted more than 320 communications to EPA regarding
establishment of effluent limitations.  Under its responsibilities for Section 307, i .e., on
toxic substances the Committee: conducted special hearings; developed extensive lists of
materials proposed for inclusion in the original  toxic substances list by EPA; prepared a
comprehensive bibliography of reference material on toxic substances and provided a posi-
tion paper for EPA on toxic substances regulations.  (Details are  provided in the Commit-
tee's First Annual Report covering the period March  2,  1973 to March  1, 1974).

SECOND YEAR ACTIVITIES, MARCH 2,  1974- MARCH  1, 1975

      In the  Committee's second year of operation a comprehensive Work Plan was developed
covering 10 key areas of Committee  interest in  providing, assessing and evaluating scienti-
fic and technical information for the Administrator of EPA,  according to its mandates.
This FY'75 Work Program is shown in Table 1.

      The Committee held 5 public meetings, and 40 industrial workshops to obtain techni-
cal information for assisting EPA in developing  industrial effluent limitations.   The Commi-
ttee also provided comments and recommendations on 56 documents and rule-making pack-
ages developed by EPA for establishing industrial effluent limitations.

      In April 1974, at the request of the House Committee on Public Works,  the Committee
provided comments on "Problems in Implementation of the Federal Water Pollution Control
Act".   The summary of this testimony follows:

      The ES&WQIAC Members have been concerned with the  following critical issues
with regard to the effluent limitations for industrial point source discharges being develop-
ed by EPA under Sections  304 (b), 306, and 307 of PL 92-500; specifically:

      (1)  EPA's interpretations of definitive terminology and the effects of these interpre-

-------
32

      tations on the promulgated limitations, i. e., Exemplary Plants;  National Standard;
      Guideline; Effluent Limitation; BPT; and,  Zero Pollutant Discharge.

      (2)  The effectiveness of the  methodology selected by the agency to develop industrial
      point source effluent limitations; i.e., selection of a descriptive and qualitative
      contractor's approach rather  than a quantitative mathematical methodology.

      (3)  Specific items which resulted because of the definitions adopted by the agency
      amd the application of the methodology selected:

          (a) % BOD reduction required of BPT by 1977;
          (b) Factors excluded generally in establishing subscategories in many of the
              promulgated regulations;
          (c) Economic equities largely imbalanced in favor of large  "most exemplary
              plants";
          (d) Lack of quantitative economic analysis of non-water impacts and energy
              requirements in the  promulgated  documents  to date;
          (e) Lack of economic and scientific  substantiation with regard for Section
               307 and toxic materials effluent discharges; and,
          (f)  Lack of cooperation by the Effluent Guidelines  Division  with the
               Committee efforts to provide, assess and evaluate scientific and
               technical information for the administrator of EPA according to  the
               mandate for the ES&WQIAC in Section 515 of  PL 92-500.

      Special information gathering conferences were planned or conducted to facilitate
exchange of technical information.  In November 1974 the Committee sponsored a Federal
Interagency Conference to  determine the quantity and quality  of technical information
derived by EPA in establishing industrial effluent  limitations.   As a result of this confer-
ence, recommendations were made to the EPA Administrator for improvement in utilization
of these government resources.

      A special  task  force was constituted in December 1974 to review and provide recom-
mendations to EPA on a newly proposed approach  to developing Toxic Substance limitations.
A report "ES&WQIAC Report on Review of Draft Advance Notice of Proposed Rule-Making
on Toxic Substances"  containing ES&WQIAC recommendations was subsequently transmitted
to EPA.

THIRD YEAR ACTIVITIES, MARCH 2, 1975- DECEMBER 31,  1975

      During the Committee's third year of  operation the Committee continued its activities
of meetings,  workshops, special task forces and reviews.

      On March 6 a  Hearing was conducted on  "Privately Owned Sewage Treatment Works"
including public interest groups, industry and Federal agency representatives.

      A conference on radioactive  wastes was conducted on April 3 with participation of
all Federal Agencies with responsibilities in the area.

      The Committee  continued its detailed  review and comments on all EPA documents

-------
                                                                                  33


related to industrial effluent standards development and the associated rulemaking packages
and evaluated 20 of these during this time period.

      Public Meetings were conducted addressing industry problems of meeting Best Avail-
able Technology (BAT) specifications for 1983.   The schedule for these meetings with
milestones is shown in Table 2.   The first meeting on  BAT was held on April 24, followed
by workshop/meetings in July, August,  and September.

      Four special studies were initiated by the Committee to provide additional  informa-
tion to EPA during this third year of operation:  (1)  Development of a Report on the Matrix
Method for Establishing Industrial Effluent Limitations;  (2) Development of an Approach
for Establishing Best Available Technology (BAT) Under PL 92-500; (3)  Analysis of  Litiga-
tion on Implementation of PL 92-500; and (4) Analysis of Toxic Substances Legislation.

      Five reports were prepared for publication in calendar year 1975:

(1)  Second Annual Report - ES&WQIAC
(2)  The Matrix Method - A Methodology to Assist in Establishing Industrial  Effluent
    Limitations under Public Law 92-500
(3)  Summary of Contentions of Industry  in Litigation Pending Under Sections 304 (b) and
    306 of Public  Law 92-500
(4)  An Analysis of Current and Proposed Federal Legislation  Seeking to  Control the Use
    of Toxic Materials
(5)  An Approach for Establishing Best Available Technology  (BAT) Under Public Law
   92-500 - With Applications to  the Organics, Synthetics  and  Plastics Industry

      A new ES&WQIAC Toxic Substances Task Force was established in November 1975
to assist EPA in implementation of Section 307 of PL 92-500.  A proposed rationale for
implementation was developed.  A workshop/meeting was also developed and scheduled
for January 15 and 16,  1967 to review the EPA plans and ES&WQIAC proposal for imple-
mentation of toxic substances standards under PL 92-500.

CLOSURE

      The ES&WQIAC is pleased with its output during its years of tenure since March
1973.  We believe the  evaluative techniques,  methods of operation and  range of inputs
are uniquely different from the typical advisory committee pattern of operation and could
possibly become  a  template for future operation of scientific  and technical advisory
committees in the Federal arena.

-------
 34
DISCUSSION

W. L. Ruggles:  Will toxic-materials restrictions generally be even more severe than the
present 1977/83 guidelines?

Martha Sager:  The toxic limitations conceptually should be more stringent because of their
direct relationship to the physiology of aquatic organisms and because of harmful effects in
minute quantities.

Irvine W. Wei:  Please describe the matrix approach.

Martha Sager:  It is a  little difficult to discuss the matrix method  in a brief time.  We are
preparing a comphensive report on the "Matrix Method" which will be available shortly.
Briefly the method utilizes a series of matrices in which the first matrix characterizes the
individual inputs and leads to the definition of the standard raw waste load for  each
category or sub category.  Then these  results are input onto a second matrix which yields
process designs for alternative choices as a function of the effluent quality.  In turn these
are then applied  to the economic matrix which generates a series  of costs versus effluent
quality relationships for the alternatives from  the second matrix.  Through consideration of
cost versus effluent quality policy decisions are next made in establishing the ultimate
effluent standards for each sub category.  This procedure is shown in Figure 1.

BIOGRAPHY
     Dr.  Sager is Chairman of the  Effluent
Standards and Water Quality Information
Advisory Committee (ES&WQIAC), a
statutory committee established under
PL 92-500.  She  is also Professor and
Director  of the Environmental Systems
Management Program within the Center
for Technology and Administration,
College of Public Affairs, at the
American University, Washington, D.C.;
Consultant in  Environmental Management;
Member, Air Quality Evaluation Panel;
Technical Advisory Board,  Department of
Commerce; Panelist, Measures of Air
Quality, National Bureau of Standards;
and a Specialist in Environmental
Engineering and Industrial Micobiology.

-------
                                                                                    35
                TABLE 1 "ES&WQIAC COMMITTEE WORK PLAN  (FY75)"
      It has been proposed that the following tasks and work plans be adopted for ES&WQIAC
 for the new fiscal year in order to accomplish the maximum production of end product in
 relation to providing, assessing and evaluating scientific and technical information for
 transmission  to the Administrator of EPA.

 I.    Document Review
      Overall  engineering responsibilities for review of documents from Effluent Guidelines
      Division.
      Don Bloodgood (coordinator)                    Ramon Guzman
      Wes Eckenfelder                               Lloyd Smith

 II.    Early Inputs for Current Effluent Limitation  Development
      Early ES&WQIAC guidance on approach and document preparation for the development
      of effluent limitations.  This will include workshops and special tasks for at least the
      following industries:
      Miscellaneous Foods and Beverages              Ore Mining  and Dressing
      Machinery and Mechanical                     Water Supply
                                     (Others)
      All Committee Members (Brossmar - Coordinate, development)

III.    Anticipating Inputs for Future Effluent Limitation Development
      Ongoing review for revision and  updating of effluent limitations.  Data input design
      and acquisition.  Development of a broader base of data sources.  Possible Matrix
      Workshops.
      Glenn Paulson                                 Martha Sager
      Wes Eckenfelder                               Martin Brossman   (utilize additional
      Blair Bower                                    technical  staff support when
                                                    available)

IV.    Public Hearings for Public Interest  Groups
      Selected series of meetings involving public interest groups  in relation to the possible
      economic implications and environmental  impact on the public of PL 92-500.   This
      task will have particular reference  to  ES&WQIAC responsibilities in order to advise
      the Administrator  about public reception and response to the progress EPA is making
      on PL 92-500 in the area of industrial  point source discharge.
      Glenn Paulson                                 Martin Brossman
      Bob McCall

V-    Federal  Liaison
      Liaison with other federal agencies on positions and problems  relating to PL  92-500
      and preparation of reports for transmittal to the Administrator.
      Martha Sager                   Bob  McCall             Martin Brossman

-------
    36
  VI.   Special Legal Tasks
       Coverage and preparation of written reports to ES&WQIAC on hearings and legal actions
       related to designated industries and toxic materials.  Maintenance of liaison with the
       General Counsel's Office in EPA and other federal agencies as required.
       Blair Bower                     Rose Mattingley           Bob Grieves

 VII.   Matrix Development
       Coordination of development and implementation of matrix method by EPA.
       Martha Sager                                  Bob Grieves
       Wes Eckenfelder                                Martin Brossman

VIII.   Coordination with National Academy of Science
       Coordination of National  Academy of Science review of EPA in relation to impact on
       ES&WQIAC responsibilities (with special reference to the  inclusion of the matrix in
       the EPA R&D Program).
       Glenn Paulson                                  Lloyd Smith
       Martha Sager                                  Martin Brossman

  IX.   Economic Analysis and Liaison with National Commission on Water Quality
       Expansion of ES&WQIAC economic analysis inputs and coordination with NCWQ.
       Blair Bower                                    Martha Sager
       Wes Eckenfelder                                Martin Brossman

  X.   Toxic Effluents
       Review of effluent limitations, analysis of legal and  technical aspects in liaison with
       other federal agencies on  Section 307 of PL 92-500.
       Lloyd Smith                     Martha Sager              Glenn  Paulson
                                                                 (participation dependent
                                                                 on current court actions)

-------
                                                                             37
                      TABLE 2 "BAT TASK FORCE SCHEDULE"
                             (Revised July 22,  1975)
                        (Organ!cs,  et al, and Paperboard)
ES&WQIAC MEETING DATE  LOCATION           MILESTONE
                                    MEETING ACTION
July 18 & 19
    (Fri./Sat.)
Aug. 18 & 19
    (Mon./Tue.)

Sept. 25 & 26
    (Thurs./Fri.)
Oct. 30 (Thurs.)
Univ. of Kentucky     Completion of   Review Data;
Lexington, KY        Data Collection  Decide on Methods
                                    of Analysis
Vanderbilt Univ.
Nashville, TN

ES&WQIAC
Wash., D. C.
ES&WQIAC
Wash., D. C.
Completion of   Develop Report
Initial Analysis  Outline
Completion of
Draft Report
Final Action
on Report
Review Draft
Report, Revise,
Produce Final
Version

Agreement on
Disposition of
Final Report

-------
38


OPERATING DATA STATISTICAL
FROM EXISTING CHARACTERIZATION OF RWL DATA
TREATMENT FACILITIES FROM ALL PLANTS IN INDUSTRY
• •
•^^ ^P*
RWL
PARA METE
FLOW
BOD
COD
SS


LJ
V.
<
$
Q

PROCESS PRODUCT CLIMATIC
R AGE SIZE TECHNOLOGY SPEC EFFECTS


1
VERIFICATION

OF PROCESS
PARAMETERS




<•
OTHER


SU8CATEGORIZATION
SELECTION OF SRWL



IN- PLANT
MODIFICATIONS ""


SELECTION OF
BPT
(BAT) ALTERNATIVES
t


SELECTION OF PROCESS DESIGN PARAMETERS
A.S.
F/U
A.L. T. F.
V

Sl/t Oh UNII S ^ ,
L.D.
A. R.
A.C. OTHER
L





"
DEF NITION OF EFFLUENT QUALITY (Sc )
S
BOD
COD
SS
OTHER
A.S.

A.L. T.F.


L. 0.

A.C. OTHER



'•



COST MODELS

OPERATING
CAPITAL

A.S.

A.L. T. F.

<
L.D.

A.C. OTHER


EFFLUENT QUALITY -COST
RELATIONSHIPS FOR
BPT (BAT) ALTERNATIVES


i
ECONOMIC
IMPACT









S DESIGN
in
LJ
0
O

-------
   "THE ROLE OF THE NATIONAL COMMISSION ON WATER QUALITY (NCWQ)"

                               Joe G. Moore, Jr.
             (formerly, Program Director, NCWQ,  Washington,  D.C.)
                 Head, Graduate Program in  Environmental Science
                         The University of Texas at Dallas

    Sometimes when I see what is going on today I wonder if the "good old days" were
as good as we sometimes want to remember.  My first experience with Bill Galegar had
to do with the adoption of the Texas water quality standards. He had the proper federal
view that nobody at the state level knew anything and  I had the proper state view that
we weren't going to have these "outlanders" telling us  what we ought to do in our state.
In one  of the early conversations with him I said, "I want you to know that we don't like
these foreigners coming down here telling  us what to  do."  Bill sort of drew himself up and
got a scowl on his face and said, "I want you  to know that my forefathers welcomed your
forefathers  to this country when they came;" that did give me pause,  I asked, "What do
you mean by that?", and he said,"Well, you know I  come from Oklahoma, you know
Oklahoma was settled by the Indians.  I happen to belong to a tribe of  Indians who  have
not yet executed a treaty with the United  States government."

    How many of you know what the National Commission on Water Quality is?  Well,
may be I had better tell you something about this operation, with apologies to those of
you who do know. The  National Commission on Water Quality, created by Section 315
of Public Law 92-500 is made up of five members of the Senate, five members of the
U.S. House of Representatives and five Presidential appointees.  The Commission was
created in the law with a statutory deadline to make a  report to the  Congress on or before
October 18, 1975. As usual  with government bureaucrats, we were two million dollars
short and six months late, if the Commission manages to make its report in March of 1976.
The Commission members — and by the way, of the Congressional  members you will
recognize  a large number of names involved in the development of Public Law 92-500 —
are as  follows:  the presidential appointees, Nelson A. Rockefeller, who was then
governor of the State of New York and now Vice-President, has remained chairman
throughout  the. life of the Commission; Dr.  Edwin A.  Gee,  senior vice president of DuPont
in the Wilmington, Delaware office; William  R. Gianelli, a consulting engineer from
Pebble Beach, California who is probably better known as the Director  of the California
Department of Water Resources, the agency responsible for developing the California
water project;  Mr. Raymond  Kuduleis, the Director to the Department of Public  Utilities
for the City of Cleveland, Ohio and Mr. S. Ladd Davies, Director of the Arkansas
Department of Pollution Control and Ecology;  from the  Senate,  Muskie  from Maine,
Randolph from  West Virginia, Bentson from Texas, Baker from Tennessee and  Buckley from
New York;  and from the House of Representatives, Jones  from Alabama, Jim Wright from
Texas,  Harold  R.  "Bizz" Johnson from California, Bill  Harsha from Ohio, and James C.
Cleveland from New Hampshire.   There were three additional Commissioners, John A.
Blatnik, a former Congressman from  Minnesota, James R. Grover, a former Congressman
from  New York and Carl E. Wright the Chairman of the Arkansas Commission on Pollution
Control and Ecology who died during the summer of 1973 and was replaced by the
Director of that particular agency.

                                      39

-------
40
     The Commission's charge in the statute Is a very narrow one, "The Commission shall
make a full and complete investigation and saidy of all of the technological aspects of
achieving and all aspects of the total economic,  social and  environmental effects of
achieving or not achieving the effluent limitations and goals set forth for 1983 in Section
301(b)(2) of this Act. "  That assignment was developed after some controversy in the
conference committee on  P.L. 92-500.  By the time  the Commission was actually operable,
nearly a year after  the passage of P.L. 92-500, attitudes even of the  congressional members
had changed, and the Commission early decided that in addition to taking a look at the
narrow requirement for 1983,  the Commission would also examine the progress being made
under the 1977 requirement to the Act and examine as well the potential impact of the
elimination of the discharge of pollutants  —  the 1985 goal of the Act. After a great deal
of discussion,the Commission,  in February of  1974, published a study design stating what
it intended to do in the time remaining to it.

     The Commission appropriated or was authorized an appropriation of $15 million
initially, has secured an additional $2 million authorization and a $2 million appropriation,
so that by  the time  it concludes its work it will have spent at least $17 million.  About
$14 million of that total will have been spent on contract research  projects defined by a
professional staff to fulfill the stated study design and represented in some 90 contractor
reports.  I  have one with  me, or at least a part of the one with me on  the petroleum
refining  industry.  This is the contractor's report that went through two drafts and has now
gone into the National Technical Information Service of the Department of Commerce.
This does not include  the Appendix which contains the data  base; but  if you  can  imagine,
some of them run to four volumes this  thick, so you can have some  idea what sort of stack
you'd get if you looked at all  the reports.  For any of you who may be interested in
finding out about any report of the Commission, or getting a list of the reports,  or getting
a copy of the draft staff report or its summary, you should write to  the  National Commission
on Water Quality,  P.O.  Box 19266,  Washington, D.C. 20036, and we will be glad to
send  you a publication list. If we don't have a particular report, we  can give you some
idea about when it  is  likely to become available.

     The contractor's report on the petroleum refining industry was  done by Engineering
Science of Texas, Inc.  You have two staff people from that concern later on in your
program  talking about various waste treatment technologies, Davis Ford and  Lial Tischler.
Rather than take up my time trying to explain to you a  technical subject from a person
not versed  in technical matters, I would suggest if you have questions you might catch
them while  they are here with regard  to the actual technical aspects of the Commission's
analysis of petroleum  refining.   Mr. Beychok, whom you have already heard, was a
consultant  to the Commission.

     There  were some  75 consultants used by the staff and an additional 30 or 40 specific-
ally mentioned in the Commission's authorizing legislation.

     The  staff itself  consisted  — the program staff — consisted of some 30-35 professionals
recruited by me  and my deputy, Jim Smith, who  came  to the Commission from the
Conservation Foundation.   That professional staff oversaw the actual execution of contracts

-------
                                                                                 41
for Hie various studies and then produced a report almost as lengthy as the one I  have
just shown you at some 900-1,000 pages; this report went through three publicly
distributed versions, and probably  three private versions that were prepared by the
staff before any were released publicly.  The staff draft report consists of several
chapters; I am going to give you the general organization of  the document.  I have with
me only the first chapter called  Issues and Findings, this is the one that contains
questions and answers conforming with  the Commission study design.  For those who wish
merely a cursory examination of the Commission's work,  this  is some 82 pages long, and
you can almost pick the subject  in which you have an interest and take a look at it in
that particular section.   If you are  more  interested in the detailed subject matter areas I
will give you the chapters so that you will know the contents of the full report.
Chapter II  is an analysis of the capabilities and cost of technology for achieving the
effluent limitations required by 1977 and 1983 as well as what remains to be done after
1983, in a  technological sense,  toward achieving the 1985 goal of the elimination of
discharge of pollutants.  This is an  assessment of the capabilities of the public and
private sectors to apply the defined effluent limitations for 1977 and 1983 as well as those
more stringent to  protect water quality.  Chapter III is an analysis of  the economic effect
of the cost of applying the necessary technologies on both a micro-economic and macro-
economic  scale as well as the social effect of these changes.  Chapter IV is a description
of present water quality and environmental water-based and related terrestrial conditions
and projections of anticipated change which may result from  implementation of the Act;
this is where Mr. Beychok got some of the information he showed in the slides a  moment
ago.  Chapter V is an identification of the effects of the Act in eleven selected  regions
in the nation.  Chapter VI is an  assessment of the public and  private response as
institutional segments finance, implement, manage and enforce the nation's water
pollution control  program.  This  Chapter concerns  itself not only with the inter-govern-
mental institutional relationships but also the response of the  various regulated segments
in terms of whether or not they can institutionally achieve the requirements of the Act.

     In addition to the more traditional aspects of municipal and industrial  technologies,
economic effects and results, there is also discussion of the requirements related  to
agriculture, particularly with regard to feed lots,  irrigation return flows and agricultural
non-point sources.

    Generally speaking, the Commission staff draft concludes that the cost of achieving
the 1977 requirements and the 1983  requirements  even in today's economic climate,  are
not insurmountable.  However, the Commission staff draft inevitably has  raised the question
as to whether or not the incremental improvement to be realized from the achievement of
the 1983 requirements, once the  1977 requirements have been met, is  worth  the probable
cost.  And  so the question is, really, do the changes in water quality, as a result of the
incremental requirements of 1983,  justify the expenditure of the  funds to achieve the 1983
requirements.  Mr. Beychok did  read for you the general conclusion of the staff  with
regard to the elimination of the discharge of pollutants.

-------
42
     Let- me just give you some figures that relate to the petroleum refining industry in
 terms of the total.  To reach the 1977 requirements for all  refineries presently in place,
 the technology assessment staff concluded that the  cost would be roughly a billion dollars.
 The economic  impact analysis  indicated that as a result of imposing these requirements
 there would be certain plants that would inevitably close or consolidations that would
 occur, and therefore the  economic impact cost would be roughly $830 million as contrasted
 with a  billion  dollar cost.  You must understand that the technology contractors were
 asked  to price the various technologies for existing facilities.  The economic impact
 analyses — and by the way, they are contained in separate volumes, so this does not
 include the economic  impact analysis — the economic  impact analysis would generally
 show lower figures because of  in those industries that may  be affected so as to cause
 plant closures, the economic impact figures are generally  lower than the technology
 cost figures.  To get that in perspective for the 1977 requirements, the total estimated
 economic  cost is roughly $37 billion, so petroleum refining would account for roughly
 one billion of the $37 billion.

     The operation and maintenance costs for those facilities once in place in petroleum
 refineries  would  be  $142 million over and above whatever operation and maintenance cost
 there is in 1975 for waste water treatment facilities.  The table you saw earlier  assumed
 that the total of all  these figures would  represent the total O & M. There is a piece
 missing, simply because  we were not always able to get the actual O & M figures for the
 base year. When you see O & M costs in Commission reports,  these are incremental over
 the base line  —  which is to say when we talk about 1977 O & M costs, we are talking
 about  the  requirements,  the O & M costs attached  to the billion dollars of expenditure
 that would be  required to get to the 1977 requirements.

     For 1983,  the technology  costs  — and unfortunately the economic analysis  did not
 analyze in detail potential closures, primarily because  most of them apparently occur in
 meeting the 1977 requirements  — the cost figures for both  technology and economic
 impact are the same for the 1983 requirements.  The total cost is roughly $1.2 billion for
 the refinery industry out of a total  industrial cost of $23 billion for industry to achieve
 the 1983 requirements.  In  just comparing these two gross figures and assuming that the
 magnitude is reasonably  valid, the cost  to achieve the  1977 requirements is greater than
 the cost to achieve  the 1983 requirements.

     In addition to these two estimates,  the Commission's staff, using a massive  macro-
 economic  model  in which the staff does  not have unlimited confidence, prepared a
 prediction as to the  cost of new source performance standards for those plants that would
 be  constructed between 1975 and 1983.  For all new plants, the estimated  cost is $20
 billion, but the staff believes that that figure is much "softer"  than the other two I have
 just given  you.  The O & M figures for the 1983 requirements would be $429 million over
 and above the  incremental  costs of 1977, and for new source performance standards,  the
 estimate was $290 million for operation and maintenance.

     I  must emphasize  that  in any tabulation you see of costs produced either in the staff
 draft reports or analyses of  the Commission's contractor  reports or staff draft reports,  it is
 inevitable that the cost figures are chosen or selected from a range.  We  did not attempt

-------
                                                                                    43
to showall  the ranges of costs that were selected or produced by the contractors; thus,
you will occasionally see a variety of figures credited to the National Commission on
Water Quality.  What this means is that someone has chosen a series of figures from the
ranges either in  the contractors reports or supporting documentation and have put together
a table of his own.  For  example,  nowhere in the Commission staff draft work will you see
complete tabulations of total costs to the nation including all three  — municipal,
industrial and agricultural  — costs,  simply because the number of variables or the range
of figures is such that it  is  very difficult to do.  for example,  you saw the figure of
$158 billion, for achieving  the treatment requirements for publicly owned treatment works.
The range from which that  figure is chosen is somewhere between about $65 billion and
$450  billion,  so that the $158 billion is a figure that the staff chose  as one that appeared
to be reasonable in terms of the  requirements of the Act.  I  must emphasize that for publicly
owned treatment works there are six categories of eligible federal construction grant
assistance.  Depending upon which of those categories might be funded, one can account
for a  wide range of costs; for example,  will municipalities be required to  collect, contain
and treat all urban storm water run-off.   If they do, and they have  to apply secondary
treatment to that volume, you can regard that cost as one that contributed to that extreme
figure of some $450 billion. I caution you that when using  cost figures, it is difficult at
best to get a series that may approach what you judge to be  realistic, but you will see
figures that  vary considerably.

   The important thing to remember is that the  cost for achieving the publicly owned
treatment works requirements of the Act are considerably higher if all categories are
funded and are considerably higher than those for achieving the industrial requirements.
At the same time it is obvious that industries  have  moved substantially farther  towards
achieving the 1977 requirements by the  deadline in the  statute than have municipalities,
because the municipal program depends substantially upon the level  of federal funding and
whether or not Congress and the Administration are providing the intended level of federal
participation.

   Generally  speaking,  most industries  believe that even  the costs as I have given them to
you are severely understated. There are those  who believe, however, that the costs are
overstated.  As you might expect, Al Cywin and his group in EPA argue that we have
substantially overstated the costs and therefore distorted the consideration of Public Law
92-500. There are those who believe,  particularly in meeting  the 1983 requirements,
industries will strive for  ways to minimize the costs through  changes  in production processes
and that therefore the  1983  costs have been substantially overstated, which is to say that
as the economics of waste water treatment approaches the point at which  the costs become
a significant factor in  the original investment or changes in investment, then industries
will find ways to adopt or adapt technologies that  will be less expensive than those that
have  been analyzed by the  Commission.

   One of the issues that confronts anyone examining this subject is  the question  of the
variability of  the effluent limitations, and you have already heard some discussion of this
topic  this morning. The  Commission generally  concluded  that in the petroleum refining
industry EPA's effluent limitations, as presently stated, will preclude 100% compliance
by all industries in which the technology specified may have been installed on a  uniform

-------
44

 basis.  This then raises the question,  must the investment be doubled in order to ensure
 stand-by equipment for every piece of the wastewater treatment process?  Of course,  it
 is in examining these kinds of questions that disputes about cost arise.

      Now without getting into the detail, let  me just go to some general conclusions,  then
 I am going to talk about what is likely to happen based upon Joe Moore's  prognostication,
 since the Commission has not yet adopted any recommendations and  now there may be
 serious questions as to whether the  Commission itself will be able to agree  on what should
 be recommended to the Congress.   First of all I think it came as something of a surprise to
 those of us, at least in the Commission staff, that there is a greater  cost for reaching the
 1977 requirements and a greater water quality improvement from reaching the 1977 require-
 ments than is estimated for the 1983 requirements and water quality  changes. Recognizing
 that  part of the difference may be accounted for  in the methodology for making  projections,
 I think it nevertheless came as a surprise to  us that the greater step — according to our
 analyses — the greater step is to achieve 1977, and the smaller step is that to incrementally
 go to the 1983 requirements provided, of course, you ignore for the moment  the new source
 performance standards which must be  met over the entire period from 1975  until  1983.

      You heard mention earlier of the four parts per milljon  minimum dissolved oxygen
 level.  Remember that that particular level  is a minimum during sustained  low flow con-
 ditions.  Obviously,  if that level of dissolved oxygen persisted in some places in the
 country for long  periods of time, it is conceivable  the biological integrity of that body of
 water could be substantially  impaired.  There has been some argument by environmenta-
 lists  that we should not have chosen four parts per million but should have  chosen six,
 which I presume  would then suggest that we could compromise on five as a logical measure
 for dissolved oxygen.  I mention that because there is in the basis for the Commission's
 analyses, grounds for differences of view, and so the Commission's results, I  am sure,  will
 undergo considerable discussion in  the months ahead.

      The Commission has just concluded five public hearings, the last one of which was
 held  in Washington on last Monday and Tuesday.  The general  conclusion  is  that there was
 substantial support for the genera! thrust of  the staff draft report with a great deal of
 quarrel about individual pieces.  Each industry has generally argued that its  costs have
 been understated and therefore the  total  should rise; environmentalists and public interest
 groups  have argued that there should  be no  relaxation in the deadlines in the statute.

      You heard some discussion this morning about the elimination of the discharge of
 pollutants; the staff early decided to  us the  "net" definition, that is, there should be no
 net increase in the constituents of the water from the intake to the discharge.  You must
 also appreciate that the staff was only allowed to examine the  elimination of discharge
 of pollutants after a great^deal of Commission disagreement; finally  there was a  trade-off.
 Those who had objected to looking  at anything but 1983 were  concerned because of the
 delay in the publicly-owned treatment-works  construction grant program and they agreed
 that the Commission could look at the elimination of the discharge of pollutants provided
 the Commission would also look at progress  being made toward  the 1977 requirement,
 particularly for publicly-owned treatment-works.   While they did take a  look at EOD
 (elimination of discharge), the Commission  staff's effort was limited when  compared with

-------
                                                                                  45
the effort that went into  looking at the 1983 requirements. Obviously, in the
elimination of discharge of pol lutants, one of the major concerns is what do  you do with
the municipal plants — the publicly-owned treatment-works — and obviously the question
of dissolved inorganic salts becomes critical if EOD is to apply to municipal facilities.
Strange as it may sound, most people  who talk about elimination of discharge want to
talk about it just in terms of industrial wastes and not necessarily in terms of the total
waste stream from all point sources.  Also, it is obvious from the Commission studies,
that in some cases,  not only will  the point source controls be obsajred by the contribution
from non-point sources even with the  1983 controls in effect, but also if all dischargers
also went to the elimination of discharge, the results from this step would in some cases be
obscured by the contribution  from non-point sources.  The real question is one of balance
between the  1977 requirement, the 1983 requirements, the elimination of discharge and
what must be done for non-point  sources where they now appear  to be the major source of
pollutants to the nation's waters.

     Some discussion about the  ESQUIAC Committees matrix approach might be appropriate,
I  must emphasize something about the way this Commission functioned.  The question was
presented early, should the Commission examine some other scheme for achieving national
water quality objectives other than that contained in Public Law 92-500? As  you should
gather from the  listing of the membership of the Commission itself, a majority was involved
in the formulation and adoption of Public Law 92-500 and therefore they had already con-
cluded that the structure of 92-500 was the preferable one for achieving water quality
goals.   For example, there was the question, should the  Commission examine  the prior
program of water quality standards to  achieve water quality objectives?  Of course, the
Congress believes it rejected that approach to the  national water pollution control program
in the  1972 Act and, furthermore,  most of them that speak on the question will say that
they became convinced that water quality standards were not achieving national water
quality objectives.   In other  words, they tend to believe that water quality standards is an
approach  to a national pollution  control program has been discredited.  I say they believe
that whether or not the arguments upon which they believe it are valid.

     Thus  in looking at the program, the staff was to some degree restricted in  what it
could examine.

    Generally speaking, the technology contractors and the economic impact analysis
contractors were asked to look at five levels of pollutant control;  (1) BPT, "best practicable
control technology  currently available, " generally as defined by the Environmental
Protection Agency;  (2) BAT,  "best available technology  economically achievable,"
generally as defined by EPA;  and (3) the " elimination of the discharge of pollutants, " as
used in the definition I gave  you earlier.  There were two other  levels for which we
attempted to  get technological  costs, one somewhere between  BPT and BAT or below BPT
and one between BAT and EOD,  so that, theoretically,  there should be an analysis of at
least five levels in  every contractor's report.  As a practical matter, the contractors
achieved  various degrees of success in their examinations. Fortunately,  in petroleum
refining,  because there  was a very good data base though, actually six levels were analyzed,
primarily  because of the question of the -- as  I understand it — of the use of activated
carbon and the possible  costs  that might be imposed upon the industry for that requirement.

-------
46
   There was an attempt to  look at various levels of treatment.  However,  the staff did
   generally not look at schemes for either achieving effluent limitations or for achieving
   national water quality objectives outside the general structure of Public Law 92-500.
   Thus we did analyze the ESQUIAC approach.

        With that,  let me go to the issues as I  see them at the moment.  First of all,  will
   there be recommendations? Despite the fact that the majority of the Commission comes
   from those who sponsored the legislation and presumably therefore endorsed section 315,
   there is some indication  that those who feel the general thrust of the staff draft report is
   inconsistent with their present or preconceptions about Public Law 92-500 believe that
   perhaps it would be advisable for the Commission to make no recommendations.  Without
   recommendations they then do not have to  contend  with any suggestion other than what is
   presently contained in 92-500.  There is a  question even now as to whether or not there
   will  be recommendations,  and it takes interesting forms.  We have a great deal of dif-
   ficulty finding an agreeable time at which  the Commission can meet, because if meetings
   can be postponed sufficiently  long and the  appropriation can be exhausted, then obviously
   the Commission cannot come to  recommendations within the appropriated funds.   The
   second question is, will  there be recommendations that  mean anything? Obviously if you
   can't stop the recommendations, then maybe the thing to do is to keep there from being
   any meaningful  recommendations.  The next question is, will there be any action in 1976?
   There are those  who believe that 92-500 was a product of a political year and the worst
   thing that could be done would be to amend it in a  political  year.  Those  who may dis-
   agree with the general thrust of the staff draft can then  argue in 1977 that the Commission
   staff's data base was in 1974,  or 1973; that a great deal of time has  elapsed and therefore
   the Commission's work is out of date —another year has passed and  so all those facts and
   figures don't really mean anything and what we should do is postpone any consideration of
   any changes until late in 1977.  Think for  a moment what that does with the deadline of
   July 1, 1977.  The argument is being made, for example, that after all, 1983 is seven years
   away, and there is plent of time for Congress to  act. My reaction is if they are as slow
   in  considering changes to 92-500 as they were in adopting it in the first place, Congress
   should have started last September in order to consider amendments to 92-500 if they are
   going to get them out of the congressional  process before the 1983 deadline.

        The next series of statements are strictly mine and you cannot attribute them to the
   Commission.  Without casting aspersions, I am going to  use the phrase that we have come
   to  use as  a result of our association with the National Academy of Sciences and  National
   Academy of Engineering  Advisory Committee.  I will credit the phrase to  Dr.  "Reds"
   Wolman,  a professor at Johns Hopkins University.  The  phrase is,  "If I were running the
   zoo," this is the way we could express our  opinions and get them credited to  us individ-
   ually.  And so,  "If I were running the zoo," I would argue that there should  be a delay
   in  the requirement of "best available technology economically achievable" for industrial
   and agricultural point source discharges and a delay in  "best practicable waste treatment
   technology over the life  of the works" for publicly-owned treatment-works beyond the
   July I, 1983 deadline, (1) provided  that in the interim, every effort is  made  adequately
   to  control toxics,  either substances or pollutants — recognizing that a large number of
   toxics are being controlled in permit conditions; (2) provided that there is an aggressive

-------
                                                                                  47
program for the adoption, modificaHon and enforcement of water quality standards
wherever achievement of the 1977 requirements will not achieve adequate levels of water
quality; (3) provided that there are stringent new source performance standards imposed
upon all new plants constructed in the future; and (4) provided that there is an evaluation
of the progress that will have been achieved by substantial compliance with  the  1977
requirements  before the course of the national water pollution control program is
irrevocably committed to BAT.

     1 wish to make clear that that does not include an abandonment of the technology
based effluent limitations in Public Law 92-500.  In my view, the achievement of either
a state or a national program requires the "marrying" or the joining of technology-based
standards with individual discharger permits.   I would like to mention that that has not
always been  the federal view.  I  was Chairman of the Texas Water Quality Board at the
time we were negotiating with old Federal Water Pollution Control Administration for
water quality standards and  I was publicly chastised in a meeting in Houston, Texas in
June of 1967 because of the Texas permit system for individual dischargers.  That system
was called by my predecessor, Mr. Jim  Quigley, the process by which one  grants "permits
to pollute. "  I couldn't wait for an opportunity,and eventually caught Jim in an  audience
at a  meeting  with  the pulp and  paper industry in  New York City, to remind him of what
can transpire in  eight short years in terms of national viewpoint and suggested that just as
the nation had decided to go with Texas in that case, they might find other ways to go
with Texas in terms of the national water pollution  control program.  I do think there  must
be a technology base and permits.

     I do have some misgivings about permits administered  from the Federal level, and I
think the experience with 92-500 has demonstrated some of the problems.  You see, as a
practical matter, the flexibility that everybody argues for the effluent  limitations for
industrial  categories very probably exists in the individual permit conditions.  Those  in
EPA  who  develop the effluent limitations don't like to admit that their associates in  the
permitting  process may not necessarily be following the effluent  limitations in every detail.
We have collected enough evidence to  indicate that the permits do not always agree with
the effluent limitations and  may not be  uniform across the  country.

     Now, "If I  were running the zoo," there should be case  by  case exceptions to the
1977 deadlines,  not only for publicly-owned  treatment-works because of the delay of
Federal construction grant funds, but also for industries as well,  because our evidence
indicates that industry will have to spend substantial sums  of  money or at least there will
have to be substantial cash flow in the next two years if they achieve BPT; therefore  there
should be a delay for industry as well as municipalities conditioned upon good faith efforts
to attempt  to meet the 1977 requirements.  Parenthetically, you should appreciate, from
the discussion already of legal issues, that in some  cases,  the challenge either to the
effluent limitations or of the permits themselves will not be concluded by July 1, 1977.
In fact, there may be some that will then have judicially-applied schedules  for achieving
the BPT requirement which could  extend as long  as  five years beyond the conclusion of the
litigation  —  which could very well run  into the 1983 deadline — just for achieving the
BPT requirement.

-------
48
     All right, "If I  were running the zoo/' I  think fhere should be priorities for the con-
 struction grant categories for publicly-owned  treatment-works with a concentration of the
 funding  where it should be.  The nation is not likely to produce the kind of money that is
 contemplated in our estimates of the costs for  publicly-owned treatment-works.

     Generally,  I believe there should  be a retention of the elimination of the discharge
 of pollutants as a goal, simply because there are some cases in which this is a technological
 possibility.  I should emphasize, in my view,  that in most of those cases  in which EPA has
 used elimination of discharge  as a part  of the  effluent limitation,  it more often  than not is
 an elimination of the discharge of water as well as pollutants  — which is to say, closed
 cycle systems.  I see no  serious problem with retention of the  elimination of discharge as
 a  goal,  retention of the  1983  interim goal —  the so called "fishable-swimmable" test,  and
 even retention of the ultimate goal of "restoring  the physical,  chemical and biological
 integrity of the nation's  waters."  I do  have a great deal of difficulty with the word
 "restore," and probably  the goal would have more meaning in the administrative sense if
 it were changed to  read  "achieve and maintain"  the physical,  chemical and biological
 integrity of the nation's  waters.

     There should be flexibility in the regulations and some delegation of authority to state
 and  local units of government, but that delegation should  be conditioned upon an accept-
 ance of  the general  philosophy of Public  Law  92-500, adequate state and local resources
 to achieve those  provisions of the statute and a post audit of performance by the Environ-
 mental  Protection Agency.  For agriculture,  there should be some special  way of control-
 ling non-point sources; there must be some consideration of the effect these have upon the
 water quality of the nation.

     I would like to  close by suggesting that we not overlook that  92-500 has built into it
 a  series of cycles.  The effluent limitations are subject  to review — conscious  review —
 at intervals of five  years.  The permits  are issued for maximum periods of five years except
 in the case of new source performance standards where the term is ten years;  but there is a
 cycle in 92-500, and  I think it might be  well  for  EPA to recognize that cycle and  make
 the necessary corrections or adjustments that should be made in effluent limitations and
 permits in  the second round.  I predict  that if there are not relaxations in the deadlines
 and some adjustment in some of the rigid  regulations  issued by EPA, the entire statute
 faces the prospect either of becoming administratively impossible — which is to say it
 would collapse by virtue of inability to administer it, or — despite what some will say —
 I think  the Act will  face serious challenges in the Congress — if not in the next two years,
 certainly the next four.  The very structure of the Act could be seriously weakened in any
 such challenge.

 DISCUSSION

 Umesh Mathur:  In the business of pursuing permits,  I think EPA policy has been
 traditionally to encourage the  states to take over as much  of this  function as possible,
 and obviously this gives  the state the responsibility of defining in  some manner the exist-
 ing water quality within the boundaries of the state.  Now I know that the State of
 Oklahoma  has declared that all the waters in  the state are "water-quality limited" as

-------
                                                                                 49
defined in the Act.  The State of Kansas which is just upstream from us on the Arkansas
has declared  that every water body in their state is  "effluent-quality limited" and I have
not seen any  determination as to whether or not either oF these two designations are in
fact valid. And to me what this means is that those of us who are in the planning business
really have no ideas as to what EPA feels is the right attitude  to take.  Certainly if you are
doing regional long-term planning for facilities as well as non-point sources in  industries
and storm water run-off, and you begin on  the basis that the whole of the  State of Oklahoma
if really filthy and the State of Kansas is extremely pure,  then I feel like  you get into
the contradictions and will have interstate  battles eventually even if all the regulations
came out in the right manner with all the right numbers.  Because what one state adds on
to the water  leaving its state will be cumulative to what the next state adds on and so when
you get to Louisiana you will never be able to do anything with  water quality because of
all the  incremental additions that have been made along the way.

Joe Moore: You have hit upon what is in the statute, a difficult dichotomy.  The construc-
tion grant program state allocations are  based  upon needs.  You  have states in the Southwest
and generally the South that do not reflect substantial needs measured against a secondary
treatment yardstick.  The reason you have the declarations that waters are water-quality
limited is to  get the needs  increased. The  State of Texas, by  the way, has done the same
thing;  most of its waters are classified as water-quality limited.  The reason is that if  it
didn't do that and set higher-requirements for publicly-owned  treatment-works construction
— if you didn't get the needs increased — it wouldn't get a very large share  in the con-
struction grant program. As you can gather from the list of names I  read,  with Bentson in
the senate and Wright in the house, both on Public Works  Committees from Texas,  the
decrease in  Texas construction grant funds wasn't very well received when this first
allocation was made under 92-500.  The question of water quality limited waters,  I think
has largely been  ignored by EPA; no judgment has really been made as  to whether or not
the designation is valid.  That is the reason I used the words "an aggressive program of
water quality standards with approval, actual  approval by EPA." To show you what has
happened,  in the President's budget message that just came out this  week — if we properly
understand the first language we saw, the construction grant program will  hereafter — if
that recommendation is adopted  — will  hereafter be limited to the achievement of second-
ary treatment only.  There will be no advantage to a state declaring its waters "water-
quality limited".  You see, I think the  "water-quality limited" designation is to get more
construction  grant funds, but affects every  other activity in which you  are engaged.
Actually, "if I were running the zoo,"  I would recommend that the  construction grant
program be terminated on a date certain  about ten or fifteen years from now,  after which
there would be zero federal funding in order to force,  not only the Agency but also the
Congress and the States to  sort out their priorities for publicly-owned treatment-works.
You asked an earlier question about section 208; you will  find tremendous reliance among
theoreticians on the whole planning structure of Public Law 92-500.  Now I have been
involved in government a long time, and planning is the ideal solution to  budgeteers.  If
you are planning, you don't have  to act and so you can plan  for ever and  you don't have
to spend large sums of money.  The problem with the planning provision of Public Law
92-500 is they will be worthless unless there is a linkage between the planner and the
actor.   The way the thing is structured at the moment, I don't think the linkage  exists.

-------
50
 To show you how far-sighted the budgeteers can get, they have now decided that they
 will withdraw all federal funding for planning after another year .  We will  then have
 plans  that will be out of date within two years, because plans in my view are worthless
 unless they are constantly updated.  I hope  I am proved wrong, but I am afraid that there
 is a tremendous amount of dust being stirred up under the guise of 208 that may settle on
 everybody and there will be no pollution control when it is over with.

 Milton Beychok:  You talked about the fact that the report, the staff draft,  did not in any
 place put the total cost for municipal industry and I agree it didn't. You also talked or
 gave  the caveats about the ranges,  and  I agree with that.  I would  point out that we  have
 not only the Public Law 92-500 but the  Clean Air Act, because essentially the American
 public believes in  clean water and clean air as it believes in motherhood, and because  it
 believes someone else will pay for it.  Despite all of the caveats, all of the  ranges you
 might have to attach to it, the greatest  service your Committee could do to  the American
 public in this election year, is to let them  know that it might cost them $150 a month.

 Joe Moore:  I appreciate that comment.  One of the statements that  was made in the  course
 of the Washington  hearings was that for  the first time, there was the cost of a program
 stated in terms that average citizens, ordinary folks, should be able to understand; for
 example, even in this section, (by the way this chapter is available like this bound in just
 an 82 page document and  we distributed many more of these than  we did the  more detailed
 report),  it actual ly says what the price rises are expected to be.  You will always know if
 we have understated them, you can expect the price rise is going to be greater than  that.
 There are some shortages that are mentioned.  The  industries and geographical areas of the
 country  that are  likely to  be most severely  impacted, insofar as we  were able to analyze
 them, are stated; and just  to show you the kinds of things you can get into,  the analysis
 shows that generally industry will be most adversely affected in the  northeastern and central
 part of the country and that there could  very well be a shift of industry  to the South and
 Southeastern part of the country.   In these kinds of  things,  even with all the caveats, I
 agree that there  is some value  in being able to state things  so that people understand them.

-------
                  "ECONOMIC IMPACT OF WASTEWATER EFFLUENT
                      GUIDELINES FOR PETROLEUM REFINING"

                                   Robert T. Denbo
            Coordinator of Environmental Control,  Exxon Company, U.S.A.
                    Baton Rouge Refinery,  Baton Rouge, Louisiana

     This paper deals with the economic impact of the guideline requirements to the
petroleum refining industry. The major points to be covered include:
     - a summary study on costs of wastewater improvement for which a status report was
      presented to an API Division of Refining meeting in the spring of 1975.
     - a case study of the experience of a major refinery  in investment cost to achieve
      1977 guideline requirements.
     - insight  into perspective of relative impact of  discharges in terms of organic loading
      from the nation's petroleum refineries compared to discharges from other sources.
     - suggestions  for corrections to Public Law 92-500, the  Federal Water Pollution
      Control Act amendments of 1972, will be presented.
The  Environmental Protection Agency has issued the wastewater guidelines for Best
Practicable Control Technology Currently Available (BPCTCA) to be achieved by U.S.
petroleum refining by July  1,  1977.  Based on a study sponsored by the American Petroleum
Institute and completed in mid-1973, the petroleum refining industry will have invested
approximately $1 billion to meet the requirements of BPCTCA.  To arrive at this figure,
the contractor made  estimates on approximately 100 refineries and adjusted this investment
to cover the entire 247 petroleum refineries in the country.  Information available on the
status of wastewater  treatment trains  as of 1972 was  used  and equipment was added to
upgrade the treating sequences to include facilities  estimated as being required to meet
BPCTCA.  The target treating sequence included facilities shown in  Table 1 .  It is apparent
that this treating sequence  incorporates the latest in today's  technology.

     This paper will not discuss the merits and demerits of the most recent guidelines for
petroleum refining.  This is covered  in detail in other places.  Suffice it to say they are
tough.  Some refineries will have real difficulty in meeting the requirements, even with
the target treating train, if they can meet them at all. Also, the daily variability require-
ments for some  parameters are quite restrictive. The assumption of the cost study is only
that the investment provides the industry  with the BPCTCA treating sequence and not that
guidelines are met.

     By now many refineries are far enough along with their wastewater treatment programs
to provide accurate investment information based on funds already expended  or appropriated.
It is of interest to consider the  investment being made by the current largest  U.S.  refinery —
the Baton Rouge Refinery of Exxon,  Company, U.S.A. — to achieve BPCTCA.  Total
investment  at this location since 1967, will amount  to approximately $61,000,000.  The
breakdown  of the investment is  shown in Table 2.  For this particular refinery the configura-
tion of the  sewer system made it less  attractive to segregate clean,  once-through cooling
water than  to eliminate it by conversion to a recirculating system employing  cooling towers.
This conversion cost $16,000,000.

                                        51

-------
52


     In-plant source  corrections to reduce wastewater flow and pollutant load cost
$2,000,000.

     Because of high annual rainfall, stormwater segregation and detention are quite
expensive at this location.  The cost of these facilities amounts to approximately $7,000,000.

     The expansion of the sour water stripping facilities cost $1,500,000.

     Site preparation was exceptionally costly, requiring expenditure of $7,000,000.
Because of the shortage of adequate sites, it was necessary to  reclaim 15 acres previously
used for storing oily  silt removed  in past years from the separator system when silty river
water was used for once-through cooling.

     The $27,500,000 for the Wastewater Treatment Plant includes — in addition to a two-
train, activated sludge unit with  aeration basins to provide 18-hour detention — sewer
segregation, a corrugated plate interceptor, equalization, dissolved air flotation, sludge
dewatering and disposal and the associated utilities required.

     The Baton Rouge Refinery came onstream some 66 years ago. These costs illustrate
that solutions to retrofitting problems for an older refinery are expensive and are not of a
nature that is readily anticipated in a generalized study approach such as the API study.
The  investment at the Baton Rouge Refinery  to meet BPCTCA amounts to about $133,000
per 1,000 barrels of  daily crude run, or almost twice the average for the entire petroleum
industry  in the API study.  This comparison does not contradict the API study, but it does
suggest that the  API  costs are not too high.

     It has been  established that the dollar costs for BPCTCA for petroleum refining  are
significant.  Let's examine what we are getting for this money.  This can be done in terms
of investment per unit of cleanup. The $1 billion  to go to BPCTCA reduces  BOD,, by some
357,000 pounds  per day for the entire industry.  This amounts  to an investment of$2,800
per pound of BOD- removed each day.

     At this point I would like to  comment briefly on certain economic facts.  When
considering the costs of improvement of water quality, let's not forget who pays these  costs.
In the cases of municipal facilities, it  is clear to us that we do — as taxpayers.  It is  not
so clear that we also pay the costs to industry since these costs are passed on as higher
product prices to us as consumers.  The essential difference is  that it is usually easier to
require industry  to spend money than it is to increase taxes.  Costs as higher prices don't
have to be voted upon.  The most important consideration in developing regulations should
be whether the citizen is getting  his money's worth.

     It now appears that by the end of 1977, the largest portion of the U.S. petroleum
refineries will have installed facilities to meet BPCTCA.  If we decide to proceed with the
1983 requirements of Best Available Treatment  (BAT) for petroleum refining, we will then
be going after reducing BODj. at  a cost per daily pound 3-5 times that of a  daily pound of
BOD- reduced by installing facilities to achieve BPT.  And we will achieve a reduction of
only about 126,000 pounds per day of BOD-.

-------
                                                                                 53
    For perspective on the impact of petroleum refining, let's look at the total of all
other sources - industrial and municipal.  On this basis, the BOD_ from total petroleum
refining amounts to about 1% of the total.  While it is true that BDDj- may not be the best
measure of pollutant  load, it was  the only parameter for which estimates of the other
categories were available.  One would  anticipate that the relative amounts of total
organic pollutants are roughly in the same proportion as
    Some of my numbers are rough and can be verified when final data become available
from studies by EPA.  The figures are probably good enough to provide perspective.  If
all of the petroleum refiners in the U.S. would suddenly disappear and we could  get the
required  energy from sunlight, the overall impact on the nation's waterways would be
almost indiscernible.

    One other comment on the relative impact of petroleum refining. None of the
comparisons have included the impacts of pollutants in runoff from feed lots, agricultural
runoffs or from  land disposal operations.  It becomes more obvious what constitutes the
greatest potential for  cleanup.

SUGGESTIONS FOR  MID-COURSE CORRECTIONS TO PUBLIC LAW 92-500

    Now for some suggestions for mid-course corrections to Public Law 92-500.

    It is generally believed that all responsible citizens agree with  the objective of
environmental legislation if it is to protect human health and well-being.  It appears
that there is much that is right with Public Law 92-500 in working toward this end.  How-
ever,  it  is not clear that technology-based guidelines uniformly applied  are the right
answer for correcting  the pollution problems of the nation's waterways.  Suggestions are
in Table  3.

    The  assessment of pollutants in drinking water known to affect human health should be
completed, and then high priority should be set for  removing the known problem  pollutants.

    In order to avoid wasting of our technical expertise, natural resources and money,
what is done after this first step should be reevaluated.  For each contemplated regulation
questions like the following should be answered:
    -  For example, does the impact of a water regulation on land destroy the  net benefit
      anticipated by  the author?
    -  Does the regulation require consumption of excessive energy?
    -  Does the manufacture of equipment necessary to implement the regulation  cause
      excessive pollution somewhere else?
    -  Do the effects of inflation, balance of payments and possible infringements on other
      social goals destroy the desired impact?
    - Are we taking control steps in priority order? We should insist on this since we are
      paying the bill .

    These questions should be adequately answered  before proceeding in our control
program.  This  is a business-like approach.  We have the time, the  expertise and the

-------
54
 commitment to do this and do it correctly.  Improvements from BPCTCA will have bought
 the time for us.  We can't afford to take the chance of wasting the very natural resources
 we are trying to conserve.

 DISCUSSION

 Sheriff Khan:  As you know, the refinery  industry has generally installed electrostatic
 collectors to control the FCC effluents whereas Exxon is  the only company which has
 installed  a wet-gas scrubber.  Would you be able to outline the economic and the techno-
 logical benefits plus any comment on the  legal aspects?

 R.  T.  Denbo:  Yes.  We are discussing control of FCC particulate emissions to the atmosphere.
 Since  it is wet-gas scrubbing, then I guess it would qualify as a water topic.  On that basis,
 I will  consider this a  just question and we will go ahead  with it.  Let me mention the reason
 for the development in the first place.  An engineer who had been an expert in the catalytic
 cracking  area for some time was not  satisfied with the previous experience with electrostatic
 precipitators.  He felt there must be a better way to achieve the  Clean Air Act requirements
 that were implemented by the States.  Further, some States such as New Jersey required the
 removal of so-called  condensible particulates in early regulations for cat  cracking emissions.
 Steel and some other  industries had employed wet gas scrubbing successfully for years and
 the application was considered for cat cracking.  The Baytown and Baton  Rouge refineries
 of Exxon  Company, U.S.A. participated  in pilot plant work that indicated that wet-gas
 scrubbing had several advantages over electrostatic precipitators  since you could remove
 SO~ and  condensibles in this system. So, consequently an economic comparison was made
 thar indicated  that investment and an operating cost were about breakeven with electro-
 static  precipitators.   Consequently,  the major refineries  in Exxon have decided to install
 these facilities.

 BIOGRAPHY

     Robert T.  (Bob) Denbo is the coordinator of
 environmental  control at Exxon Company, U.S.A.
 Baton  Rouge Refinery.  In this capacity,  he is
 responsible  for development of long-range goals
 for air and water conservation and solid waste
 disposal problems.  Mr.  Denbo received a B.S.
 degree in chemistry from Louisiana State  Univer-
 sity in 1948, finishing the work for his degree
 after spending  four years in military service during
 World  War II.  Mr. Denbo has been with Exxon in
 Baton  Rouge since 1948.  He  is currently chairman
 of an  Exxon committee concerned with environ-
 mental control for all  Exxon refineries.  He is
 vice-chairman of the API Committee on Refinery
 Environmental  Control, a member of the Sub-
 committee of the API Division of Environmental
 Affairs Water Quality Committee and other
 industry committees.

-------
                                                                               55
                                     TABLE 1

      efficient in-plant water use practices
      segregated storm runoff and detention
      sour water stripping
      gravity separation
      equalization
      added removal of suspended oil and solids by either dissolved air flotation,
      filtration or coagulation
      activated sludge biological treatment
      postfiltration
      facilities for dewatering  oily and biological sludges and sludge disposal
          TABLE 2 "INVESTMENT REQUIRED TO ACHIEVE BPCTCA AT THE
              EXXON COMPANY, U.S.A., BATON ROUGE REFINERY"
            Project
          Purpose
Investment
    $
River Water Replacement Project    Eliminate the use of once-through  16,000,000
                                  cooling water to reduce effluent
                                  rate by about 90%
In-Plant Waste Load Reduction


Storm Water Handling


Sour Water Stripper
Reclamation of Old Oily Silt
Storage Pond by Sand-Lime
Stabilization

Wastewater Treatment Plant
(Activated Sludge)
To further reduce flow and abate    2,000,000
pollution at the source

Segregate, detain and treat        7,000,000
storm water
Remove ammonia and H~S from      1,500,000
effluent
Develop site for Wastewater        7,000,000
Treatment Plant
Provide treatment to achieve      27,500,000
BPCTCA
                                     Total
                                61,000,000

-------
56
    TABLE 3 "SUGGESTED MID-COURSE CORRECTIONS TO PUBLIC LAW 92-500"

    (Steps assume BPCTCA is implemented as currently scheduled  in mid-1977.)

    Complete the assessment of pollutants known to affect human  health.

    - Determine sources and abate.

    At that point provide an adequate period for reassessment of any additional steps.

    - For each new,  contemplated regulation evaluate:

            the impacts of that regulation on other areas of the environment — air;
            water; land

            energy consumption

            use of natural  resources to install the control facility

            impacts in  other segments of the economy

    Impose only those regulations that are determined to provide net benefits that justify
    the net costs to society,  with special consideration for the specified uses and the
    assimilative capacity of each waterway.

-------
                  SESSION II

"PERMITS FOR THE POINT-SOURCE CATEGORY"
   Chairman
   Jerry T.  Thornhill
   Regional Representative
   Office of Program Integration
   U.  S. EPA, Region VI,  Dallas, Texas
   Speakers
   Carl J. Schafer
   "The Environmental  Protection Agency's Permit System
   Robert F. Silvus
   "Current Approach to Refinery Permits"
   William  K. Lorenz
   "Petroleum Industry Experience With The NPDES Permit System1
   Harless R. Benthul
   "The Enforcement of Permits"
   F. T. Weiss
   "Measuring the Parameters  Specified in Permits"
                      57

-------
BIOGRAPHY         Jerry T. Thornhill

    Jerry T. Thornhill holds a B.S. in Geology from the
University of Texas at Austin.  Mr. Thornhill has been
with the  EPA in Dallas, Texas since 1969.  During this
time he has  served as: - Contingency Plan Officer; Chief,
Federal Activities Branch; Chief, Oil & Hazardous
Substances Branch; and Deputy and Acting Director,
Hazardous Materials Control Division  - before assuming
his present position of Regional Representative, Office
of Program Integration in 1974.  Mr. Thornhill has also
served with  the Texas Water Development Board  as Water
Quality Specialist and as Assistant Director,  Ground
Water Division; the Texas Water Commission; and the
Texas Board  of Water Engineers.
                                       58

-------
                  THE ENVIRONMENTAL PROTECTION AGENCY'S
                                   PERMIT SYSTEM

                           Carl J. Schafer,  Acting Director
                                  Permits Division
                             Office of Water Enforcement
                         U.S.  Environmental Protection Agency
                                 Washington,  D. C.

      Created by the Federal Water Pollution Control  Act Amendments of 1972 (Public
Law 92-500), the National Pollutant Discharge Elimination System (NPDES) replaces and
improves upon the old permit system under the 1899 Refuse Act.   It is part of the compre-
hensive effort set in  motion by the 1972 law to prevent, reduce, and eliminate water
pollution.

      After listening to all the excellent speakers  this morning, I  have to tell you they
make up a  hard act to follow.   I thought,  too, of all sorts of clever opening remarks as
I  sat in the audience but as I stand here, they've either vanished or have been used by
somebody else.   I would  just  like to say that it is  a pleasure to be here, and it is a
special pleasure to renew acquaintances with many of you with whom I  have worked in
years past  in setting  up some of the  initial considerations for permit limitations before
there were guidelines.  The  Permit Program is established under Section 402 of the
Federal Water Pollution Control  Act as amended in 1972 and constitutes one cornerstone
of a comprehensive program to insure that the intent of the Act is complied with by all
point source dischargers to the waters of the United States.   Under the Act, National
Pollutant Discharge  Elimination System (NPDES) permits specify effluent limitations
and schedules of achievement which involved both compliance monitoring requirements
on the part of the discharger and compliance monitoring on the part of the regulatory
agency, be it State or Federal.   Obviously, the purpose of all NPDES compliance
monitoring  is to assess whether or not the limitations are being met and, if not, whether
enforcement actions  are necessary.

      The Program is designed for ultimate delegation to each  State, and State participation
with and without delegation has  been most gratifying.   While the States retain primary
responsibility to combat water pollution, they do so within the framework of the Federal
law.   Wherever States encounter difficulties in  administrating the NPDES, the Environ-
mental Protection Agency may intervene to assist in resolving the problems,  be they legal,
technical,  or administrative.  Currently, a total of 27 States manage their own NPDES
programs, all of which are monitored constantly  by Federal EPA Water  Enforcement
personnel.   The  Environmental  Protection Agency is a highly decentralized effort in
itself as it  delegates authority to its ten regional offices, each of which exercise a high
degree of autonomy.

      I'd like to address today the actual substance and content of NPDES permits.  By
and large,  you can  break  an NPDES permit down into four general components: (1) the
limitations; (2) the compliance schedule; (3) the self-monitoring and reporting requirements;
and (4) the boiler plate.    Limitations, of course, are based upon effluent guidelines, or

                                       59

-------
  60


water quality standards in their absence, under Section 402 (a) (1) of the Act.  The indi-
vidual determination by EPA for a particular plant regarding what constitutes  Best Practicable
Treatment (BPT) is generally written in pounds, depending upon production levels at the time
the limitations  are determined.   Concentration limits are also specified depending upon
various other requirements.    Compliance schedules cite specific steps  to achieve compliance
with the Act by July 1, 1977, as well as reports of progress on the achievement of  each step
and, in particular, reports of failure to make progress.

      Each of the requirements in an NPDES permit, whether it be interim or final, is
legally enforceable.   Each permit is  for a fixed period of time not  to exceed  five years and
application  for renewal must be made 180 days prior to  the expiration date.   Most of the
major permits were issued by December 31,  1974,  and we therefore expect the next round
of permit issuances to take place during 1979 and  1980.

      I would like to give you a little bit of insight into where we stand right now, what
the status of the program  is, and what is going on.   There has been a great deal of interest
expressed regarding the  Seventh Circuit Court decision by Judge Flannery on  feedlots and
I  don't think we know yet exactly how many new point  source discharges this is going to
add to the list, but for the time being at least —  that not withstanding	we  have a total
of 59,048 identified dischargers requiring permits.   Permits have been  issued  for 43,420 of
that number, which is a 74% issuance rate.   The breakdown on  those figures indicates  that
16,000 municipal permits have been issued out of  19,000 dischargers identified and in the
non-municipal  field, which includes industrial, agricultural and  Federal facilities, we
have issued  just under 26,000  permits out of the approximately 40,000 dischargers that  have
been identified.

      Out of the 59,000  permits to be issued,  of which EPA has issued 43,000, you might
be interested in what our adjudicatory hearing experience has been.  To date there have
been 1,863  adjudicatory  hearings requested.  In  other  words, only 3 percent  of all the
permits issued have been  contested.   Of those adjudicatory hearings requested, 1,105  are
still pending.   A look at those pending requests reveals that it is not necessarily the major
discharges that give us the most grief,  because by and  large we will have devoted more time
and energy and thought to the limitations going into those permits.   The minor dischargers,
on the other hand,  pose some very difficult problems in peripheral areas which must be
resolved  before a reasonable permit can be issued.   Such is the case for petroleum bulk
terminals, two or three other sub-categories, some water treatment  plants and other classes
of permits which make up a good portion of the 1,105 adjudicatory  hearing requests yet to
be resolved.   It is  also interesting to note that only ten of the 750  adjudicatory hearings
that have been  resolved required the use of all the formal adjudicatory  proceedings avail-
able under the  law.  By  far, the majority of all adjudicatory hearings requested are settled
prior to trial.

      Insofar as the petroleum  refining industry is  concerned, most — if not all — of the
approximately 250 petroleum refineries in the  nation have now  been issued NPDES  permits.
Of those, only  about 20 are currently pending under the adjudicatory hearing request
review process,  having been contested on some requirement or condition of the permit.
Typical issues in the adjudication of petroleum refining permits include deep well injection

-------
                                                                                   61
controls/ storm water runoff treatment, supply-water silt disposal and other miscellaneous
items.

      Overall, the NPDES program has proven quite workable in the petroleum refining in-
dustry.   The basic requirements for Best  Practicable Treatment of process water do not seem
to have constituted a major source of disagreement between the industry  and  EPA.  Earlier
statements provided at this Forum give further proof that the industry as a whole  is well on
its way to achieving the installation of the equipment necessary to meet  Best Practicable
Treatment.

      To sum up, NPDES requirements were designed to make sure that discharges of pollu-
tants into the nation's waters are subject to uniform minimum controls and that the restrictions,
terms,  and conditions of a permit are substantially the same whether issued by EPA or a State.
In brief, water pollution is a national problem, and the permit system is  part of a national
program to combat that problem.

      I  will be happy to take any questions at this time.

 BIOGRAPHY

      Mr. Schafer received his degree in Chemical
 Engineeeing from  Georgia Institute of Technology, and
 has extensive experience in industrial production and
 development with Shell  Chemical Company (Petro-
 chemicals,  Houston, Texas) and Celanese Figers Company
 (cellulose fibers,  Cumberland, Maryland).   He joined
 EPA in January 1972, and is currently Chief of the
 Permits Assistance Branch,  Office of Water Enforcement,
 in Washington, D. C.   As such, he  is responsible for
 quality control,technical assistance and  policy coordina-
 tion and policy coordination for EPA regions and states
 writing National Pollutant  Discharge Elimination System
 permits for industrial point source discharges.

-------
                "CURRENT APPROACH  TO REFINERY PERMITS"

                               Robert F. Silvus
                   Texas Water Quality Board, Austin, Texas
HISTORY
     Much of Texas is water short, and water is a valuable resource which has always
needed protection.  The first official  recognition in Texas came in I860, when a law was
passed to make water pollution illegal.   But, it wasn't until the 1930's that the degree of
allowable pollution was quantified by a State Health Department requirement of secondary
treatment of sewage effluent from cities.  From that time until  1967, the State Health
Department was responsible for water  pollution control in Texas.  The Texas Water
Pollution Control Board functioned (as a  part of the Health Department) from  1961 to 1967
and  initiated a permit system for both municipalities and industries.  The Texas Water
Quality Board was established by the  legislature as an  independent agency in 1967.  It
presently performs the administration of the Texas Water Quality Act and the Solid Waste
Act  and has responsibilities under the Subsurface Disposal  Act.  The Board  is composed of
three public members and representatives from four other State agencies with interest in
water.

     Most State permits  were written 10-15 years ago and revised since then as water
quality problems were recognized, plant expansions occurred, etc. The initial efforts
were more of a cataloging operation to denote the location and approximate quality and
quantity of discharges.  State law requires control of effluent volume as well as average
and  instantaneous quality.  A self-reporting system was established in 1970, and com-
puterized data for each outfall of each discharger  is available for review.  Traditionally,
the State has relied on treatability studies to set limitations for industrial permits with the
overall goal of "good secondary treatment".

SINCE  EPA

     There has been quite an impact on the State permit system  since the implementation
of the  NPDES permit program beginning in 1972.  The  nationwide guidelines adopted by
EPA  provide another and important means to check industrial permits for reasonableness of
effluent requirements and equity among  permit holders.  The State permit format has been
revised to resemble the  NPDES format, and a working agreement established with the EPA
whereby we  write most NPDES permits for the EPA for Texas dischargers.  Currently, we
consider (1)  waste load  evaluations and allocations for the various stream segments, (2)
stream standards, (3) treatability studies, (4) self-report data, (5) the existing State  per-
mit and (6) guidelines in considering a permit revision. It should be emphasized that
before the guidelines were issued most refiners in Texas had either already  built or con-
tracted for wastewater treatment plants which can now meet or exceed  guidelines.  In
fact, six of the twelve exemplary refineries  used to develop the guidelines are located in
Texas.  These are Coastal States and Champlin at Corpus Christi, Union Oil at Beaumont,
Marathon at Texas City, Shell at Deer Park  (near Houston) and Phillips at Sweeney.
Most treatment systems include oil-water separators, dissolved air flotation units, equali-
zation basins, biological treatment units, clarifiers, and a sludge handling system.
                                        63

-------
  64

     Having taken care of the major treatment problems during the late 1960's and early
 1970's, most Texas refiners can concentrate their efforts on the remaining problems of
 storm water handling and hazardous metals control and should be able  to meet guidelines
 in 1977 without additional major expenditures.  However, a  few old refineries do not yet
 have adequate treatment facilities and will have difficulty meeting even the 1977 guide-
 lines. It is almost impossible to carry out a large wastewater handling and treatment
 program  over a period  of years without the continuous commitment of corporate manage-
 ment to the program.  The State strongly advocates a management  policy which
 encourages timely investments in concrete  and steel rather than  in a good staff of
 attorneys.

 PERMIT PREPARATION

     Now that you know how we arrived at our current 1976 state of affairs, you  would
 probably  like  to know  what is involved  in the preparation of  a waste discharge permit.
 The Texas Water Quality Board has divided the various water bodies in the State  into
 segments and each segment is regularly  monitored.  A set  of stream standards was adopted
 in 1967 and subsequently revised several times.  Those stream segments which have
 problems are designated  "Water  Quality Segments", and some special  limitations designed
 to correct the problem  are placed on all dischargers to the segment. In the case  of the
 Houston  Ship Channel, a computer model was developed,  and a  waste  load allocation
 expressed in pounds of BOD^ and ammonia-nitrogen was established by the State  for each
 discharger (including municipalities).  The EPA  reviewed  and concurred with these
 limitations.  In addition, the State has  issued several special statewide orders which affect
 many dischargers.   Chief among these is the Board Order governing the discharge of
 hazardous metals.   The latest revision of this order  is stringent and may require special
 treatment to remove or replace catalysts or cooling water  additives which were previously
 discharged.  When an  engineer first begins to draft a permit, he looks  at the stream
 standards, waste load allocations and special orders to determine their applicability.

     A second question for existing dischargers concerns their current performance.  The
 Texas Water Quality Board has 12 district offices around the  State staffed by about 100
 employees.  By reading their inspection reports and reviewing the  self-reporting data from
 the permittee, the engineer can answer questions about current performance.  What he
 does not  know is whether refinery operations were normal  and whether  the refinery was
 operating near its  capacity.  This information along with unit capacities,  etc.  is usually
 obtained by direct contact with the company.

     There is an existing  State permit  for each refinery. Most State permits were written
to include concentration limitations and dry weather flow. The  engineer will calculate
the limitations (in  pounds) which are established in the existing  State permit for later
comparison with guideline calculations.  The guideline numbers  will be used in the new
permit unless the existing State permit requires something  more restrictive.

     From the NPDES or Corps of Engineers application, we would  attempt to determine
the unit capacities so that size factors and  process configuration factors can be calculated.
 Frequently,  additional  information from the company is required.  A storm water

-------
                                                                                  65
management plan is normally a very important ingredient in every permit, and a ballast
water treatment system is important in many cases.  Storm water runoff can be estimated
from a plot plan of the refinery and historical  rainfall data.  A better source of data is
historical in-plant concentration and volume measurements on storm sewers.  Historical
ballast water receipts can be used  to estimate  the ballast water contribution to the permit.
Ballast water is being treated successfully in combination with other refinery wastewaters.

     Normally, storm water data is the most difficult item to quantify in a permit. Since
storm water runoff is a problem common to all  refiners, some further elaboration of the
type of recommendations which we make is in  order.  First, we recommend that process
and storm water sewers be separate. Second,  we recommend that all storage tanks be
diked and that tanks not be drained into the diked area; draining water from tanks into
the firewall  area has been  common practice at some locations.  Third, we recommend that
all process areas be curbed and provided with  drains so that contaminated storm water will
not be discharged without treatment.  It has been the  practice in many plants for
operators to rinse sample bottles and pour the  contents down the drains or on the slab and
to drain  equipment which is opened for  maintenance on the slab. If it rains before this
material  is flushed to a treatment unit,  the storm water will be  contaminated.  Our fourth
recommendation is that all contaminated storm water be captured for eventual treatment
and that  clean storm water be discharged directly.  The permit  usually defines clean
storm water in terms of (1)  pH, (2) TOC and (3)  Oil and Grease content, and this storm
water can be released without treatment. As  mentioned earlier, historical volume and
quality measurements provide the best data on the magnitude of the storm water treatment
problem. Impoundment of the contaminated runoff is  the most common approach.  This
allows the collected water to be treated over  a period of time;  this reduces the hydraulic
loading of the treatment plant which, in turn, reduces the investment.

     Impoundment is not always practical because of the geography of the plant or avail-
ability of land. In such cases, exceptionally  good housekeeping practices are necessary,
and the additional investment for an oversized treatment  plant may also be necessary.  A
final recommendation is that runoff from non-point sources including those areas held for
future development be diverted away from potentially contaminated areas through an
adequate system of dikes and ditches.  In many cases, efforts expended in preventing the
contamination of rainfall runoff can greatly reduce the magnitude of the  problem.

     Another common problem involves the  compliance with chromium limitations. Limi-
tations are usually based on the volume of chromium-treated cooling water discharged
and not on the total  plant effluent.  Since  not every refiner has had success in substi-
tuting other water treatment chemicals for chromium,  some plan to continue to use
chromium but to treat the cooling  water blowdown prior to discharge.  There are  three
processes under evaluation to achieve removal.  These are (1) chemical precipitation and
clarification, (2) ion exchange, and (3) electrolysis.  Chemical precipitation with clari-
fication  is the most proven process but each has  its advantages.  Each of the processes
will  be installed at some manufacturing plant  in Texas before July 1, 1977, so we will
soon learn how well  they perform in routine service.  It is  very difficult for a company to
meet the required chromium levels if the discharges are not confined to a  few large
volume streams.  For this reason, the engineer who prepares the permit will consider each
plant situation individually.

-------
  66


     Having reviewed the stream standards, waste load allocation,  Board orders, operating
 history, NPDES applications and other company input, and the existing  State permit,  the
 engineer will simply  compare the calculated limitations using the existing State permit or
 allocation with calculations using the guidelines and pick the more stringent limitations
 for the July  1,  1977  requirements.   Limitations are norrfraUy placed on BOD5, TSS, COD,
 oil and grease, phenol,  ammonia-nitrogen, sulfides, chromium and pH.   Volume limi-
 tations and requirements on other hazardous metals,  temperatures, and occasionally on
 chlorides and sulfates are added.  Interim  limitations (until July  1,  1977) are usually
 based  on self-report data,  field reports,  current State permit limitations  and federal
 applications.  Limitations for each  time  period on individual samples and on volume are
 also placed in the State  permit.  At this point, it is  normal to discuss the proposed  permit
 with the company to  determine the  accuracy of assumptions, applicability of particular
 guidelines to the refinery, currentness of the data, time required for compliance, and
 approved expansions  or other major changes to take place during  the five year period
 covered by the permit.  Once all points have been explored, the State permit is set for
 public hearing and a  similar NPDES permit is sent to the EPA for  review.  The two  permits
 then proceed  through the various public  notice and hearing procedures where  they may be
 modified as a result of new evidence introduced by the public or  other governmental
 agency. Finally the  permits are adopted by the Texas Water Quality Board and the EPA.
 With good  communication between  EPA and the State during the  public hearing procedures,
 the permits will contain  the same limitations and  the  same sampling and reporting require-
 ments.

     These  procedures have been completed for most  refineries in  Texas,  and 1977
 objectives  are set. There remain questions concerning 1983  limitations.   It is assumed
 that final filtration will  be a standard treatment step  if the 1983 guidelines remain
 unchanged.  Additional  biological  treatment or physical-chemical  treatment or some
 combination will also be required.  The technology is available,  and considering the
 excellent progress which most refiners have made in  the last decade and  the knowledge
 accumulated  during this  period, I feel that the major problems in meeting the 1983  guide-
 lines will be  financial and not technical.

 ENFORCEMENT

    State enforcement action is initiated primarily by the district offices around the
 State.   In addition to regular inspections,  all dates in permit compliance schedules are
 entered into a computerized data system  which is regularly checked.  It  is an understate-
 ment to say that it is  more difficult  to achieve a continuous quality level with biological
 treatment units than with other refinery unit operations.   Recognizing this fact, the State
 has not normally pursued enforcement actions for  isolated  violations of permits caused by
weather changes, etc.  However, where repeated violations occur, enforcement action
will be taken.  This frequently involves an enforcement hearing and subsequent recommen-
dations to the Texas Water Quality  Board for remedial action.  Flagrant  violations are
 normally referred to the Attorney General  for prosecution either directly  or following an
enforcement hearing.   It remains the intent of the State to enforce  State  permit require-
ments .

-------
                                                                                67
SUMMARY
    A partial picture of the background and current procedures for the development of
permits for petroleum refineries has been presented. Most State permits were originally
issued  10-15 years ago and revised as necessary.  Traditionally,  the State of Texas has
relied on treatability studies to set limitations with the overall goal of "good secondary
treatment". Following the establishment of EPA, a working agreement was reached
whereby the State drafts most NpDES permits  for Texas dischargers along with a parallel
State  permit.  Current permits are derived by combining information from waste load
evaluations, self-report data,  existing State permits,  NPDES  applications and guidelines
and company input.  Most refiners already have secondary treatment systems which will
meet or exceed 1977 NPDES requirements.  In general, storm water handling and
chromium and zinc removal constitute the problems to be solved by 1977.

BIOGRAPHY

    Robert F.  Silvus is Branch Chief of the Industrial
Permits Branch, Central Operations Division , Texas
Water Quality Board.  He received a B.  S. degree
in Chemical Engineering and an MBA degree  from
the University of  Texas at Austin.   He is a professional
engineer registered in the State of Texas.   He had 15
years  of industrial experience  in Research and Developmenl
with Ehtyl Corporation and Jefferson Chemical Co., Inc.
before joining the Texas Water Quality Board.

-------
68

-------
     PETROLEUM INDUSTRY EXPERIENCE WITH THE NPDES PERMIT SYSTEM

                                Harold F.  Elkin
                       Coordinator,  Environmental Affairs
          Sun Oil Company,  1608 Walnut Street, Philadelphia, Pa. 19103
                                     and
                              William  K.  Lorenz
                         Senior  Environmental Engineer
          Sun Oil Company,  1608 Walnut Street,  Philadelphia, Pa. 19103

ABSTRACT

     Sun Oil reviews its experiences with refinery NPDES permits.  This discussion
emphasizes the special problems encountered with water quality limited streams, effluent
sampling, laboratory analysis, and self-reporting requirements.  Recommendations are
made that may simplify negotiation and compliance with future NPDES permits.

INTRODUCTION

     Most of the 250 U.S. petroleum refineries have now been issued final National
Pollutant Discharge Elimination System (NPDES) permits. Compliance and enforcement
are now actively underway.  Over the last three years,  the refining industry has had many
problems with the NPDES permit system and has hopefully learned  many important lessons.

     We are speaking with you this afternoon primarily on the basis of Sun Oil Company's
experience at our six domestic refineries.  Sun has accepted final  NPDES permits at all
six refineries.  Our permits are based  almost entirely on the EPA petroleum refining
effluent guidelines with certain more stringent conditions based on  water quality-related
effluent standards.

     We are planning to achieve our 1977 permit limitations through the use of conven-
tional treatment technology including sand filtration. At this time, we are  not certain
that this  technology can achieve the daily 24-hour maximum effluent limitations for all
parameters. Sun's largest refinery at Marcus Hook,  Pennsylvania will  join a regional
treatment plant in early 1977. We believe this experience  is typical of the  problems
encountered by most U.S. refiners. Many of these experiences are also relevant to the
marketing and production phases of the oil industry.  The purpose of this paper is to dis-
cuss our most significant  NPDES problems and to make some recommendations that may
simplify negotiation and compliance with future NPDES permits.

     Several of the more vexing NPDES problems occur throughout all the stages of the
refinery NPDES  permit process.  For clarity, we will discuss these issues on  a functional
basis, rather than chronologically.  Table 1  lists the  three subject headings  we feel are
worthy of special discussion at this conference.  We have excluded reference to the  EPA
petroleum refining  effluent guidelines, and will  limit our discussion to the basic NPDES
problems which are somewhat independent of any specific set of effluent regulations.
Further, we assume our audience has some experience with  the NPDES permit system so
                                       69

-------
70
that we will not dwell on the procedural and legal aspects of negotiating and appealing
NPDES permits.

1.   Water Quality Limited Streams

         According to the Engineering-Science Report for the National Commission on
     Water Quality, 68 refineries, having a cumulative crude capacity of 6,800,000
     bbl/day,  now discharge to water quality-limited streams (Table 2).   Thus, over one-
     third of U.S. refining capacity is partially, if not totally, controlled by water
     quality standards, rather than the EPA technology-based effluent standards.

         The impact of the water quality-related effluent limitations can range anywhere
     from Best Practicable Control Technology Currently Available  (BPCTCA) limitations
     to zero discharge.  Individual states are responsible for  classifying specific river
     segments as effluent quality-limited or water quality-limited on  the basis of EPA-
     approved state water quality standards.

         Fora  refinery discharging into a water quality-limited segment, permit
     negotiations can be a complex and frustrating experience (Table 3).

         First, it is necessary  to establish the water quality standards upon which the
     effluent limits are based.  We have found  that these standards are not always
     available in specific form in the legally-promulgated state regulations.  In some
     states,  the regulations are narrative with the specification of numerical  water
     quality standards up to  the state agency.   In the  latter case, it is very important
     to determine if the internal state agency decisions have  been open to public comment
     according  to the relevant Administrative Procedures Act.

         The states have employed a variety of methods for calculating the total
     allowable effluent load that will achieve water quality standards on a particular
     stream.  These techniques range from sophisticated computer models to mere
     educated guesses.  No  individual method is satisfactory for all circumstances.   A
     major stumbling block appears to be the unavailability of historical data.  The
     important fact is that the  individual company must analyze the state  methods in
     sufficient detail to determine if there is an adequate  basis for establishing its
     effluent standards.

         The next step is to allocate to all individual dischargers on  the stream segment
     a portion of the total allowable segment load determined by the  mathematical  model
     or educated guesswork. This process opens up a complex set of economic equity
     problems.  Mathematical models are available to set  load allocations in a variety
     of ways.  As you all know,  it is very difficult to  quantify the relative environmental
     impacts of a variety of different effluents to an individual  stream segment; thus, it
     is almost impossible  to objectively determine individual  abatement schedules.
     Nevertheless, these decisions are being made every day and all dischargers,
     municipal and industrial,  should play an active role in setting  load allocations.

-------
                                                                                 71
         Wafer quality-limited river segments will also be classified for the  1983
    water qualify standards.  We can expect that our next round of NPDES permits
    will also include a significant number of refineries discharging  to these  river
    segments.

2.  Effluent Sampling and  Laboratory Analysis

         One of the most significant changes brought by the NPDES system is the
    imposition  of uniform requirements for measuring wastewater flows and the obtaining
    and analyzing of representative effluent samples (Table 4).  Prior to  PL  92-500,
    all Sun refineries had evolved their own monitoring programs based on local and
    state regulations.  However, some waste streams like  tank farm storm runoff were
    never measured or sampled before the 1972 Federal Water Act.

         As you know, in most cases the  interim and final  refinery effluent limitations
    are based on mass-weight (Ib/day) discharge rates.  This requires that the flow
    measurements,  sampling techniques and analytical methods be of comparable
    precision and accuracy.   It makes no sense, whatsoever,  to monitor a multi-million
    dollar treatment facility with trained chemists using sophisticated analytical
    instruments if the flow measurements are only within - 20% of the real  flow rate.

         Inadequate historical flow data  is one of the main reasons that our refineries
    have had some  non-compliance  incidences with  interim permit conditions.  Such
    problems stem from both the lack of the quality and quantify of historical flow data.
    Hopefully, these problems will not reoccur since we are now establishing flow
    measurement systems that should provide an adequate amount of accurate and
    precise flow  data.  For example, our Marcus Hook, Pennsylvania refinery will
    join a large  municipal-industrial regional  treatment system that requires us to
    install a  flow meter accurate to within _+ 1%.

         Prior to  PL92-500, our refinery  practices were about evenly split between
    grab and composite samples. We are now running more composite samples as
    required  by,NPDES permits.  However,  we expect to  continue  faking grab samples
    on segregated storm runoff. In developing our specifications on sampling methods
    and sample preservation, we have relied on the guidance of the EPA technology
    transfer handbook entitled, "Monitoring Industrial Wasfewater." This reference
    is quite complete and is an excellent source of information on sampling  equipment
    and flow measuring devices.

         Laboratory testing is rapidly becoming the most critical part of the  NPDES
    compliance monitoring system.

         As an  example, we are all spending a lot of money to  build sophisticated
    treatment systems.  We measure our performance by laboratory  results.  But  what
    does it all  mean if another laboratory analyzes an identical sample and  obtains
    results five or ten  times higher than our own?  This is a very significant problem

-------
72


     if the second laboratory happens to be an EPA laboratory and the samples  are being
     analyzed as part of an EPA annual site inspection.

         We  have only begun to look into the areas of analytical precision, analytical
     accuracy, and  interlaboratory variability. Our fundamental interest is in producing
     reliable  information that is legally defensible in administrative and judicial
     proceedings.  If we really do have a non-compliance situation, we need accurate
     data to identify and solve the problem.  On the other hand,  if we are in compliance
     with the effluent limitation, we must have accurate data for use in any governmental
     enforcement action.  In the long run, we need accurate and precise effluent data so
     that the  1983 Best Available Technology Economically Achievable (BATEA) effluent
     standards better reflect reality.

         By way of  example,  the State of Texas is now implementing an analytical
     quality control  program for use  in  NPDES compliance monitoring.  Once this
     program  is operational, the Sate will back up its effluent sampling results with
     information on  the precision and accuracy of the test method.  If an industrial
     discharger wishes to challenge the State analytical  results, he will be at a distinct
     disadvantage without comparable precision and accuracy data on his own laboratory.
     These quality control programs are expensive and time-consuming to operate, but
     they may be necessary in order to maintain a balance in  the enforcement of NPDES
     pe rm i ts.

3.   Enforcement of Self-Reporting Requirements

        As we all know, the bases for the enforcement of NPDES permits are  the many
     self-reporting requirements that the permittee must satisfy.  Table 5 shows the major
     headings of a typical NPDES permit issued in  EPA Region VI. The format may vary
     throughout the  country, but the main subjects are included in all NPDES permits.

         Part 1,  Section A details the specific effluent limitations and monitoring
     requirements, but does not stipulate the form or frequency of any self-monitoring
     reports.

         Part 1,  Section B sets forth  the specific timetables for implementing the limit-
     ations  in Part 1, Section A.  These timetables require progress reports from design
     of facilities through the attainment of operational levels.  We have been  successful
     in negotiating reasonable compliance schedules,  but still recognize that the interim
     dates are only educated guesses  based on very little prior experience in designing
    and building refinery wastewater treatment facilities.  As long as the final deadline
     is satisfied,  we believe that the interim dates should have some flexibility. The EPA
    or the states cannot now agree to interim date modifications without formally
     modifying the permit including  proper public notice.  We suggest that some red
     tape could be cut by a revision in EPA's regulations to allow them some internal
     flexibility in modifying interim compliance dates as  long as final deadlines can  be
     achieved.

-------
                                                                                  73
         Part1 1, Section C covers the details of analytical test procedures, reporting
    of results and retention of records.  Under this section, the report of all
    analytical  results is prepared for EPA or the states every three months.  The
    regulatory  agencies can seek enforcement actions against the permittee on the
    basis of the quarterly report.

         The procedures and reports of this section do not reflect the accuracy and
    precision of the compliance monitoring process nor the problem of interlaboratory
    variability.  We know that the EPA effluent guidelines attempt to reflect the
    variability of wastewater effluents.  However,  we are not satisfied  with  the EPA
    methodology.  The total  variability of a wastewater effluent must be broken down
    into its component parts.  Basically, we  need to differentiate between the
    controllable vs. uncontrollable sources of variability.  Then, a compliance
    monitoring system  could be developed  that 1) does not try to limit the uncontrollable
    variability and 2) does attempt to limit the controllable variability  sources in a
    probabilistic manner.  We are still  learning about the variability issue and have
    no specific systems to offer at this  time.  But,  we are convinced that the existing
    data base is not adequate.

         Part II,  Section A includes a subheading on non-compliance notification.
    Within five days that a permittee becomes aware of exceeding any daily maximum
    effluent limitation, the permittee must notify the agency.  This notification must
    describe the problem and specify what, if any, remedial actions are being taken.
    We consider this requirement to be an  example of unnecessary red tape.   Further,
    we have had some difficulties in implementing the five-day requirement, especially
    when commercial laboratories are performing the analyses. After all,  it is our
    legal responsibility to achieve both the daily maximum and monthly average
    effluent limitations.  Why should we have to take special steps on the daily
    maximum limitations and not on the monthly average  limitations? Each daily
    non-compliance is a separate violation subject to considerable  fines and adverse
    publicity.  We believe that we can manage our wastewaters without the  added
    burden of this daily maximum non-compliance notification. Our recommendation
    is to delete this provision from NPDES permits.

         The remaining sections of the NPDES permits do  not include any specific
    self-reporting requirements that are worthy of specific mention  in this paper.

CONCLUSIONS

    This concludes our prepared remarks on experiences with petroleum  refinery NPDES
permits.  To sum up:

    I.   Over one-third of all U.S.  refining capacity discharges to water
         quality-limited streams.  Special  considerations  must be taken  in
         order  to negotiate an optimum  set of permit conditions.  Water
         quality-limited discharges will also  be an  important factor in the
         1983 permit negotiations.

-------
 74
     2.   Effluent sampling and laboratory analytical procedures are now becoming
         more standardized and more sophisticated.  All phases of the compliance
         monitoring system must be of comparable  precision and accuracy.
         Analytical quality control programs and round-robin  testing  may become
         a necessary part of the NPDES permit process.

     3.   Self-reporting is the primary basis for enforcing NPDES permits.  Most of
         the existing  requirements are necessary, but some improvements can be made
         to cut red tape and  improve the identification of an actual non-compliance
         incident.

     We believe that our experiences are typical of other refiners and also relate  to
problems that will be  encountered by the marketing and production activities of the oil
industry.  We have purposely not dwelled on the EPA effluent guideline regulations
believing that these criticisms are being handled in a separate forum.
BIOGRAPHIES

     William K. Lorenz is Senior Environmental
Engineer in the Environmental Affairs Department
of Sun Oil Company.  Bill holds the following degrees:
Bachelor of Chemical Engineering from Villanova
University, Master of Science in Environmental
Engineering from  Drexel  University, and  expects to
receive a Masters in Business Administration from
Drexel University in 1977.

     Harold F.  Elkin is Coordinator of Environmental
Affairs for Sun Oil Company in Philadelphia.  He
holds the  following degrees:  B.S. in Chemistry,
University of Pennsylvania;  B.S.  Engineering,
Drexel University; M.S.  Environmental Engineering,
University of North Carolina.  He is a professional
engineer registered in Pennsylvania.

     Mr. Elkin has served on the Pennsylvania Air
Pollution Commission, advisory panels to EPA, and
numerous industry environmental committees.

-------
                                                                                            75
                    TABLE 1  SELECTED PROBLEM AREAS IN NPDES PERMITS

                    1  -  Wafer Qualify Limited Segments
                    11  -  Sampling and Laboratory Analysis
                   111 - Enforcement of Self-Reporting Requirements

                TABLE 2  DISCHARGE LOCATION OF REFINERIES IN THE U.S.
                                                                           Crude Throughput
Type of Effluent Limitation             Number of Refineries                   1,000 bbIs/Stream Day

Effluent Quality Limited                     106                                 5,729
Water Quality Limited                       68                                 6,954
Municipal System                            34                                 1,720
Zero Discharge                              38                                   615

                        TOTALS           246                                14,918

    Source:  Engineering  Science, "Petroleum Refining Industry Technology and Costs of Wastewater
            Control" Table II -- 6, page 11-25


                  TABLE 3  WATER QUALITY BASED EFFLUENT LIMITATIONS

                    o  Applicable State Water Quality Standards

                    o  "Modeling" Techniques
                           - Computerized River Model Studies
                           - Manuel Computation River Model  Studies
                           - Educated Estimates

                    o  Individual  Load Allocations
                        TABLE 4  EFFLUENT SAMPLING AND ANALYSIS

                     o  Lack of Historical Data
                     o  Flow Measurements
                     o  Analytical Quality Control
                     o  Interlaboratory Variability
                     o  Legally Defensible Monitoring Programs
                            TABLES  OUTLINE OF NPDES PERMITS

                                  EXAMPLE: EPA REGION VI

 PART I               A - Effluent Limitations and Monitoring Requirements
                      B - Schedule of Compliance
                      C - Monitoring and Reporting

 PART II               A - Management Requirements
                      B - Responsibilities

 PART III  -  OTHER REQUIREMENTS

-------
76

-------
                     "THE ENFORCEMENT OF PERMITS"

                             Harless R. Benthul
                 Chief, Legal Branch, Enforcement Division
                        U. S. E.P.A., Dallas, Texas

     Before I  get into the substance of my discussion I would like to extend personal
thanks to Bob Silvus.  He did something in  his talk that  I have been waiting for for some
time, and that  is Bob acknowledged that two people, non-lawyers presumably,  can look
at the same piece of information or the  same data and reach different conclusions.  I had
heard since I got a  law license that it was unique to attorneys that we could look at the
same piece of English language and forever argue two different conclusions, but I am
grateful  to Bob that he has made progress for us, who knows tomorrow we may be con-
sidered people  as well.  I am going to talk a little bit about enforcement of federally
issued NPDES permits.  Federally  issued because in the  Region where I work, Texas and
the four  bordering states, we don't yet have a state which has been formally delegated
NPDES permit authority, although we do work with those states on various cooperative
bases. Nevertheless my discussion will be  on enforcement of federally issued permits.

     We  are now just over three years from  the enactment of public law 92-500.  Many of
the deadlines of the Act have come and gone.  Others are yet before us - you as the
operators of regulated  facilities, we as  EPA the regulatory agency.   Two deadlines are of
specific  interest today.  They are  first,  the December 31, 1974 moratorium expiration
date which became the target by which all major permits (and some others) were to have
been issued and second the July 1, 1977 requirement for attainment of best practicable
control technology  currently available (BPT).

     The  December  31,  1974 date  saw the issuance of virtually all  permits for major dis-
chargers  in Region VI and EPA  nationally.  What is a major industrial discharger is a
matter of evolution.  It has included generally  large production facilities such as
refineries, chemical plants, certain power  plants,  metal processors and others,  not based
solely on size or diversity of waste,  but on the strength  or toxicity of waste or effect on
the receiving stream.  Those to be added to the major permit list in the future will
probably be based  on considerations such as potential harm to the receiving stream,
significance  of pollution abatement facilities, extent of hazardous material handled and
the like.  Finally,  it should be remembered that designation of major permits is in large
part a label for purposes of allocation of our resources.

     The  number of  such industrial  permits in Region VI, is now about 340.  It should be
noted that the claim of having  issued those permits rests in part on  the procedural system
which allows uncontested portions  of permits to become  effective notwithstanding the
fact that an adjudicatory hearing is pending on other portions of the  permit.  Depending
upon the day of the count,  the number of such pending adjudicatory  hearings is around
100  in Region VI and about 1100 nationally.

     Parts of almost all of those 100 contested permits, all the other 200 plus majors, as
well as many minor industrial and some municipal permits are effective.  They have been

                                        77

-------
78


effective for a year or more in many cases.  As you are aware all of these industrial
permits require the discharger to attain effluent limits equal to BPT by the statutory date
of July 1, 1977 or other more stringent limitations established pursuant to Section
301(b)(l)C based on water quality standards  or other federal or state law.  BPT may be
defined by effluent guidelines promulgated pursuant to Section 304 or  in the absence of
guidelines on a case by case basis by application  of professional engineering judgment.

     As is well known to this particular group, many effluent guidelines remain in liti-
gation.  Yet if one can detach himself from  this hopefully temporary delay in  our joint
implementation of the Act, it is  clear that if the July 1,  1977 mandate is to be realized,
one thing that must be done is to uphold the  credibility of the issued permits.

     Parenthetically, I would point out that  EPA administrator Russel Train recently in a
case involving Bethlehem Steel,  concluded that the July 1, 1977 date is a mandatory
date and  permits must require permitees to meet it.  This position upheld a previous
decision of the general  counsel.  That was No. 26 if you want to look at it or have your
attorneys look into it.  Enforcement of these permits is an  element of maintaining that
credibility and enforcement is well under way.

     The principal  legal basis for enforcement of NPDES permits is generally Section 309
of the Act.  I am not going to bore you by reading that whole page, I  will read a
section of it  of particular importance.  Potentially available also for direct enforcement
of permits is  Section 504.  Also available  for collateral support of a municipal permit is
Section 402(h) by which injunctive relief  is  available to prohibit additional pollutant
load being introduced where there is an existing violation of that municipal permit.

     However for most violations of NPDES permits, Section 309 will be the primary
enforcement  tool.  Since  we are here concerned with  federally issued permits  the parts
of Section 309 of most immediate interest are Sections 309(a)3, 309(a)4 and 309(b),(c)
and (d).  This results from  the fact that Sections 309(a)l and (2) are  concerned with
state-issued permits and 309(e) deals with  municipal permits.  I will not in this discussion
deal with enforcement of true municipal permits where there are violations caused by
industrial or  other  non-domestic waste contributors.  There are other classes of permit
holders which claim some attributes of municipalities as defined in the Act, but treat
almost exclusively industrial waste.   These applicants have been  in our Region and will
continue to be dealt with on the basis that each industrial contributor  is a party to the
permit and responsible directly to EPA along with  the  applicant for performance of the
permit.  These permits are not generally addressed here either.  Thus we will  be  dealing
with enforcement of single-party, federally-issued permits for industrial facilities.

     Section  309(a)3 reads as follows,

           "Whenever on  the basis of any  information  available to him, the administrator
           finds that any person is in violation of  Section 301,  302, 306, 307 or  308 of
           this Act, or is  in violation of any permit condition or limitation implementing
           any of such sections in a permit issued  under Section 402  of this Act by him
           or by state, he shall issue an order requiring such  person to comply with such

-------
                                                                                  79


         section or requirement, or he shall bring a civil action in accordance with sub-
         section (b) of the Act. "

    A few remarks are in order about word  usage in this section.  First, the word "any"
appears twice in the first two lines preceding both information and person.   This would
appear to be all inclusive and  cover self-reported, agency-discovered and  citizen-
reported information (information relating to violations).  Then note that the language
encompasses both violations of certain enumerated sections  of the Act and violations of
permits which implement those sections.  Then follows the directive to the administrator.
He sha 11 issue an order requiring compliance or he shall commence a civil action in
accordance with Section 309(b) (emphasis added).  I  will touch later on this directive
language.   The authority to so act  has been delegated to  regional administrators.
Section 309(a)4 provides for sending a copy of these administrative orders to states; it
provides that service of orders  shall be by personal  service, that they  shall  state with
reasonable specificity, the nature of the violation and it  also puts an  outside limit of 30
days on the time for compliance.  This section also creates  a right on  the part of the
recipient of the order, to a conference as a condition to effectiveness of the order
relating to a violation of Section 308.  Section 308 deals with monitoring,  reporting,
sampling, inspection and the like.  Parenthetically I would say that in orders we  have
issued dealing with monitoring requirements of the permit or violation of monitoring
requirements we have been very careful, we think, to be sure  to post date the effective-
ness of  those orders and to explicitly advise people that they had a right to a conference
before the order became effective.

    Section 309(b) authorizes  civil actions for what is called "appropriate  relief",
including (but not limited presumably to) injunctive relief for any violation for which an
order  could have been issued under Section 309(a)3 (that  is the section I read just a
minute ago).  Section  309(b) also establishes venue,  that is where the suit is to be filed
for such proceedings and requires immediate notification of the appropriate state.

    Section 309(c)1 contains the criminal  sanction for violation of enumerated sub-
stantive provisions of the Act or a permit implementing  them.  The criteria  for Section
309(c)l criminal conduct is "wilful or  negligent".  Liabilities  of $2500 to $25,000  or up
to one year in prison or conceivably both,  per day of violation are imposed and the
limits are double for a second offense.

    Section 309(c)2 makes criminal a  knowing false statement, representation or certifi-
cation of certain documents filed or required to be  maintained under the Act.  Tampering
with monitoring equipment is a criminal offense. Penalties range up to $10,000 or  six
months imprisonment or both.

    Section 309(c)3 adds for the purpose of 309(c),  any responsible corporate officer to
the general definition  of the term  'person'  as contained in Section 502(5).  502(5) defines
person generally for purposes of the Act.  309(c)3 adds to that definition for  the purposes
of 309(c),  the term "responsible corporate  officer".

    Finally, Section 309(d)  creates a  civil penalty of up to $10,000 per day of violation

-------
80

of the same sections listed in 309(a)3 or a permit condition implementing them.  On its
face this liability,  the civil penalty liability,  is strict.  That is, it applies without the
necessity of proof of fault on the part of the violator.  The question of how many
penalties are assessable for multiple violations  in one day has been addressed so far by
only one court that I am aware of.  That was the Detrex case which indicates a $10,000
penalty per day only may be assessed. That is, not several $10,000 penalties in the
same  day for multiple violations and that is based on a comparison of language in Section
309 and  Section 311 which is the oil-and-hazardous-material section.  One case
addressed the point and that was the result.  In my opinion there will be more heard on
that point as to whether more than one $10,000 penalty could be assessed for more than
one violation - for  instance, for violation  of two different effluent parameters.

     Let me talk now about the mandatory aspects of Section 309(a)3 which was again
the language that I read for you out of the statute a few minutes ago.  I  have referred
previously to Section 309(a)3 as directive in  nature.  This is because it says that the
Administrator shall, and I emphasize shall, issue an order or commence a civil action
upon  the finding  that any person is  in violation of a permit as well  as in violation of the
enumerated sections of the Act.  Literally  read, there is no discretion except as to
which remedy to  exercise - issue an order or  commence a civil action.

    To date we have had to exercise some discretion  because there have been more
violations (of various levels of seriousness I should add) than we have had resources with
which to develop orders or prepare  referrals or  court actions.  Consequently we have
tried  to issue orders and commence  civil actions in cases where we  felt it most important
to do so.  We have noticed others by warning letter.  Notwithstanding resource limits on
the part of EPA and the States for that matter,  the language of Section 309(a)3 remains
what  it is.

    This language using the term "shall" is pretty clearly mandatory upon the Admin-
istrator once there  is a finding  of a violation.  The requirement of 309(a)3 is qualified
elsewhere in Section 309 only if a state agency has been delegated NPDES authority  by
the provisions of Section 309(a)l.  Now if you refer to Section 309(a)l, it  suggests the
argument that the only circumstances where the Administrator has an alternative,  and  I
will come back to that in a minute, to issuing an order or bringing  a civil action under
309(a)3, is where a state  has been delegated NPDES authority.  However the EPA may
be free to seek civil or  criminal penalties in  addition to the order or civil action con-
templated by 309(a)3.  In any event if we  are operating  under 309(c)l, where a state
has authority to issue  permits under this law,  even then the alternative to issue an order
or commence a civil action appears to me to  be limited to a 30 day waiting period after
giving notice  to a state of a violation to see  if the state  does or does not commence
appropriate  action.  This  option is not available where,  as here, we are dealing with
federally issued permits.  Now the  word alternative that I used, was used advisedly be-
cause of  language in another section of the Act, 505(a) of the Water Pollution Control
Act, which authorizes citizens suits to compel  the administrator to  perform  any act or
duty which is not discretionary, and I will  risk boring you by  reading part of section 505.
It reads - "except as provided in sub-section (b) of this section, any citizen may
commence a civil action on his own behalf against any person including the United

-------
                                                                                  81


States and any other governmental instrumentality or agency, to the extent permitted by
the eleventh amendment to the constitution,  who is alleged to be  in violation of a) an
effluent standard or limitation under this Act or b) an order issued by the Administrator
or state with respect to such standard or limitation, or against the Administrator where
there is alleged  a failure of the Administrator to perform any act or duty under this Act
which is not discretionary with the Administrator". Then it goes not to say that federal
district  courts will have jurisdiction  without regard to  the amount  in controversy or
citizenship of the parties to enforce  such an effluent standard limitation or such an order
or to order the administrator to perform such act or duty as the case may be herein to
apply any appropriate  civil penalties under Section 309(d).  Well, so what does all  that
legalese get for us? Referring back  to 309(a)3, which is the part  you will recall that
said that on the  finding of a violation by any person the  administrator shall  issue  an  order
or commence a  civil action, a compelling  argument can  be made that once a finding of
a violation has been made on  the basis of any information available by  the Administrator
in a pre-NPDES delegation situation,  he has no discretion but to issue an order or
commence a civil action per 309(a)3 and that such duty is  enforceable by a citizen
through Section  505(a).

     This argument is very similar to  the result of a case that was decided  under the Clean
Air Act called Wisconsin Environmental Decade versus Wisconsin Power and Light et al;
one  of the et als being the administrator of the EPA.  This  was in the  U.S. District Court
for the western district of Wisconsin  and decided in 1975.  You will pass it on  to your
attorneys that case  is printed in volume 7,  Environmental Reporter Cases,  page 2022.
Here the court construed part  of Section 113(a) of the  Clean Air Act an conferring a non-
discretionary duty on the Administrator to issue a notice  of ciolation, which duty is
enforceable under citizen suit provisions of the Clean Air Act.  Now I  am going  to bore
you again with some reading,  which I  think you will find interesting.  This is Section
113(a)l of the Clean Air Act.

         "Whenever on the basis of  any information available to him, the administrator
         finds that  any person is in violation of any requirement of an applicable
         implementation plan, the administrator shall  notify the person in violation  of
         the plan and  the state in which the  plan applies of such finding.  If such
         violation extends beyond the 30th day after the date of the administrator's
         notification, the administrator may (emphasis added) issue an order requiring
         such person to comply with the requirements of the plan or bring a civil action
         in  accordance with sub-section (b).11

     The "may"  that I read refers to  what can be done  after the  30 days expires.  "Shall"
was back up earlier - whenever on the basis of any information  the administrator finds
that any persons in  violation he shall notify the person in violation.  It reads very much
in words like 309(a)3 in the water act which  we have been reading.  The  citizen suit
provisions of the Clean Air Act are very very similar to Section 505(a)  of the Water Act
which I read to you.  The court in the Wisconsin case  also indicates that the act of the
Administrator 30 days after the notification is discretionary and that the initial finding
which generates the notice of violation must be a good faith effort based on investigation
of any evidence available.  By analogy, and I emphasize  by analogy,  similar  criteria

-------
82

could apply to a founding of a violation in connection with Section 309(a)3 orders or
civil actions.  The case is also interesting in the fact that the violation was on alleged
violation of a state implementation plan as compared with an NPDES permit.  The court
holds that the EPA may,  in determining whether a violation has occurred which requires
that a  notice of violation be issued, defer to the state's interpretation of its plan as long
as the  state's interpretation is reasonable and consistent with the Act.

     Thus there appears to be a compelling conclusion, if not inescapable,  on the basis
of this analysis and that case that I have cited,  that the 309(a)3 requirement to issue an
order or commence a  civil action is non-discretionary once a violation is found to exist
and  that that duty is enforceable by citizen suit through Section 505.   In checking this
last  week, I discovered the existence of one unreported case in a Federal Court in the
northern district of Alabama which reached a contrary result on whether this is a dis-
cretionary or non-discretionary duty.  I  would caution that the  result was reached on a
preliminary motion and that the case is on appeal.  It has not turned up in the reporters
yet to my knowledge.

     Now this is not to suggest that issuing an order or commencing a civil action such as
an injunction  under 309(a)3 is an exclusive remedy.  We believe, and  we have operated
on the premise that the civil penalty and criminal sanctions of 309(c) and (d) compliment,
in appropriate cases, Section 309(a)3.  There has been one case that held that the
309(a)3 remedies are  not exclusive.  That is a case which was decided, I believe, in the
District Court for the district of Arizona where it was claimed that EPA had  issued an
administrative order and  therefore was not entitled to collect penalties for violations that
were covered  by the administrative order. The judge said that's not right,  the two go
together.

     Finally, it should be noted in the face of all this grimness that there may in fact be,
and  may in law be, a place in the execution of  Section 309 for what has traditionally
been called prosecutorial discretion.   However,  the extent to which this may be
engrafted onto the Act when Congress  used the language it did presumably being aware
of the  concept of prosecutorial discretion remains to be seen. And. also if there  is dis-
cretion it may  lie  in the  area  of the finding of a  violation.   I think this is sort of hinted
at in the rational behind the Wisconsin case that I mentioned.   Time and experience will
tell. This doesn't mean that we, EPA,  are going to issue  administrative orders on trivial
violations and it doesn't  mean we are going to refer bad cases to the U.S. attorneys,
but that we see our enforcement responsibilities  as rather explicit and rather demanding.
I  would like to conclude by making a few observations on some  of our experience to date
in enforcement of  permits.  We have been attempting to enforce these permits in Region
VI,  and I think nationally,  as long as we have had them issued. During fiscal year 1975,
approximately 65 administrative orders were issued to a variety  of dischargers for various
violations.  Beginning in fiscal year 1976 or July 1975, a greater emphasis on enforce-
ment of permits has taken place.  During the first six months of  fiscal year  1976, that is
July to December  1975 inclusive, we have issued 48 administrative orders involving  38
violations of effluent limits,  13 violations of reporting requirements and 6 violations of
schedule of compliance.   During the same period there were a total of 4 referrals to
U.S. attorneys for prosecution.  The prosecutions have all been for civil penalty

-------
                                                                                 83


collections and one case for injunctive relief involving a municipality.  The referrals
involve for the most part, violations of effluent limitations or serious bypassing,  that is
running waste around  the treatment plant, and we have recommended collection  of civil
penalties for each day in violation.

     It is worthy of observation that a problem keeps recurring which we are trying to
discourage by making referrals as it comes up in appropriate cases.  The pattern is that a
difficulty of some nature will arise within a  production facility  or perhaps independent of
production, but in any event the facility operator is faced with a choice of whether to
discharge untreated or partially treated waste  in violation of his permit or curtail
production.  In the instances we have had occasion to carefully examine so far,  we have
concluded that a referral was  in order.  I should say in some of  the cases, not all of the
cases.  The point we are trying to make is that if there is a choice of violating a permit
and keeping production going, then  that choice is  in all likelihood going to have a price
attached to it, and that concludes my prepared remarks.  I  will  attempt  to answer
questions  if anyone has any.

DISCUSSION

Milton  Beychok:    I have a question and I am not sure whether  it should be addressed  to
you or one or more of the previous speakers.  We have heard a great deal this afternoon
about compliance schedules for industry, violations by industry  and enforcement  of
sanctions against industry,  be  they discretionary or  non discretionary.  I would like to put
the shoe on the other  foot.  There is one  part of PL  92-500 that places what is in my
opinion the mandatory responsibility upon the  EPA and it says that before granting a new
source permit the EPA must develop an environmental impact statement.   So I have two
questions, how many new source permits has the EPA granted, and how many environ-
mental  impact statements have they written?

Harless Benthul:   In  Region VI,  we have granted,  I want to say a handful of new source
permits and by that I think and I really honestly don't know, I think the number is on the
order of maybe 10 or perhaps less or  perhaps a few more.   How many impact statements
have we prepared? I  don't believe we have published any impact statements Mr. Beychok.
We have asked for and have received environmental assessments in at least some  of those.
In one or two cases that come  to mind, we have received an impact statement for review
which has been prepared before  the facility  owners  got to us because they had to get a
federal  license or permit earlier in the game.  Now if I may, let me add one other point.
The requirement is as  I recall not that self executing that automatically-if it's a  new
source, there must be an impact statement.  As I recall 511(c),  it says that only  two
things are "major federal actions" under this law, one of which is issuance of new source
permits. Well given that it is a major federal action because it is a new source permit
then there may or may not be a requirement  for an  impact statement depending on the
various criteria that are contained in NEPA and in the cases that  have developed under
NEPA.  Does that answer your question?

Milton Beychok:    Well  I understand that you could have a  negative declaration and get
around the EIS and  that is what you are really saying that all it required was that NEPA

-------
84

 apply.  Are you saying then that out of the ten new source permits, if there were no
 impact statements written,  that in each case a negative declaration was filed.  And
 simply telling me that you asked for one doesn't answer the question because it is EPA
 who must prepare that statement.

 Harless Benthul:   Your latter statement is correct.  We do have responsibility for pre-
 paring the  impact statement.  We have asked for and received the assessments in those
 cases.  What I  was trying to say to you was in at least some of those, and don't hold me
 to the ten, that is a matter of memory and I honestly don't know if there were ten or less
 or more, but in some of those, by the time the people  got to us,  there had already been
 an impact statement prepared as a result of their dealing with some other agency,
 before they got to us, and some other agency had to do something like granting a license
 of one kind or another which they felt like required an impact statement.

 Milton Beychok:   Just one further question, does PL 92-500 spell  out any sanctions
 against the EPA, should they not follow Section 511 properly?

 Harless Benthul:   Well we are back to 505; if need be.  The sanctions I think are there
 in abundance under NEPA,  if we take a major federal action significantly affecting the
 human environment and don't prepare the 102(2)C statement its "back to go" for us.
 Whether 505(a) of the Water Pollution Control Act applies I honestly don't know.  I
 would have to give some consideration to whether or to what extent preparing an impact
 statement is discretionary and  again,  horseback, off the top of my head, I would say
 there is a good likelihood it is not discretionary,  that  if a court found that we issued a
 new source permit to a facility which was a major federal action significantly affecting
 the human  environment and we hadn't prepared an impact statement, I would say the
 judge would say go back and prepare  it.

 Frank Manning:  You have mentioned many times in your talks when a violation occurs.
 Is this always cut and dried, doesn't a violation occurring depend upon a measurement
 and analysis?  Don't you find  cases where one analyst might say yes it's over a  guideline,
 another might say it's under, in sort of a  case  like this. One, do these kind of cases
 occur and if they do occur, what do you  do then?

 Harless Benthul:   The answer is that  yes both on the self-reporting data and as a result
 of our own inspections we turn up apparent violations which are to use an electronics
 term "in the noise level", either because it's such a small percentage over  the  limit or
 because  it's within the accuracy of the analytical technique or in another area  if a
 schedule of compliance  date is missed by a week or two weeks or a month,  technically
 that is a violation.  As a practical matter,  going back to what 1 said earlier, we try not
 to issue administrative orders on relatively trivial violations and we hopefully are not
 going to send any bad cases to the U.S. attorney.  So again as a practical  matter there
 is some discretion exercised.  Does that answer your question?

 Umesh Mathur:    I noticed in  your presentation that you stayed away from the whole
 area of pretreatment,  particularly for toxic materials.  Obviously the NPDES permit
 system applies only to a direct discharger to the navigatable waters.  And so if somebody

-------
                                                                                  85


was to say,  Well if 1 am going to have to discharge to the river then obviously I am going
to have to get a permit from EPA, so why don't I  just discharge  into the sanitary sewer
system because there is no regulation that covers that, except EPA's 40CFRI28 which is
really saying nothing,  because 40CFRI28 does not specify concentration levels or any of
the mass loadings permissible per unit of production, and so this area has  been very grey
and so what I would like to point out here  is that the construction grants program under
Title 2 has been running around all over the  country requiring municipalities that are
going  to get federal funds for  building facilities to pass ordinances that will supposedly
control these pollutants.  Now the problem of that is that there is no uniformity in  these
various and sundry ordinances that have  emerged and as  a result of that nobody in industry
that is using sanitary sewers for discharging what are called incompatible or toxic
pollutants depending upon what you choose, has any idea as to what those numbers are
going  to say eventually when  and if EPA comes around to defining pretreatment standards
for these materials.  Now the difficulty  that I am  trying  to point out here is very real
because once an industry was  given an ordinance which  stayed in effect for two years, if
EPA took two years to  determine what pretreatment they  should require, and  this industry
went  in and built pollution control facilities to meet both standards that have been
established  on a completely arbitrary manner in the face of any definitive guidance from
EPA,  then for the municipality to go back  to the industry and say look now that EPA has
come out with a new set of numbers you  are going to have to shoot for this new set  of
numbers;  I feel is going to cause a management problem for the municipalities involved.
I  was wondering why the construction grant problem cannot be restrained and told not  to
require facilities that are going to require  federal funds  to necessarily have the industrial
waste  pretreatment ordinance  in the face of the fact that EPA itself is under suit to
produce those pretreatment guidelines and  they have failed to do so for I think they are
already two years too  late, and they have  still not produced  and so what is the point of
forcing municipalities  to do it.

Harless Benthul:   Well thank you for raising a very pertinent point in this whole dis-
cussion"!I can't very reliably answer your question as it relates directly to the con-
struction grant program. Hopefully, I can give you some reassurance with respect to the
rest of it.  The pretreatment regulations  generally required that as of the end of the three
year period which I think about upon us  in the order of a year or so, contributors of
incompatible waste or  waste which would pass through or interfere with the operation of
the treatment works, have some pretreatment satisfactory to prevent that happening
implemented.  Now you are quite correct that that is not implemented is a general  rule or
any specific basis,  that is in terms of limitations in either concentrations or flow or pounds
per day from the industrial contributor to the municipal treatment facility.  Nevertheless
the regulation is there, it has been on the  books for a period long enough that we feel
like it is about ready to be implementing.  Coincidentally, we had a crew investigating
a municipality in an industrial facility last week which fits this circumstance,  and  there
very well may be some enforcement activity  resulting from that.  It doesn't by the way
concern anybody in this room. It is not  a petroleum or refinery operation.  The rest of
the reassurance  is,  I hope, that is something I mentioned earlier in my talk,  and that is
that category of industry permit applicants which has some attributes of a municipality
under the Act,  that is  they are a creature of statute under state law, they have bond-
issuing powers and the  like, but treat totally if not largely,  industrial waste.  We are

-------
86

 permitting those applicants on the basis that the contributors of that waste is a party to
 that permit and is therefore directly responsible to us for a violation of that permit. My
 answer to you is not specific  as it relates to the grant program.  What it is, is that we
 think we have some  enforcement  power on  pretreatment by virtue of the regulation which
 you mentioned and by virtue  of the reference in Section 309(a)3 to violations of that
 portion of the statute that requires pretreatment. So I would have to agree that pretreat-
 ment has probably not been implemented on a specific uniform basis to the degree that
 you would like or that EPA would like, but we  do recognize that there is a problem and
 we try to deal with it to some degree at least.

 Pete Hoffman:    The question I have deals with those permits which have not been given
 by the EPA,  on applications  that were filed say back in the original Corps of Engineers
 which I think was July  1971 . If  during the period of time in which this permit is pending
 conditions change significantly,  on  that discharge, what  is the burden upon the operator
 to file a revised permit or revised application on changing conditions on a permit that
 has never been issued in the  first place and secondly,  is it not true what you were saying
 that if a citizen wants to they could file suit and force the EPA to fine that company for
 not having a permit?

 Harless Benthul:   I will answer your last question first.  Yes it is possible.  Again the
 citizen has to prove that or has to show, make  a prima facie showing that the duty
 involved is not discretionary  and that we haven't performed it.  You are describing a
 facility for which there was an application filed under the Refuse Act Permit Program
 and which has never been permitted?

 Pete Hoffman:   Yes

 Harless Benthul:   I didn't think  there were any of those left around Mr.  Hoffman.
 Alright, you filed the  application timely,  I assume,  and you say  there have been some
 changed conditions.  Have you notified us of those changed  conditions?  Therefore, if
 we were to proceed  to issue a permit tomorrow, we would be issuing a permit based on
 facts which are not current.  Would that be a fair assumption?

 Pete Hoffman:   Yes,  Sir

 Harless Benthul:   Mr. Hoffman,  I would have  sworn that we have  issued permits to all
 your facilities, but obviously we haven't.  I  would say that, and had very good
 experience with you in doing it I would add, if there have been changed circumstances
 i think  I would make a record of  it if nothing else by letter to the Regional Office and
 at least by doing  so you  have put us on notice that the facts are different than we
 apparently think they are.  Someone else was going  to help me and far be it from me to
 turn down any help.  Whoever was going to Mr. Hoffman I would  like a little expansion
 on  that last thinking.

 Bob Funk:   We have a similar situation to Mr. Hoffman.  Being  specific we  have a
 number of offshore production facilities which have  permits filed which have  never
 received a permit from Region VI and these kind of facilities are subject to change as

-------
                                                                                  87
time goes on due to the nature of the discharge.  You mentioned a letter to inform EPA
of what the change is, what level of change would trigger a  letter, this is our real
problem.

Harless Benthul;   While I doubt you are ready for this, what I am referring to is a major
change, significant change and  those are only words, I  realize. You are  talking about
offshore facilities, you are absolutely right you haven't got any permits for those and
Mr. Hoffman are you talking about minor industrial facilities, not what we would probably
call a major permit? You  are right there is a bunch of applications lying  around for
minor, what we have in the past called minor discharges,  and they have not been issued.
Several have, on the order of 1500-2000 in Region VI total.   We will get to you sooner
or later and again, if there has  been a change in circumstances, again for the like of a
better phrase, major, significant and I know that is not  much help, but those you should
notify us.

     I am  delighted to hear it and thank you.

Bob Funk:   During your discussion you cited a court case limiting penalties for  violations
to $10,000 per day regardless of the number of violations and you mentioned that you felt
like there would be further challenges to this interpretation.   Can  I draw the conclusion
that Region VI  would contemplate requesting penalties above $10,000 a day for  multiple
violations?

Harless Benthul:  Yes, you can draw from that we think  we will encounter some cases
where it is appropriate to ask for a penalty, conceivably $10,000 per day for more than
one violation which  results in conceivably  $20-$30-$40,000  per day violation.  I don't
believe we have done  that yet.  I don't think we have asked  for more than $10,000 per
day,  but it has been because we haven't run across a case where we think it fits. I  do
believe though that there will be cases where it is appropriate even within the language
of the judge in that Detrex case to ask for  penalties, multiple penalties, for multiple
violations on the same day.

Morris Wiley:   I have a question I think that concerns  many of the audience here about
the pretreatment requirement. EPA in publishing the guidelines for municipal secondary
treatment has included fewer parameters than they have  in some of the other industrial
guidelines. For example,  COD  is not included in municipal  secondary treatment but it is
included in the refinery guidelines.  Now  if the refinery or other industrial plant dis-
charged a pollutant to a municipal treatment plant which is required by  its guidelines to
treat for implementation of the 77 standards, but is not included in the municipal
monitoring program,  what would be the view of the pretreatment requirements which might
be required of that discharger?

Harless Benthul:   Sounds like it might be a case where  it would be in order to put a COD
limit on the municipal  permit.

Morris Wiley:    Is that being done at all?

-------
Harless Benthul:   It is being done in those, that class of cases I mentioned earlier which
are sort of quasi municipal type permits.  I don't know that it is being done in ordinary
municipal permits.  We have a case pending in one of the states in this Region,  where
that is precisely the issue. We have a court decree from a couple of years back that
imposed a requirement on a municipality and us as parties to  the settlement, that there
will be certain pretreatments of metals and influence  limits on metals coming into the
treatment plant and it  also is a requirement that there be effluent limits on those same
metals which, of course,  would be significantly above what you normally see in a
municipal permit as you are aware. And that one is still up in the air, I don't know
where it is going to wind  up.

Morris Wiley:   Metals would be a good case in point because the refinery effluent
guidelines do place a  limitation  on chromium and it is well known that this is removed
by either a municipal or industrial primary and secondary treatment and yet I don't
think there is any basis for evaluating what chromium  limit municipalities should dis-
charge in a permit.

Harless Benthul:   Well I  wouldn't place the greatest  emphasis on the fact that one set
of guidelines requires controls of COD and chrome and lead and another one might not.
Those as you  undoubtedly know, vary from sub-category  to sub-category.  For instance,
one of the organic chemicals or may be all of them for instance, don't have a BPT
limitation for COD, it doesn't come up until  1983. Well I personally don't know  about
that,  but nevertheless  that is what the  regulation says.

Morris Wiley:   If the COD limitation were placed on a  municipal  discharge and they
received discharges from both the refinery and the organic chemicals plant then how
would that be resolved to determine who was responsible  for what pretreatment?

Harless Benthul:   The total COD load  coming from the municipal  treatment plant
couldn't be any more  than what would  be BPT for those two industrial facilities.  So how
you translate that back up through the  municipal  treatment plant, back to the effluent
from the two  industrial facilities, I am to many years away from engineering to know if
I  ever did and I  probably didn't.  Sorry.

Bob Wortman;   Isn't  the  intent of the  law where you  have an industry discharging  into
a municipal system the actual intent I would think is that any discharge in the system
should not cause any harm to the  sewer  or harm to any of the biological treatment  pro-
cesses.  And  yet the effluent from the sewage treatment plants should meet stream
quality standards and  I would think if you meet the intent of  the law this would be  the
essential thing to do regardless of all of the other parameters.

Harless Benthul:   Well I  go along with you up to the point where you said it would be
sufficient  if you met the stream standards, remember we have technology standards which
would apply to  the municipality  for starters.  If you start backing away from placing
requirements  on however many industrial dischargers there are to the municipal  facility,
if you'd start  backing  away much from  the technological  standard that they would  other-
wise be required to meet,  then you are  placing them at a competitive advantage with

-------
                                                                                 89
the people who are discharging and treating their own waste and just that advantage
might or might not be offset by the cost recovery ordinance under the construction grant,
assuming  there is one.

Bob Wortman;  But I thought  that was one of the concerns to encourage more central-
ization of treatment processes  and to me there should be some built in advantage to an
industry going to the city,  otherwise he  is not going to go and you are going to have
more permits, more administrative expenses and everything. In other words there are a
lot of things to consider here.

Harless Benthul:   You are right there are and one of them is not placing people at
relative advantages or disadvantages when they should  otherwise be  similarly situated.

BIOGRAPHY

     Harless Benthul holds a B.A. from Texas A&M
University and a Bachelor of Law degree from S.M.U.
in 1964.   He has vast experience in the  engineering
field, seven years total  and has been in  the  private
law practice four years and three years in  the
practice of corporate law.  His present title is Chief
of the Enforcement Legal Branch in Environmental
Protection Agency Region VI  in Dallas.

-------
90

-------
                 "MEASURING THE PARAMETERS SPECIFIED IN  PERMITS"
                                       F.  T. Weiss
                               Shell Development Company
                                     Houston,  Texas

                                       ABSTRACT
       Much effort has been expended in recent years to develop and evaluate methods for
the determination of wastewater parameters.  It is  important to review the validity of the
analytical methods and to determine that they relate to the ecology of the receiving waters.
The detailed procedures of the methods must be carefully standardized and cooperatively
tested on samples representing actual effluents.

INTRODUCTION

       In order to establish and enforce environmental regulations it is necessary that there
be a reliable, accurate and meaningful measuring  system.   The measuring system is com-
posed of a complex interrelation of several factors.  First, of course, are the test methods
themselves.   These should be chosed to determine those components and properties of the
effluents for which there is a need for measurement.   The methods must  be carefully
examined to determine the significance of procedural operations and standardized by inter-
laboratory comparisons.   Precision of the methods should be determined from interlaboratory
testing and accuracy determined, if possible, from analysis of Standard Reference Materials.
Well established procedures must be set up and  followed to provide for proper sampling  and
to ensure stability of the samples prior to actual analysis.   Analytical chemists should
always be a part of the overall planning process.

       A essential consideration in establishing a  meaningful measuring system  is to make
certain that the analytical data obtained on the effluents relate to the ecology  of the
receiving waters.   Factors that should be considered in assessing the health  of the receiving
waters include processes acting upon the  components of the effluent, as  it passes into the
receiving water, as evaporation, dilution, chemical reactions, and biological degradation.

ORIGIN OF THE METHODS

       The  methods specified in the permits represent various stages in the history of
analytical technology,  some quite current while others derive from an earlier period of
water analysis methodology.   Among the more  current techniques are atomic absorption
for determination of metals and specific ion  electrodes for determination of individual
ionic  species.    These powerful new techniques now permit the determination of specific
components  with a higher  degree of accuracy.   Examples of test methods which derive from
an earlier period of technology in water analysis include the determination of oil and grease
and of biochemical oxygen demand.  It is interesting  to review the history  of these  older

                                        91

-------
92

methods in regard to the significance of their present applications.

      The present methods for determination of oil and grease in water stem from a commi-
ttee review held in Los Angeles by the  City Bureau of  Engineering some years ago.  The
paper covering their work (1) was published in 1941 and used the now familiar extraction
methods for determination of grease in sewage sludges and industrial wastes.   The procedure
described initial acidification of the sample, extraction of the  grease with a solvent,
evaporation of the solvent, and weighing the residue.  The extraction procedure itself,
however,  was taken from yet an earlier  method published in 1933 by the American Petroleum
Institute (2) for the analysis of oil-field wastewaters.

      The history of the test for biochemical oxygen demand (BOD) in water analysis goes
back many more years.  The test was recommended in  1912 by the British Royal Commission
on Sewage Disposal (3).   Almost fifty years ago, in 1928,  it was possible to present a
review paper entitled "Technic and Significance of the Biochemical Oxygen Demand
Determination" (4)  and, even then, make the statement that the BOD determination had
been in use for many years.   The dilution method, taken from the 1917 issue of "Standard
Methods"  (5), is essentially the BOD method  in current use.  The data reported (4) were
taken from analysis of sewage effluents, the drainage  canal of the Chicago Sanitary District,
and of an  effluent from a corn products manufacturing plant.

      A characteristic of these older methods, and that also of  certain others being used as
test procedures, is that the results are empirical.   The data obtained  are dependent upon
the test variables themselves and do not provide compositional data.  In certain of these
methods, the parameter being measured  is defined by the methodology itself.

STANDARDIZATION GROUPS

      The long history of the analysis of water led inevitable to a confusion of methods and
interpretation of results.   To overcome  these problems, standardization groups were
organized many years ago.  A committee of the American Association for the Advancement
of Science reviewed methods for the analysis  of water and published a report (6) in 1889
which covered methods for determination of "free" and  "albuminoid" ammonia, nitrites,
nitrates, and oxygen-consuming capacity.   Further developments led the American Public
Health Association  to appoint a committee who produced in 1905 the first edition of the
now familiar "Standard Methods".  Revised and enlarged editions of  "Standard Methods"
have been produced over the years under the  combined sponsorship of the American Public
Health Association, the American Water Works Association, and the Water Pollution Control
Federation (7).

      The  American Society for Testing and Materials (ASTM)  made a review in  1925 and
established that there was a need for cooperative testing in the sampling and analysis of
industrial  waters (8).   As a result,  ASTM Committee  D-19, on water analysis, was
established in 1933 with eleven members and has grown in scope and membership, parti-
cularly in the past ten  years.   Membership now is over 500 and the number of methods
currently printed are 135 (9).  Active groups meet twice yearly to review and introduce
methods covering such topics as biological monitoring,  sampling techniques, radiochemical

-------
                                                                                   93

analysis, determination of inorganic and organic components, analysis of brackish water,
and the identification of waterborne oils.  An essential part of the ASTM process is inter-
laboratory comparison of results on practical samples in round-robin fashion to determine
precision and reliability of the methods.   Such testing goes on during the course of the
year and results are reviewed at the semi-annual meetings.

        The U.S.  Environmental Protection Agency (EPA) has developed a valuable Manual
covering "Methods for  Chemical Analysis of Water and Wastes"  (10) now in its second
edition.  This  set of methods was collected, reviewed and published by the Method Devel-
opment and Quality Assurance Research Laboratory at the National Environmental Research
Center at Cincinnati, Ohio.  Assistance was provided by other governmental laboratories,
both state and federel.   The Manual is a basic reference for monitoring water and wastes
in compliance with the  Federal Water Pollution Control Act.  Although other test procedures
may be used, it is stated in  the Manual that the methods therein  described will be used  by
the EPA in determining  compliance with  applicable water and effluent standards.

        European standardization groups are also active in promulgating methods for the
analysis of wastewaters, but their publications do not directly impinge upon the practice in
the United States on measuring wastewater parameters.  Exceptions,  of course, will be with
the development of new analytical technology or in topics which have a large geographical
impact.  For instance,  the  characterization of waterborne oils has been the subject of
activity not only by the ASTM and others in the United States, but also by the Institute of
Petroleum in England (11).

FEDERAL GUIDELINES  FOR ESTABLISHING TEST PROCEDURES  FOR THE ANALYSIS OF
POLLUTANTS

        The specific approved  test procedures to be used for measuring the parameters
specified in  permits derive from Part 136, Title 40,  Code of Federal Regulations.  Section
136.3 identifies the test procedures which must be used in any permit application pursuant
to sections of the Federal Water Pollution Control Act.   Updated lists of approved methods
in the Code  of  Federal Regulations are published from time to time in the Federal Register.
The most recent proposed amendments to  the test procedures for aqueous samples were pub-
lished in the Federal Register issue of June 9,  1975 (12).

        These proposed amendments covered a great deal of ground since they include not
only correction of minor errors but also allowed new techniques and added a considerable
number of additional parameters for possible measurement.  The  corresponding proposed
table of approved test procedures from  the Federal Register of October 16, 1973 (pages
28759 and 28760) lists 71 parameters for measurement.  The 1975 table lists 115 parameters.
Although the text briefly reviews the fact that additional parameters have been added,  the
magnitude of the additions is obscured  by the way in which the 1975 table was constructed.
In the effort to  keep  the parameter listings of the 1973 table unchanged, the 44 new para-
meters have  been  introduced as subheadings.   For instance, the  new listing of Rhodium is
introduced as 35 (b) under Potassium  (35) and Potassium - dissolved (35a).   Ruthenium Is
then introduced as 35 (c).

        The recent issue of  the Federal Register is an example of a way in which test

-------
 94

procedures have been introduced without undergoing a thorough standardization process and
 interlaboratory review.   For example, a number of the new test procedures which have
 been introduced are directed to  the determination of the dissolved portion of metal com-
 ponents, generally using atomic absorption for final measurement.  The procedures differ-
 entiate between the portion of the metallic constituents which will pass through a 0.45
micron filter and those which will  be retained.  In nearly all cases the reference is to a
 single page of the EPA Methods  Manual (10).  This Manual does not provide sufficient
 procedural details for method standardization.   In fact, a note is included  in the EPA
Methods Manual in the general method detailing techniques for atomic absorption which
 suggests the need for (unspecified) special precautions for stabilization of certain types of
 metal containing samples.   Large  differences can be  expected in behavior between metal
 components of such diverse properties as sodium, for example, and titanium.  Because of
 these factors new sets of proceudres should always be  subjected to careful and detailed
 review and proper standardization  and validation before being applied to the permit process.
 This should include crosstesting of  samples amongst a group of laboratories for intercompo-
 sition and development of data on  the reliability of the procedures.

        Although the current Federal Register lists 115 parameters, in  actual practice con-
 siderably fewer are used for control of petroleum refinery wastewaters.   Table I lists twelve
 parameters which are most commonly required with a brief statement of the method recom-
 mended.

 REQUIREMENTS FOR SATISFACTORY TRACE ANALYSIS

        Measurement of certain components specified  in the permits is a practical example
 of trace analysis which is generally considered to be the determination of a component or
 components in a sample vastly diluted by other materials.    Many methods have been used
 for trace analysis. (13),  (14)  During 1972 an intercompany project was initiated to develop
 techniques for measuring trace concentrations of metals in oil  (14). Although their analyti-
 cal medium  was different from water, their detailed studies of the principles of trace analysis
 apply to the determination  of trace components in water.

        Sampling.   It  is well acknowledged that analytical  results can be no better than the
sampling and preservation techniques employed.  It cannot be overstressed  that samples
must accurately represent the conditions existing at the point of sampling and that samples
must be properly preserved  in that  condition until the analysis is made.   Although each of
the reference compilations  provides guidelines on sampling, in actual practice good sampl-
ing techniques are not always employed.  For accurate results in trace analysis, clean
sampling containers of the proper material, caps, thorough sampling procedures,  and proper
preservation techniques must be  carefully  established  and  rigorously follwed in actual  use.
If these points are not properly observed,  the results obtained can be meaningless.   It
must be emphasized that anything which contacts the  sample,  whether apparatus, storage,
or  shipping containers and closures,  the reagents or the atmosphere can serve as a potenH
ial source of elemental contamination or provide a sink for loss of some components, such
as  by adsorption.

        Storage and Stability.  The instability of dilute aqueous solutions has been widely
observed. (10, 14)  The best rule is to analyze samples as soon as possible after they have

-------
                                                                                   95

been taken.   Changes that take place in a sample can be due to chemical or biological
reactions.   Metal cations may precipitate, form complexes or adsorb onto surfaces.  Organ-
ics7 including hydrocarbons7may undergo biochemical oxidation.   A general rule to
increase the stability of aqueous samples is to acidfy approximately to a pH of 2.  HgClj
can be added for bacterial inhibition.  Refrigeration at  temperatures near freezing or
below can  be helpful in preserving a sample which cannot be analyzed immediately.

        Standards.   The best means to evaluate laboratory performance is  the analysis of a
standard sample in which  the composition of the determined component is  known exactly.
A series of quality control samples is available on request from  the Environmental Protection
Agency (15) for the  determination of many components including certain trace metals,
ammonia, nitrate, phosphate, pH, and so on.   These have  been widely used to provide
checks on reagents and laboratory procedures.   More exact primary standards are the
Standard Reference Materials being prepared by the U.S.  National Bureau of Standards
for trace metals in water.   The composition of NBS Standard Reference have been certified
by  at least two independent analyses.   Suitable Standard Reference Materials are the best
means of evaluating the entire  proceudre and to establish its accuracy.  In the absence of
a standard  sample,  "spiked" samples prepared by adding  known amounts of a constituent
to an effluent can be used.

SIGNIFICANCE OF THE  DATA

        It is important to divide the test results into those which provide compositional data
and those which are empirical.   In the first category, compositional, are included those
methods for determination of ionic constituents such as dissolved metals, nitrogen compounds
such as  ammonia, nitrite or nitrate, and sulfide ion.  Since these are all  defined compon-
ents and determinable by  at least two independent analytical methods, their concentrations
can be determined with some degree of accuracy and  the effect of interferences investigated
on  the precision and accuracy of the results.  Interlaboratory comparisons of methods and
data, preferably, by several techniques can establish the validity,  accuracy, and precision
of the data.

        On the other hand, those methods which provide empirical results, including the
determinations for oil and grease and  biochemical oxygen demand,  defy a definition of
accuracy.   For the  empirical methods the best that can be done is to standardize the
procedural  descriptions and then determine precision by interlaboratory round-robin testing
under exactly defined conditions. It  is not possible to run any  of the empirical tests by
two sets of procedures to determine accuracy since the results are defined by the procedures
themselves.

        Relationship  to Ecology of Receiving Waters.   The prime consideration in  establish-
ing effluent standards should be to relate the analytical data of the effluent wastewaters to
the ecology of the receiving waters.  The factors which should be considered in the
receiving water are  toxicity (both acute and long-term) to the flora and fauna, depletion
of oxygen, odor, and detrimental  changes in the esthetics of the receiving waters  and the
adjacent land areas.   Factors that should be considered  in assessing the effects of effluent
composition on the receiving waters include such processes as evaporation, dilution, air-
water interchange, chemical and photochemical reactions,  and biological degradation. (16)

-------
96

The relationship between  laboratory bioassays and actual results in the field are not necessarily
straightforward.  For example, the chemical form in which a trace metal is released or the
form it adopts after release is highly significant in its biological  impact (17).   There  is an
increasing body of evidence  indicating that there are natural processes operating to reduce
both the concentration and toxicity of trace metals dissolved in water.   In order for a
heavy metal to be toxic it must be  in the ionic state. (18,19,20)   In  most natural waters
much of the free metal ions would probably be rapidly bound to organic substances, nat-
urally present in the water,  decreasing the relative percentage of the ionic species   There
is indirect evidence that organically chelated heavy metals in  aqueous solutions do not have
as great an effect upon organisms as do solutions of the metal salts.  (20)  This could be due
either to the fact that  the organo-metallic complex is too  bulky to enter a biological  system
or it could be due to the  lack of availability of the metal  for reaction with enzymes within
the cells.

        Precision of Data.   From the standpoint of quality control and agreement with
regulatory standards it is important to have an understanding of the significance of precision
of the results obtained by the various methods.   Both ASTM Committee  D-19 (21) and the
Methods Development  and Quality Assurace Research Laboratory of the  Environmental
Protection Agency  (22,23,24,25)  have practices for evaluating analytical methods for
water.   After a method has been sufficiently  investigated to determine the significant
variables and establish their  control, it is possible to organize  an interlaboratory, round-
robin testing program.   A written copy of the method being investigated plus the necessary
samples for each test are supplied  to each laboratory.   From the interlaboratory tests, it is
possible to derive terms for method precision, which is  usually  done by calculating standard
deviations.  Standard deviation is the measure of the dispersion of the data around their
average,  expressed as  the square root of the quantity obtained  by summing the squares of the
deviations from  the average of the  results and dividing  by  the number  of observations minus
one.   The standard deviation is generally given as a measure of the precision of methods
which have been subjected to comparative evaluation.   The next factor to consider is the
confidence level of the data obtained at two different laboratories. As Youden (26) points
out, it  is well known that a reasonable number of measurements must be made to give
confidence in the results.  A number of factors enter into establishing confidence limits on
data sets.   At a 95 percent confidence level  an approximate rule (26) is that,  for sets of
4 or 5 determinations,  average results between two laboratories should not differ by more
than approximately 3 times the standard deviation.  When analytical  methods are used at
or near  their lower level of sensitivity, there  is a greater tendency to error and  loss of
reproducibility.

PROCESS-TYPE INSTRUMENTATION

        The use of continuous monitors to determine effluent parameters on flowing systems
is particularly valuable, if it can  be made effectively.   Such  continuous measurements
provide for the fully automatic regulation and control of certain  parameters and avoid the
delay and error in sample taking and handling.   It is always necessary to carefully review
the operation of a continuous monitor under field conditions for a sufficient period of  time
to make certain its  behavior  is satisfactory and that its  response is to the parameter to be
measured and not to an artifact in the system.

-------
                                                                                  97

        Continuous instruments are available and have been commonly used for many years
 for measurement of pH by use of the glass electrode (ASTM D 1293).  Commercial process-
 type water quality analyzers are currently becoming available for a number of other com-
 ponents.  These include the use of specific ion electrodes for ammonia and sulfide,  and
 ultraviolet spectroscopic instruments for phenols.   An excellent detailed coverage of such
 instruments was recently published by  the Lawrence Laboratory of the University of Califor-
 nia (27).

 EXAMPLES OF CRITICAL ANALYSES

        Oil and Grease  Determination.   A considerable number of methods have been
 applied over many years for determination of "oil and grease" or extractable materials in
 aqueous samples.   Table II provides a summary of the principal features of some twenty
 such methods.   For each of these, more  detail is given in a recent report (28) prepared
 for the American Petroleum Institute.   The kind of  results that may  be obtained from use of
 this large array of empirical procedures may be quite different.   For instance, Methods 4
 and 10 both involve initial  extraction  with  Freon 113.  Method 4 utilizes a gravimetric
 determination following solvent evaporation, which causes also some evaporation of light
 hydrocarbons.   Method 10  utilizes an intrared spectrophotometric measurement and  does
 not involve evaporation.  Consequently, it can  be  anticipated that comparative measure-
 ments, v/here light hydrocarbon components are present in the sample, can be higher with
 Method 10.   It is interesting that the  Federal Register of June 9, 1975 (12) lists both the
 gravimetric and infrared methods as alternative procedures without commenting on differences
 that could be anticipated in results.

        Another area of concern is the relative solvent power of common solvents for  petro-
 leum  constituents.   The components of wastewater are a complex mixture of compounds of
 different nature and are not all  equally dissolved by different solvents.   For instance,
 Freon  appears to extract less material of an asphaltic nature and a "rag" or undissolved oily
 film has  sometimes  been  noticed.

        Another significant  difference  in results can be obtained when an absorbent is used
 to separate polar from nonpolar materials.  Methods of this type can provide for the deter-
 mination of hydrocarbons in effluent waters.  A review  of Table II shows that there are
 several available methods utilizing adsorbents to separate polar materials from  hydrocarbons
 to allow the more specific determination of the latter.   These involve different applications
 of adsorbents such as the use of alumina (Method 5), Florisil (Methods 11 and 17) or  silica
 gel (Method 13).  An extensive round-robin test was recently managed by the Robert S.  Kerr
 Laboratory of the Environmental Protection Agency, Ada, Oklahoma to  ascertain the validity
 and reproducibility of Method 13.  It  is generally the experience that the efficiency of
 separation of polar organics from aromatic hydrocarbons  may depend to a considerable extent
 on the polarity of the solvent, the activity of the adsorbent and  the adsorbent/sample ratio.
 A study of this type has been  recently published (29) and indicates that the variables should
 be carefully established with  such methods for accurage  and meaningful  results.

        Continuous monitors have been considered for determination of small quantities of
 oil in  effluent  waters.   The principles and experience with several  of these techniques are
 indicated under Method No.  21.   Most of those methods involve interaction of the com-
ponents of the  sample with  light of several wavelengths.  It is important that the spectral

-------
 98

region (that is, ultraviolet,  visible,  or infrared) relate properly to the components of the
samples.  Although these techniques  do show promise, several have shown limitations in
field use in that they may be somewhat fragile and some tend to undergo fogging in the
optical path.  It is necessary that the optical instruments properly discriminate between  oil
and interferences such as suspended solids.

        ASTM Committee D-19 is currently conducting interlaboratory comparisons with  many
participants of several of the methods of oil and grease analysis.   Samples being circulated
are practical effluents and contain both petroleum and non-petroleum constituents.

        Chemical Oxygen Demand.  The dichromate oxidation method for  chemical  oxygen
demand is another empirical test with a history  of use  of some years (30.   As is pointed  out
in "Standard Methods" (7) it is important to follow a standardized technique since only a
part of the organic matter is oxidized.  This is particularly important with nitrogen com-
pounds since the extent of their reaction can be variable, depending upon structure, as  can
be seen from Table 111.   The data  in Table III are taken from references being reviewed for
consideration in the ASTM D-19 Committee studies on the method for Chemical Oxygen
Demand.   Because of the way  the test is performed, volatile, nonpolar components, such
as benzene and toluene can  be partially lost.   The limited reaction of many nitrogen com-
pounds depend upon the structure of the compound.   For instance, pyridine with one heter-
ocyclic ring  is essentially nonreactive while quinoline undergoes extensive reaction,  no
doubt due to initial attack on the homocyclic ring.

        Biochemical Oxygen Demand  (BOD).  As pointed out earlier, this empirical method
has a  long history of use in the analysis of wastewaters.  The test has significance as an  index
to the  susceptibility of the components of wastewater to biodegradation.   The test involves
measurement of the amount of oxygen consumed in a 5 day period after the sample has been
inoculated with sewage bacteria.  Despite its long history of utility, the  BOD method has
been recognized to have many  limitations  (31).   The  limitations include such factors as the
differential rates of biochemical oxidation of various compounds,  including the effects of
nitrifying bacteria which convert nitrogen compounds  to higher forms.   The rates of reaction
may vary markedly with different sources of inocula.   "Standard Methods" (7) points out that
the BOD test is of limited value in measuring the  actual oxygen demand of surface waters.
The extrapolation of test results to actual stream oxygen demand is highly  questionable since
the laboratory environment does not reproduce actual  field conditions, particularly as
related to temperature,  sunlight, biological population,  water movements, and oxygen
concentrations.

        ASTM Committee D-19 came out with a similar assessment of the BOD method.  The
action  of the ASTM was to discontinue without  replacement, ASTM D 2329, "Test for Bio-
chemical Exygen Demand of Industrial Water and  Wastewaters". (9)  The  ASTM group con-
cluded that the BOD concept is important in determining water quality but that the laboratory
bottle  incubation procedure  typified  by the usual  BOD test is ill defined,  producing data
which  are limited.  The ASTM  recommends several alternate tests which have been  better
standardized to assess pollution potential.   These tests include those for Total Organic
Carbon (ASTM D 2579)  (9) and Total Oxygen Demand (ASTM D 3250) (9).  The subject  is
still under close review by this ASTM group.

-------
                                                                                  99

       Cyanide Determinations.  The issue which has become of concern in the determina-
tion of cyanide in effluents is the differentiation between simple cyanide and complexed
cyanides.  In common practice simple or "free"  cyanide is defined as the CN  ion or the
HCN molecule.  Complex cyanide is taken to mean any molecular composition in which
the CN group is bound to a metal through coordinate bonds.   Examples of complex cyanides
are those of iron: Fe (CN),  and Fe (CN)7 .   The practical distinction between simple
and complex  cyanides depends upon the production of HCN  under relatively mild conditions.
For example, the procedure of Roberts and Jackson (32) provides for distillation of hydrogen
cyanide, in the presence of zinc acetate,  under  reduced pressure.  Under the conditions
of the test the decomposition of ferrocyanide is completely prevented.   The cyanide in  the
distillate is determined by colorimetric pyridine-pyrasolone  procedure.

       There is currently an extensive effort in ASTM Committee D-19  involving inter-
laboratory round-robin tests with many participants examining several of the methods which
have been recommended for cyanide determination.  The aim of this testing program is to
establish the  validity and precision of the cyanide methods and to establish the significance
of interferences.

CONCLUSIONS

       1.  The large scope of the procedures for measuring the parameters of aqueous
effluents has  given rise to methods whose variables  have not all been completely under-
stood or standardized.   Results from some of these methods are dependent upon the proced-
ural steps employed.

       2.  It is important to review and decide on  what components in effluent waters should
be measured.   The  purposes and definitions of  the measurements must be tied to real  needs
relating to  the  ecology of the receiving waters.

       All methods must be thoroughly examined to determine the significance of the
procedural  operations.   The final methods must  be carefully standardized and cooperative
testing on samples representing actual effluents.

REFERENCES

(1) Pomeroy, R. and Wakeman,  C. M. (1941), "Determination of Grease in Sewage,
    Sludge,  and Industrail Wastes", Ind.  Eng. Chemistry,  Analytical  Edition, 13,
    pages 795-801.
(2) American Petroleum Institute (1933), "Disposal of Refinery Wastes", Second Edition,
    Section I.
(3) Royal Commission on Sewage Disposal  (1912),  Eighth Report, I. "Standards and Tests
    for Sewage and Sewage Effluents Discharging into Ribers and Streams",  Cd 6464,
    HMSO,  London.
(4) Mohlman,  IF. W., Edwards, G.  P., and Swope, G.(1928),  "Technic and Significance
    of the Biochemical Oxygen Demand Determination", Ind. Eng. Chemistry,  20,  pages
    242-246.
(5) "Standard Methods for Examination of Water and Sewage" (1917), page 71.

-------
 100
 (6)  Caldwell, G. C. (1889), "A Method,  In Part,  For the Sanitary Examination of Water
     and for the Statement of  Results, Offered for General  Adoption", J. Analytical and
     Applied Chemistry, 3, page 398.
 (7)  "Standard Methods for the Examination  of Water and Wastewater"  (1971), Thirteenth
     Edition, American Public Health Association, New York.
 (8)  Clarke, F. E. (1972), "The History of Committee D-19 on  Water" in ASTM Committee
     D-19 Handbook.
 (9)  Annual  Book of ASTM Standards (1975), "Water", Part 31.
(10)  "Methods for Chemical Analysis of Water and Wastes"  (1974), Second Edition,  U. S.
     Environmental Protection Agency, National  Environmental Research Center, Cincinnati,
     Ohio.
(11) Whitham, B. T., Duckworth, D. F., Harvey, A. A.  B., Jeffrey,  P. G., and Perry,
     S.  G. (1974),  "Marine Pollution by Oil - Characterization of Pollutants, Sampling,
     Analysis, and Interpretation", Institute of Petroleum,  Great Britain.
(12)  Federal Register (1975),  Vol. 40, No.  Ill,  Monday,  June 9, pages 24535-24539.
(13)  Weiss,  F. T. (1970),  "Trace Analysis"  in "Determination of Organic Compounds:
     Methods and Procedures",  Wiley-lnterscience, pages 317-334.
(14)  Hofstader, R. A., Milner, O. I., and Runnels, J. H. (1976), "Analysis of Petroleum
     for Trace Metals", to be  published by the American Chemical Society.
(15) U.  S.  Environmental Protection Agency,  Analytical Quality Control  (1975), "Newsletter"
     Environmental Monitoring and Support  Laboratory, Cincinnati, Ohio 45268, October,
     page 6.
(16)  "Petroleum in the Marine Environment" (1975),  National Academy of Sciences,
     Washington,  D. C.
(17)  "Trace Methods" in "Marine Bioassays" (1974),  Workshop Proceedings,  Marine Techno-
     logy Society, 1730 "M"  Street N.W.,  Washington, D. C.    20036, pages  76-93.
 (18) Lerman, A. and Childs,  C. W. (1973), "Metal-Organic Complexes in  Natural Waters:
     Control of Distribution by  Thermodynamic, Kinetic, and Physcial Factors",  in "Trace
     Metals and Metal-Organic Interactions in Natural Waters," Edited by Singer,  P.  C.,
     Ann Arbor Publ., Inc., Ann Arbor, Michigan.
 (19) Morel,  F., McDuff,  R.  E., and Morgan, J. J. (1973), "Interactions and Chemostasis
     in Aquatic Chemical Systems: Role of pH, pE, Solubility,  and Complexation", In
     "Trace Metals and Metal-Organic Interactions in Natural Waters",  Edited by Singer,
     P.C., Ann Arbor Publ.,  Inc., Ann Arbor, Michigan.
(20)  Wilson, R. C.  H.  (1972), "Prediction of Copper Toxicity in Receiving  Waters", J.
     Fish.  Res. Bd.  Canada, ^9, pages 1500-1502.
(21)  ASTM D 2777-72 (1975),  "Standard Recommended Practice for Determination of
     Precision of Methods of Committee D-19 on Water", In "Annual Book of ASTM
     Standards",  Part 31,  American Society for Testing and Materials,  Philadelphia,
     Pennsylvania.
(22)  Winter, J. A.  and Clements, H. A.  (1975), "Interlaboratory Study of the Cold Vapor
     Techniques for  Total Mercury in Water",  Water Quality Parameters, ASTM Special
     Technical Publication 573, pages 566-580.
(23)  Winter, J. A.  and Midgett, M.  R.  (1969), "FWPCA Method Study I, Mineral and
     Physival Analysis, An Evaluation of the FWPCA Analytical Methods for Surface Waters",
     Analytical Quality Control Laboratory, Cincinnati, Ohio  45202.

-------
                                                                                  101

(24)  Winter, J. A. and Midgett, M. R. (1970),  "Method Study 2,  Nutrient Analyses,
     Manual Methods, An  Evaluation of Analytical Methods for Water and Wastewater",
     Environmental Protection Agency,  Analytical Quality Control  Laboratory,
     Cincinnati, Ohio  45202.
(25) Winter, J. A. (1971), "Method Study 37 Demand Analyses, An Evaluation of
     Analytical Methods for Water and Wastewater",  Environmental Protection Agency,
     Analytical Quality Control Laboratory,  Cincinnati, Ohio 45202.
(26) Youden, W.  J.  (1951), "Statistical Methods for Chemists", John Wiley  & Sons,  New
     York.
(27)  "Instrumental for Environmental Monitoring" (1975), Water, Lawrence Berkeley
     Laboratory, University of California,  Berkeley, California 94720, Second Update, June.
(28)  "Review of Analytical Methods for Determination of Oil and Grease in Produced
     Waters from Oil and Gas Extraction Industry Operations" (1975), Prepared for the
     American Petroleum Institute Committee on Environmental Conservation -  Division
     of Production, 300  Corrigan Tower, Dallas, Texas   75201.
(29) Bridie,  A. L.,  Bos, J., and Herzberg, S. (1973),  "Interferences of Non-Hydrocarbons
     in Oil-in-Water Determination," J. of the Inst. of Petroleum, 59, 570, pp.263-267,
     November.
(30)  Moore, W. A., Kroner,  R. C., and Ruchhoft, C.  C.  (1949),  "Dichromate Reflux
     Method for Determination of Oxygen Consumed", Anal. Chem., 21, pages 953-957.
(31)  Marske, D. M.  and Polkowki,  L.  B. (1972), "Evaluation of Methods for Estimating
     Biochemical Oxygen Demand Parameters", Journal Water  Pollution Control Federation,
     44, pages  1987-2000.
(32) Roberts, R. F. and Jackson, B.  (1971),  "The Determination of Small Amounts of
     Cyanide in the Presence of Ferrocyanide by Distillation Under  Reduced Pressure",
     Analyst, 96, pages 209-212.

DISCUSSION

Larry Echelbergen  I am wondering what Shell does to ensure the accuracry of its numbers.
What do you do at the individual  refineries?  Do you spike samples or periodically on a
routine basis, for example, send them to independent laboratories?  What do you do or
what do most of your refineries do to ensure accuracy of your results?

Fred Weiss:  In the Company for certain types of analyses we have for some years run our
own type round-robining for some product materials.  For effluents,  which I take your
question to be, because the transport of effluent samples is very difficult, we do run cross
comparisons with commercial  laboratories in each  of our areas.  The data  I have seen from
a number of the cross comparisons is generally fairly good.  Another thing that we have done,
although I don't know how  widely it  is done throughout the refinery laboratories, is to get
those samples from the EPA that are essentially standard  materials.   There are  a whole
series which have been made available and we have run those.

-------
  102
BIOGRAPHY
     Fred Weiss is Senior Staff Research Chemist,
Analytical  Department, Shell Development Company,
Houston, Texas. He holds degrees in Chemistry from
fhe University of California  at Los Angeles
(B.S., 1938) and from Harvard University (M.A.,
1939 and Ph.D., 1941).  He has been involved in
research with the Shell Companies since 1941.  His
special  interests have been in the detailed analysis
of organic compounds and the application of methods
to trace analysis.  In recent  years much of his work
has been devoted to environmental measurements.
   TABLE 1  PARAMETERS COMMONLY REQUIRED FOR PETROLEUM REFINERY PERMITS
             Parameters
Biochemical Oxygen Demand 5 day (BOD,)
Chemical Oxygen Demand (COD)
Oil and Grease
Total suspended solids
Total Organic Carbon (TOC)
Chromium, Total
Chromium (VI)

Cyanide
Phenols
Ammonia
Sulfide
PH
              Method
Modified Winkler or probe
Dichromate reflux
Extraction
Glass fiber filtration
Combustion-infrared
Atomic absorption or  colorimetric
Extraction and atomic absorption,
colorimetric
Distillation and colorimetric
Colorimetric
Several  techniques
Titrimetric or methylene blue
Electrometric

-------
                                                                                   TABLE II

                                                      SUMMARY OF ANALYTICAL METHODS FOR DETERMINATION OF OIL & GREASE'
TYPE


u
t-i
H O

> a
<; s
o

VOLUMETRIC
METHODS


,j
S w
| §
5 H
H §
2


SAMPLING
en
11
fe H
M M
O S
NO.

1
2
3
4
5
6
7
8

9
10
11
12
13
14
15
16
17
18
19
20


21

METHOD DESIGNATION

Chloroform Extractable Matter
Solvent Extraction
Soxhlet Extraction
APHA Oil & Grease
Hydrocarbon and Fatty Matter
Reflux Distillation Method
Petroleum Ether Extraction
Hexane Extractable Material

Flocculation - Extraction Method
Oil and Grease - Freon Extractables
Hydrocarbons in Water or Soil
Combined Methods
Petroleum Hydrocarbons
Volatile and Non-Volatile Oily Material
Ultraviolet Fluorescence
Ultraviolet Absorption
Gas Chromatography
Visible Color Method
Color Comparator

Ultraviolet
Visible
Infrared
Fluorescence
Light Scattering
REFERENCE

ASTM D-1178
ASTM-2778
EPA Manual 00550
APHA - 137
Std. Methods P. 413
ASTM D-1340
a)
ASTM D-1891

API 732-53
EPA - 00560
Concave 1/72
API 0 & G Comm
1975 EPA Study
API-733-58 5. Lit.


Concawe 111-72
Hach Handbook-1973
Champion Chem. Co.
EPA, ASTM, APHA




STATUS

Discontinued
Limited Use
Wide Use
Wide Use
Limited Use
Discontinued
Limited Use
Discontinued

Limited Use
General Use
Limited Use
Proposed
Proposed
Limited Use
Limited Use
Limited Use
European Use
General Use
Very Limited
Wide Use

Spsc if ic
Applications

CONDITIONS
OF EXTRACTION
pH 3-4
pH Variable
pH - 2 Soxhlet
Acidic
Follov Method 4
Acidic
No pH Adj.
pH-4

pH-4
pH-2
Acidic
pH < 7 > 7
Acidic
Acidic
None
Acidic
Acidic
None
None
EXTRACTANT

Chloroform
Any Solvent
Hexane or Freon-113
Freon or Pet. Ether
Freon-Hexane
Benzene, Others
Petroleum Ether
Hexane

Ether
Freon 113-IR
CCL4 (IR)
CCL4 (IR)
CCL4 (IR)
CCL4 (IR)
CCL4-Hexane (UV)
Hexane (UV)
CCL4 or Pentane (GC)
3-4 Solvents
Special Chemical
ADSORBENT

None
None
None
None
Alumina
None
None
None

Fe(OH)3Floc
None
Florisil
None
Silica Gel
None
None
None
Florisil
None
None
COSTS

Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Expensive
Moderate

Inexpensive
Expensive
Expensive
Expensive
Expensive
Expensive
Expensive
Expensive
Expensive
Inexpensive
Inexpensive
pH Control, Refrigeration, Time-Dependent, Glass Bottles


None



As Received



None

Moderate
Inexpensive
Expensive
Expensive
Moderate
•For more information, see Reference 28.
                                                                                                                                                                   O
                                                                                                                                                                   CO

-------
 104
         TABLES  BEHAVIOR OF VARIOUS TYPES OF ORGANIC MATERIAL IN
              ASTM METHOD D-1252 FOR CHEMICAL OXYGEN DEMAND

        Component                 Reactivity, Percent of Theoretical

References (a)                (1)           (2)            (3)           (4)

Aliphatic Compounds

   Acetone                   98            96          94
   Acetic Acid                              98
   Dextrose                   95
   Diethylene Glycol          -                         70
   Ethyl Acetate              -                         85
   Methyl Ethyl Ketone        -                         90

Aromatic Compounds

   Acetophenone              89
   Benzeldehyde              -                         80
   Benzene                   60            41           -
   Benzole Acid               98            -          100           100
   Dioctyl Phthalate           83
   Ortho Cresol               95-95
   Toluene                    -                         45

Nitrogen Compounds

   Acrylonitrile               48            -           44
   Adenine                   -             -           -             59
   Aniline                    80            -           74
   Butyl Amine                57
   Pyridine                    0             I                          2
   Quinoline                                                         87
   Trimethylamine             1             -
   Tryptophane                                                        87
   Uric Acid                                                          64

(a) References:
      (1) Unpublished data.
      (2) Dobbs, R. A., Williams, R. T., Anal. Chem. 35, 1064  (1963)
      (3) Buzzell,  J. C., Young, R.  H. F.,  Ryckman, 57 W., "Behavior of Organic
         Chemicals in the Aquatic Environment: Part II, Dilute Systems,"  Manufacturing
         Chemists Association,  April 1968, page 34.
      (4) Chudoba, J. and Dalesicky, J., Water Research,  V. 7, No. 5,  663, (1973).

-------
                                BANQUET

Master of Ceremony --Dr. J. Paschal Twyman
                      President
                      The University of Tulsa, Tulsa, Oklahoma

Paschal Twyman;       "I am  Paschal Twyman, and, as President of the University of
Tulsa it is my privilege to welcome all of you to this Open Forum Banquet. As you are
well aware this highly successful symposium on "Management of Petroleum Refinery Waste-
waters" is co-sponsored by the EPA, the API, the NPRA, and T.U. Just before this
banquet I  was told that T.U.  were the only initials which should be spelled out as Tulsa
University; because the others - E.P.A., A.P.L, and N.P.R.A.  - are so well known to
all  of us.

                      "As your M.C. for this banquet it is my pleasure to introduce to
you the people  at the head table.  Starting at my left, that is your right, they are:"—

Fred Pfeffer,          of the R. S.  Kerr Environmental Research Laboratory,  Ada,
                      Oklahoma.   Fred is the Project Officer for  the EPA Grant
                      sponsoring this Open Forum.

Milton Beychok,       Consulting Engineer, Irvine, California. Our distinguished key-
                      note speaker.

Herb Bruck,           Technical Director, N.P.R.A. Herb is representing the NPRA,
                      who are, of course, one of the sponsors of this symposium.

Nick Gammelgard,    Senior Vice President, Environmental Affairs, A.P.I.  Nick  is our
                      banquet speaker and is also representing the A.P.I.,  the other
                      industrial sponsor of this symposium. "

"Now on the other side starting at my right hand  is:" —

Bill Galegar,          Director, R.  S.  Kerr Environmental Research Laboratory, Ada,
                      Oklahoma.   Bill, as you remember, welcomed us this morning.

Nancy Manning,       Research, biochemist at St.  Francis Hospital  in  Tulsa.   Dr. Nancy
                      Manning is also Mrs. Frank Manning.

Frank Manning,        Professor of Chemical Engineering at Tulsa  University.  Frank is
                     Project Director for the EPA Grant financing this Open Forum."

"Frank Manning will now introduce  the banquet speaker."


Frank Manning  then introduced Mr.  Gammelgard.  Mr.  Gammelgard's address and
biography follow on page 107.

                                      105

-------
106

BIOGRAPHY         J. Paschal Twyman

     Dr. Twyman holds the following degrees
from The University of Missouri at Kansas City:
B.A. in Sociology; M.A. in Educational
Administration; and a Ph.D. in Education.  Dr.
Twyman served as a Professor of Education and
Associate  Director of the Research Foundation
at Oklahoma State University and as Director
of Research, Assistant to the Chancellor and
Professor of Education at University of
Missouri at St. Louis/ before joining Tulsa
University in 1967, as Vice President for
Research and Development.  Dr.  Twyman be-
came President of Tulsa  University in 1968.

     Dr. Twyman has extensive research
experience and has published  extensively on
higher education.  Dr. Twyman is also most
active in numerous educational,  charitable and
community-service organizations.

-------
                                  BANQUET ADDRESS

                         "LET'S CHANGE COURSE AND SPEED"

                           P-  N. Gammelgard, Vice President
                    American Petroleum Institute, Washington,  D. C.

      I am very glad to have been invited to participate in your Forum here in Tulsa.   The
wastewater management subjects that you will be discussing in your sessions this week  are
matters of great concern to refiners in this country —and, I have  found, in other countries
as well.   On a recent trip to the USSR, for example,  I found Russian refiners talking  about
some of the same  things that you are talking about here.

      A good many of the wastewater management problems that refiners face in  this country
can be traced to  the Federal  Water Pollution Control Act.   That  is hardly news.   At  the
time of the passage of the 1972 Act, Congress was frustrated by the lack of progress that
had been made under earlier versions of the Act.  It determined that major legislation,
with teeth in it,  was necessary.  But it also knew perils and  pitfalls would be involved,
as is the case with all legislation of such broad sweep and bite.

      In this case,  for example, many important water pollution control problems had  not
yet even been clearly defined.  And certainly, to meet the ambitious  goals of the Act,
it would be necessary to develop new control technology, always a tricky business.  Back
in 1972, therefore, Congress wisely built into the Act  some mechanisms that would facilitate
making any necessary changes that might be indicated  in the  future.

      Among other  things, Congress established the National Commission on Water Quality.
As you know, the Commission has just completed hearings on  the staff draft of its report
to Congress,  which is due in March.  As a result, some of the tools needed for making
the kinds of changes envisaged  by Congress are now at hand.   I am referring to the really
impressive amount of data in the draft report itself and the public hearing record.

      This evening, I would  like to offer for your consideration a general proposition which
I  believe would be  helpful to bear in mind  as we go about making such changes.    Briefly,
it is that our country is best served by the kind of environmental legislation that  responds
to clear environmental needs, has been  carefully weighed against other pressing national
needs, and promises to yield benefits commensurate with the costs involved.

      In light of  this proposition, I  would like to talk with you, first, about the  impact of
the Federal Water Pollution Control Act on the petroleum  industry — and on  the economy
as a whole, for problems like the ones you  are discussing here have reprecussions far beyond
the refinery gate.  Second,  I would like to make some suggestions  for  amending the Act
in ways  that will  enable this country to  move ahead toward reasonable  water  quality goals,
efficiently and effectively.

      First, regarding impact, as this audience is well aware, life under the  Act is very
difficult for the regulated industries now and, unless some sound changes are  made, will

                                       107

-------
 108

be a great deal more difficult in the future.  As things presently stand,  the EPA Administrator
must require what is, in his judgment, the best  practicable control technology currently
available to be installed by 1977 and the best available technology  economically achievable
by 1983.   Beyond that date, the Act sets a national goal of elimination of discharge of pollu-
tants, or EDOP, for 1985.   Implicit in  these three dates and in the  increasingly stringent
requirements for each date  is a rapidly rising cost forecast that is of  great concern to the
petroleum industry and, indeed, to all affected industries.

      That concern is justified.   Many  of you  are familiar with the  cost findings of API's
1973 Brown and  Root study  of the economics of  refinery wastewater treatment.   The study
concluded that the  cumulative capital investment in refineries alone to achieve the three
levels of wastewater treatment required  by the  Act, adjusted to  1974 dollars,  would reach
$2.5 billion in 1977,  $4.7 billion in 1983, and $6.7 billion in  1985.

      Of even greater concern to us in the petroleum industry is the fact that much more
attention needs to be given to the cost-benefit  ratio associated with  water pollution control
requirements.   The Brown and Root study plotted capital costs against water pollution
reduction — measured in terms of the three increasingly stringent levels of treatment techno-
logy and Biochemical Oxygen Demand, or BOD, discharges.  The cost per annual pound of
BOD removed  from  refinery effluent streams by  the first level of control is estimated, again
in 1974 dollars,  to  be about $10.  The cost of application of the second  level of technology
for 1983, although  it would bring only a small  reduction in  discharges,  would just about
double to some $20 per annual pound.   And the cost of elimination  of discharge of pollutants
by 1985 would soar to more than $80 per annual pound — about eight times the cost of the
first level of control.   That, in our judgment,  is completely unreasonable.

      The industry expenditures on water quality have already been  substantial — some
$2.8 billion between  the years 1966 and 1974.   The very serious question now arising is
whether, after already making notable progress, the petroleum  industry should be required
to continue to commit ever larger amounts of money to achieve even smaller reductions in
water pollution.   The anticipated reductions are so small, in fact,  that in many, if not
most cases,  they could not even be accurately  measured in the receiving waters.

      In other words, whatever degree of success the petroleum  industry  and other industries
should attain,  at whatever cost, the quality of  the receiving waters  - - the name of the game,
after all  — would  remain about the same.

      That bleak outlook is not the petroleum industry's alone.  A few months ago, the
Environmental  Reporter  carried a story on several of the regional assessment studies sponsored
by the National  Commission on Water Quality. These studies confirm  the dismal cost-benefit
picture.  With regard to  the Commission's study of the densely populated and  industrialized
Houston Ship Channel-Galveston Bay area, for example, the Reporter had this to say:

                The study  concludes that the costs of meeting  1983  and 1985
                scenarios would be disproportionate to the  benefits  of
                achieving the goals of the Act.  Changes  in water quality
                and hence recreational opportunities/lumber of aquatic and
                terrestrial biota,  and aesthetics would be minimal  . .  .

-------
                                                                                   109

      Yellowstone country, on the other hand,  is far from being densely populated and
industrialized.  Yet here, too, the Commission's investigators reached the same conclusion.
Again according to the Environment Reporter:

                The report for the Yellowstone River Basin .  . says the high
                concentration of nutrients in the river basin appears to be
                of natural origin.   The river's tributaries contain high
                concentrations of sediments and total dissolved solids,
                mostly from natural and diffuse nonpoint sources  .  . . and
                they would  not be significantly affected by implementation
                of the Act.

      With so little  to be gained in the way  of improving actual  water quality, even after
heavy expenditures, it would seem wise — indeed mandatory —  to assess our position in
light of genuine environmental needs, other  national needs, and, especially,  costs and
expected benefits.

      Everyone I  know in the petroleum industry is acutely aware of the  need for such
assessment.   The primary mission of the industry, of course, is to help supply this country's
energy needs.  That is our  reason  for being.  And we know that environmental expenditures
on the order of magnitude required by the Act are bound to have an adverse effect on the
industry's ability to  generate the capital necessary to maintain a satisfactory growth pattern
and meet future energy needs.

      Capital formation is one of the most important problems — possibly-- facing the
petroleum industry.   In the years ahead,  if the nation is  to achieve reasonable energy
self-sufficiency,  the industry's capital needs are going to  be staggering.   It is therefore
essential that legislators and government environmental planners —  at some point  along the
ling, and the sooner the better — consider all  of the industry's environmental  expenditures
together and then relate them to the industry's  expected capital requirements for all purposes.

      The plain fact is that the amount of capital available to the industry in a given year
is limited.   The capital used to meet  environmental requirements simply is not available to
meet other pressing requirements,  including, particularly, exploration and production —
the bedrock upon which a major part of the nation's energy  supply structure is built.

      What these environmental costs  are likely to mean is becoming clearer and more urgent
by the day.   In the  same year that the Brown and Root refinery study was  completed, API
did what it has long  been recommending that legislators and government environmental
planners do.   It embarked upon a much broader study covering all environmental costs to
all  segments of the industry.   Phase I of that study, a literature survey, was published in
1974.   Phase II,  forecasting environmental costs to the industry  for present and future
environmental regulations, is now being finalized.

      According to the preliminary findings of  Phase II  of the study, the cost of meeting
anticipated federal,  state, and local  environmental regulations between 1975  and 1985
to the petroleum industry  —  expressed in 1974  dollars —  will be $20.1 billion.   This figure,
incidentally, does not include the  cost of the 1985 goal of EDOP, which, again, we

-------
 no

believe to be unreasonable and most probably unattainable.  Excluding EDOP,  the cost of
meeting water quality regulations is estimated to be about $8 billion, or some 40 percent of
total environmental expenditures.

      The magnitude of the additional  $20.1  billion  environmental  investment for the 1975-
1985 period can be put  in perspective by comparing it with recent and projected levels of
total capital spending by the petroleum industry.   That $20.1 billion,  for example, is almost
as much as the  $21.3 billion  that the U.S. petroleum industry laid out in total capital and
exploratory expenditures during the two-year period  1972-1973.   In the next decade, however,
and for as far ahead as we can see, the industry's total capital  requirements -- in current
dollars, with a five per cent inflation factor -- are expected to be on  the order of $40 billion
a year.  The question  is:  Where is all of that capital going to come from?

      The coming  capital bind is  there for all to see.  The pool of funds available to the
nation for pollution control is limited — including, certainly,  funds for water pollution
control.  We simply do not have  the wherewithal to  accomplish all that has been promised by
the Federal Water Pollution Control Act within the time frame specified.

      I would like to turn now to some suggested amendments to the Act.   Like everyone
else who must deal with the provisionsof the  Act,  we in the petroleum industry have compiled
a long list of such suggestions.   This evening,  however, I  want to mention just  three of the
most important ones, designed to  provide some of the things now sorely lacking — namely,
achievable  deadlines,  realistic water quality goals,  and ensuing environmental  benefits
commensurate with the  costs  to society:

       1 .  Extension of the  1977 Deadline.   EPA has already concluded that because  of  lack
of funds, 50 per cent-- some 9000 — of the  municipalities that will be serving  60 per cent
of the 1977 population  will  not be able to comply with the 1977 sewage treatment require-
ments of the Act.   The Agency therefore recommends that the  1977 deadline be extended to
about  1983 —  for municipalities.    That may or may not be the optimal date, but we in the
petroleum industry agree that an  extension certainly is indicated.  We also feel, however —
and very strongly  — that the deadline should be extended  for industry, too,  especially since
such extension would not prevent EPA or the  states from requiring earlier compliance  where
possible through the individual permit program.

      As you know, a  part of the problem is  that  some refineries and smaller facilities like
terminals  and service stations discharge to municipal sewers.   Such facilities must rely on
municipal  treatment plants for removal of compatible pollutants in order to comply with  1977
requirements.   It is worth adding that industry must  bear 100 percent of the  costs, while
municipalities  need bear only 25 per cent of the capiral  costs for federally approved  projects.
Clearly, to extend the deadline for municipalities and not for industry would be inequitable
in the extreme.

      2.  Modification of the Requirement for Best Available Technology Economically
Achievable for 1983.    Once both municipalities and industry are in compliance with  the
first treatment  level requirements — at whatever date compliance is achieved — it would
be unwise  to continue to require  application of tremendously expensive second  level  tech-
nology everywhere, indiscriminately.   The  National Commission on Water Quality's studies

-------
                                                                                  Ill

show that water quality is improving faster than expected and that implementing the 1977
requirements of the Act will produce better water quality than was anticipated at the time
of enactment.   The Commission's studies also show, as I mentioned earlier,  that the benefits
of implementing the second level of technology would not in general be commensurate with
the costs.  The limitations and other controls imposed by the Act for second level technology,
therefore, should be applied only in those cases where they are necessary to achieve
receiving water quality and then only when  the environmental benefits exceed the costs to
society.

      3. Redefinition of the Term "Pollutant."    In general, the Act fails to distinguish
between natural and man-made pollutants.  The Act also fails to recognize that substances
now termed pollutants are not in all cases harmfn'    Salt and suspended solids, for example
— whether natural or resulting from man's activities — often occur without harm to waters
or to the biota in them.   What is needed is a definition of "pollutant" which excludes
substances which are discharged in  quantities too small to cause damage.  It would then be
possible to establish —  not the Act's present national  goal  of elimination of discharge of
pollutants, which just is not going to happen — but a national goal of elimination of dis-
charge of harmful quantities of pollutants, which well could happen.

      These amendments,  if enacted, would not solve all of the nation's water quality
problems.  They would clarify some of the nation's water quality goals and set reasonable
deadlines and procedures for attaining them.   And from clear goals and reasonable dead-
lines come the capacity of regulatory agencies to regulate wisely and the capacity of industry
and municipalities to comply at acceptable  cost.

      This position was championed last week by Bill  Ruckelshaus,  "Congress has an almost
neurotic fixation for setting deadlines  on environmental matters that are unachievable.  Then
by promising something we cannot achieve we become incapable of measuring progress.   A
delay of one year in meeting the standard that denied reality to begin  with makes it look
as if EPA is copping out and all the progress that has been made is ignored.   Adjustments to
some of the schedules are necessary for a sane and a reasoned approach to such a vital
national goal.  The question  should not be  whether an environmental  law is tougher  but
whether it is wise and in the best interests of the whole country. "

      John Queries,  last  December, also espoused these views in a letter to me, "On the
side of disappointment, we are all now wiser but sadder as to the ability and solve national
pollution problems quickly.   It is now evident that the work of cleaning up pollution will
require intensive efforts throughout the next ten or fifteen  years and that many important
environmental goals will not be reached any sooner than that.  Passing a law does not by
itself solve a  problem.  Manpower, money  and in some cases technica, advances must also
be provided and experience leads me to conclude that in a national program tackling complex
problems the achievement of results takes patience as well as persistence.  The risk this
creates is that if the intensity of effort weakens along the way some of the goals may never
be achieved.  "

      The national badly needs to move ahead toward its water quality goals at a more
considered pace,  allowing ample time to take into account ciM of its peoples' needs  and

-------
 112
how best and most economically to meet them.   To answer to that need, let's change course
and speed.

BIOGRAPHY
Peter N. Gammelgard is the Vice President of the American
Petroleum Institute responsible for Industry Affairs. He holds
a B.S. degree in Engineering from Northwestern University.
He served as a Commander in the U.S.  Navy during World
War  II.  He has been a Vice President  of the Air Pollution
Control Association and past President of the National
Petroleum Refiners Association.   He has participated in
environmental conferences around the world and has been
active in the affairs of the World Energy Conference.

-------
              SESSION III

       "BIOLOGICAL TREATMENT"

Chairman

George W. Reid
Regents Professor of Civil Engineering
University of Oklahoma, Norman, Oklahoma

Speakers

W.  Wesley Eckenfelder
"Activated Sludge Treatment of Petroleum Refinery Wastewaters
 an Overview"

James F.  Grutsch
"A  New Perspective On The Role Of The Activated Sludge Process
 and Ancillary Facilities"

Charles E.  Ganze
"Case History On Biological Treatment Of A Petroleum Refinery
 Wastewater"

Mohammed A. Zeitoun
"Optimization Of The Activated  Sludge Process Through Automation"

Fred M. Pfeffer
"The National Petroleum Refinery Wastewater Characterization
 Study"
                 113

-------
114

  BIOGRAPHY

       George W. Reid, born 1917, Indianapolis, Indiana,
 attended Shortridge High  School, Purdue,  Harvard and Johns
 Hopkins Universities, studying Civil and Sanitary Engineering.
 Employed by the Indiana State Health Department, a
 sanitary engineer with USPHS, MCWA and CDC.  Has
 taught at Purdue, Johns Hopkins, Florida and George Tech.
 Currently teaching at the University of  Oklahoma as
 Regents  Professor of Civil Engineering and  Environmental
 Sciences and is the  Director of the  Bureau of Water and
 Environmental  Resources Research.    Holds degrees from
 Purdue and Harvard Universities.

       Has published numerous articles,  etc., in the field  (170);
 is and has been consultant to the USPHS, EPA, Interior, USAF,
 WHO (PAHO), AID and others in water and public resources systems.  Recently involved
 in extensive research on modeling of natural  and man-made public systems.   Lectures at  r
 recent Urban Water Systems Conferences and seminars in Colorado,  New Mexico, Bogota,
 Lima, Europe,  etc.

-------
 ACTIVATED SLUDGE TREATMENT OF PETROLEUM REFINERY WASTEWATERS
                              AN OVERVIEW

                        W. Wesley Eckenfelder, Jr.
   Distinguished Professor of Environmental cTnd Water Resources Engineering
                 VanderbiIt University, Nashville,  Tennessee

     This discussion will consider aerobic biological  treatment in the activated  sludge
process.  The two basic equations which describe the aerobic biological mechanism are:

                 (a1)             cells              (a)
     Organics  +  O2  + N  + P 	>  new cells + COo  +  H2O
                                  (k)
(BOD,COD,TOC)
     (s)
                      + nondegradable soluble residue                            (1)
             (b')    (b)
     Cells  + O2 	> CO2  + H2O  +  N  +  P  + nondegradable cellular
     (Xy)                                                   residue              (2)

     Equation (1)  is the synthesis equation and occurs in the presence of degradable
organics and biological cells with oxygen, nitrogen,  and phosphorus present.  Equation
(2) is the endogenous breakdown of the biological cells, and  occurs continuously in an
aerobic environment.

     In order to design an aerobic biological process it is necessary to balance equations
(1) and (2). Since  petroleum refinery wastewaters contain an unknown mixture  of
organics, it is necessary  to operate a laboratory or pilot plant program to define the
coefficients to  balance these equations.  In some cases experience elsewhere may be
employed.   The pertinent parameters for design purposes in equation (1) are  the coefficient
a1 which is the fraction of organic matter removed in terms of BOD, COD or TOC
oxidized for energy; the  coefficient a which is the fraction of organic matter removed
converted to new cells expressed  as volatile suspended solids; the rate coefficient k is a
measure of the  rate of oxidation of the organics present in the wastewater.   The
coefficient b in equation (2) is the  fraction per day of the biodegradable  volatile
suspended solids oxidized.  Since it requires approximately 1 .4 pounds of oxygen  to
oxidize one pound of biological cells as volatile suspended solids, the coefficient  b1 is
generally approximately  1.4b_.  For all practical purposes  organic matter  removed in the
process is either oxidized for energy or converted to biological  cells (except for the
small fraction of organics which result in nondegradable by-products as shown in equation
(1) ).  Therefore, on a COD or ultimate BOD  basis,  the coefficients a +a' = 1 . As
shown in equations  (1) and (2), the biomass as volatile suspended solids is denoted  by the
symbol X  and  the organics  expressed as BOD, COD or TOC  by the symbol s.   It should
be noted that the  nondegradable soluble residue indicated in  equation (1) results from the
accumulation of nondegradable organic by-products  from the  biological oxidation process.
It has been found  that the magnitude of this value usually  falls  between 1 and 5 percent
of the influent COD. The nondegradable cellular residue as  shown in equation (2) will
amount to approximately 23 percent by weight of the volatile

                                     115

-------
 116

suspended solids generated in equation (2).  It can therefore be concluded that the pertinent
parameters required for bio-oxidation design are those indicated in equations (1) and (2),
namely a1 ,  a, k,  b and b1.   Typical values found for petroleum refinery wastewaters are
summarized in Table 1.   These reactions as they  would occur in a batch oxidation system
are shown in Figure 1 .

      Figure 2 illustrates the breakdown  of  1 pound of degradable COD  in accordance with
equations (1)  and  (2).   Analysis of the wastewater in Figure 2 indicates that the COD/BOD-
ratio is l:0.75;a' and a are 0.5 on a COD basis.   It is assumed that 1.4 Ibs CL will oxidize
1 Ib cell mass as VSS.

DESIGN RELATIONSHIPS

      As can  be deduced from equations  (1) and (2) nondegradable volatile suspended solids
will accumulate in the process as auto-oxidation of the cellular mass proceeds.

      In order to develop process design  relationships it is therefore  first necessary to define
the degradable fraction, x, of the volatile suspended solids (biomass) in the aeration process.
The degradable fraction is a function of the sludge age or the length of time the volatile
suspended solids are in the process as shown in Figure 3.  Sludge age in the activated sludge
process can  be defined by the relationship:

                                     X    (Ibs)
                       Sludge Age =^- ^^                                (3)

in which AX  is the volatile suspended solids  wasted per day from the process.  As the sludge
age is increased the nondegradable materials accumulate and the degradable fraction decreases
as one would  expect.   The relationship shown in Figure 3 represents  average conditions for a
petroleum refinery wastewater.

      The kinetic  relationship employed  for the design of a  completely mixed activated sludge
process is based upon  the removal of most soluble simple organic compounds as a zero order
reaction.   In a multi-component wastewater a number of zero order reactions occur con-
currently.   The summation of these zero order reactions  can be expressed in an exponential
form for a batch reaction or employing equation (4) for a completely mixed continuous flow
reactor.
                            v
In cases where the influent concentration of the wastewater varies markedly,  equation (4)
may be modified to equation (5) which considers variable influent wastewater strength.

                         S  - S
                                       S A                                       (5)
X  t    '  " V o
  v
When COD or TOC are employed as a parameter of organic strength, the nondegradable
portion of the TOC or COD must be substracted from the total as shown  in Figure 4.  It

-------
                                                                                   117

should be noted that the reaction coefficient k in equations  (4) and (5) relates to a specific
wastewater composition.   If the wastewater composition varies the reaction rate coefficient
k may also be expected to vary.

      It should be noted that the effluent concentration defined by equations (4) and (5) are
soluble concentrations.  In order to determine the total BOD in the effluent it is necessary
to add that contributed by the effluent suspended solids.   While the BOD contributed by
the effluent suspended solids will be a function of sludge age (increasing sludge age de-
creases the degradable fraction) for most activated sludge processes in the petroleum refining
industry, a value of approximately 0.3 mg BOD/mg SS may  be  used.

      The total oxygen requirements in the biological process are related to the oxygen
consumed to supply energy for synthesis  (equation 1) and the oxygen consumed for endo-
genous respiration of the biological cells (equation 2).   Total oxygen requirements can
therefore be defined by equation (6) and shown in Figure 5.

                        09 /day = a  '  (S   - S )  +  b ' x.(X  )                         (6)
                         z        —     o   e     —    v
If nitrification is occurring in the process this oxygen will also  be  included in equation  (6)
and considered in the  coefficient a1.

       Sludge accumulation from the  biological treatment of petroleum refinery wastewaters
can be similarly calculated.   Volatile suspended solids accumulation, AX ,  is equal to that
resulting from synthesis  (equation  1) plus the influent volatile  suspended solids not degraded
in the process, minus the volatile suspended solids oxidized  by  endogenous metabolism
(equation 2).   This relationship is  shown in equation  (7) and Figure 6.

                     AX  =  a (S  -  S  ) + S.  -  bx(X )                              (7)
                        v   —   o   e     i        v
      One further consideration in the activated sludge process is  that the biological sludge
generated in the process flocculates,  settles and  compacts in the final clarifier.   In most
cases the sludge settling characteristics have been found  to relate  to the organic loading to
the process (F/M) expressed as  pounds of BOD applied per pound of MLVSS.
                                       SQ
                               F/M  = --
In order to insure good settling properties, the F/M should be maintained within an accept-
able range as shown in Figure 7.   If the  F/M becomes too high, frequently non-settling
filamentous growths will result creating sludge bulking.  If  the F/M remains too  low,  auto-
oxidation of the biological floe and resulting floe dispersion causes an increase in effluent
suspended solids and a bulking condition.

      The final clarifier is an important consideration in the process design of the activated
sludge process.   The solids flux to the clarifier and hence the underflow solids concentrations
attainable are related to the zone settling velocity of the biological sludge mixture.  These
relationships are shown  in Figure 8.

-------
 118

ATTAINABLE EFFLUENT QUALITY

      Considering the performance of biological treatment plants the primary factors which
will affect effluent quality are aeration basins, temperature, changes in organic loading
(as reflected in the F/M) and changes in product mix (as reflected in the k value).  Where
temperature is critical for effluent quality during winter operation the aeration system should
be selected to minimize heat loss, usually a submerged aeration system.   By contrast,  high
temperatures during summer operation can be minimized by the use of surface aerators which
maximize heat loss from the aeration basin.   Covered basins as employed in the  high purity
oxygen process will maintain high temperatures during cold weather operation.

      Changes in organic loading can best be dampened by equalization.   Since,  all other
things being equal, the effluent BOD will be directly proportional to the influent, the
equalization basin can be designed for whatever maximum  effluent levels are permissible.
Variations in organic loading can also be compensated by increasing the sludge recycle,
thereby maintaining  the F/M constant.   The most difficult effect to assess is changes in the
reaction rate  k due to changes in wastewater composition.   In some cases the plant operating
conditions can be modified for  a  change in plant production schedule.   Equalization will
also help to dampen  this variability.   While the effects of temperature and organic loading
variation can be  calculated from  the design relationships,  it  is not possible in most cases to
calculate the effect  of product mix.  In this case a statistical  approach  using operating data
is  necessary.

      Long-term  case histories  of a number of activated sludge plants in  the petroleum refining
industry  have been reported in  a  study for the National Commission on Water Quality (2) and
are summarized in Table 2.   The variability in effluent quality which might be expected for
these parameters  are shown in Table 3.

      It  may be expected, therefore, that as a refinery becomes more complex there will be
a greater variation in wastewater composition and hence a greater variation in k.   The net
effect of this considering equations (4) and (5) is to  increase  the variability in effluent con-
centration, S , on a daily basis.   Refineries in the northern part of the  United States would
be expected to have a greater monthly variability due to seasonal temperature effects than
those in  the south.

LIST OF SYMBOLS

a     Ibs VSS produced per Ib BOD, COD or TOC removed
a1     Ibs oxygen  required to  oxidize one  Ib of BOD,COD  or TOC
A     Area
b     Ibs VSS destroyed per day per Ib oxidizable MLVSS
E1     Ibs oxygen  per day per Ib oxidizable MLVSS
C.    Oxygen concentration  in the basin
F/M   Food-to-Microorganism Ratio
hr    Hour
k     Organic removal  rate coefficient, days
MLVSS  mixed liquor volatile  suspended  solids

-------
                                                                                   119
N    oxygen transfer coefficient, corrected for field conditions
N    Oxygen transfer coefficient under standard  conditions
OR   Clarifier overflow rate  (gpd/sq ft)
Q    Flow rate
R     Recycle flow
S     effluent organic concentration
S     Influent organic concentration
x°    T-
t     Time
T     Temperature
T     Basin temperature
S.    Average influent suspended solids
X    Volatile suspended solids
a     Ratio of oxygen transfer rate of waste to water

REFERENCES

(1)  Eckenfelder, W.  W., "Principles of Biological Treatment," Vanderbilt University,
    Nashville, Tennessee,  1975.
(2)  Engineering  Science, Inc.,  "Petroleum Refining Industry Technology Costs of
    Wastewater  Control," Report to the National Commission on Water Quality, June,  1975.
(3)  Engineering  Science, Inc.,  "Activated Sludge Treatability of Petroleum Refinery
    Wastewaters," 1975.
BIOGRAPHY

W. Wesley Eckenfelder holds a B.E. from Manhattan College;
a M.S.  from Pennsylvania State University; and a M. E. from
New York University.   Professor Eckenfelder has served on
the faculty of Manhattan College, the University of Texas at
Austin,  the University  of Delft in Holland, and is now a
Distinguished Professor of Environmental and Water  Resources
Engineering at Vanderbilt University.  Wes is also  Board
Chairman of AWARE, Inc. and has written over 200 publi-
cations and 11 books on water quality management.  Pro-
fessor Eckenfelder has received many awards.

-------
120

DISCUSSION

Bob Worton:   Would you care to comment on the effect of po I/electrolytes on keeping sus-
pended solids down in the final effluent when operating with a high degree of nitrification.

Wes  Eckenfelder:  The polyelectrorytes probably would not help much, since  denitrification
occurring in  the final clarifier would lead to floating sludge due to the nitrogen gas bubbles
released during the process of denitrification.

Bob Worton:   Do you know any research in this area?

Wes  Eckenfelder:    Specifically on nitrification, no.  I think the answer to this problem,
however7~is to remove the sludge from the final clarifier before active denitrification sets
in.

Juanito Chavez:   My first question is also on the  polyelectrolyte injection.   In commercial
application is there any specific point where you would inject the polyelectrolyte in such a
way  that you effect good  distribution and at the same time not destroy the biological floe.

Wes  Eckenfelder:   To my knowledge, I know of no full-scale plant that has been using
polyelectrolytes for effluent suspended solids control for long periods.  All of the experi-
mental studies that I have been associated with  have  injected the polymer prior to going
into  the final settling tank.   In view of the fact that most of these have not been pre-
engineered to optimize the process, much of the data is erratic.

Bill Oberman :    Would you care to comment on the use of dissolved air flotation instead
of secondary  clarification, where poor settling sludges are involved?

Wes  Eckenfelder:   Some  years ago I  was  involved in engineering an activated sludge system
employing BAF in  lieu of  secondary clarification.  While a reasonably concentrated sludge
was obtained for re-cycle, the effluent suspended solids from  the flotation unit were highly
variable.  It is possible that advances in flotation practice may overcome this problem to
date.

Larry Eckelberger;    What type of variability factors can  be achieved in your opinion?

Wes  Eckenfelder:   I quote here data that generates out of the Petroleum Industry report to
the National Commission  on Water Quality.   Monthly variability in BOD is  2.7 times the
average with daily variability in BOD being 5.4 times the average.  For effluent suspended
solids, monthly variability is 2.5 and daily variability 3.8.  I consider these realistic limits.

John Garner: This question may be a little premature and may be that it will be answered
later,  but I have  found in raw waste BOD that I require 20 to 25 aeration horsepower and in
practical application I find I need 50-55  horsepower.

Wes  Eckenfelder:    I have no figures specifically for the petroleum refining industry, but in
the pulp and  paper industry,  we generally estimate 45 pounds of BOD removed per day per
horsepower as a rule-of-thumb.   I  see no reason why this  would be markedly different in
petroleum.

-------
                                                                                               121
             TABLE 1  "BIOLOGICAL TREATMENT COEFFICIENTS FOR PETROLEUM WASTES (3)"
    Refinery
Category  Size
(EPA)    (MBPD)
 E
 C
 D
 C
 E
 B
 C
   406
    60
    90
    63
   427
   205
   88.4
      k*
    I/days
 BOD    COD

4.15    2.74
                            Sludge Growth Coefficients   Oxygen Requirements Coefficients   Influent
                              BOD Basis    COD Basis    BOD Basis   COD Basis          mg/1
                               a    b1       a    b1      a1    b1     a1    b1         BOD
0.595
                 3.84
97
86
92
0
79
0.5 0.08
0.5 0.05
        7.24
0.5
0.5
0.44
0.26
0.2
0.43
0.06
0.06
0.1
0.03
0.08
0.10
0.47 0.28
0.57
0.06
0.34
0.35
0.46
0.40
0.52
0.1
0.11
0.06
0.08
0.05
0.01
0.14
                              0.6 0.05
                                           0.88 0.10   -
244
575
396
153
600

345
160
* at 24 C k = organic removal rate at 24 C
1  based on total  VSS
  Sub-
category

  C
  C
  B
  B
  B
      (lOOOb/S.)
              a

         60
         63
        135
         63
        205
            BOD

              13
              16
              18
              28
              12
              COD

              174
              130
              132
              183
              112
                    mg/1
                TSS      Oil
                78
                30
                23
                41
                15
            9.4
            9.0
            9.9
                 Phenol

                  .053
                  .15
                  .04
                     NH3-N

                       1.1
                       6.7
                       8.2
COD

509
980
782
428
170
563
806
TABLE 2  "LONG TERM EFFLUENT CONCENTRATIONS FROM THE ACTIVATED SLUDGE PROCESS (2)"

       Capacity
                       S

                     0.27
           TABLE 3  "VARIABILITY OF EFFLUENT LEVEL IN ACTIVATED SLUDGE TREATMENT (2)"
Daily Min
      Max
      Ave

Monthly Min
        Max
        Ave
              BOD

              3.8
              7.6
              5.4

              2.2
              4.6
              2.7
                   COD

                   1.9
                   5.0
                   3.6

                   1.5
                   1.9
                   1.8
                                                TOC
                      3.5
                      2.9
                         TSS

                        2.4
                        5.2
                        3.8

                        1.5
                        3.3
                        2.5
                       Oil

                       3.9
                       5.2
                       4.7

                       1.8
                       2.9
                       2.2
                            Phenols

                            4.5
                            5.2
                            4.9

                            1.4
                            3.4
                            2.5
                                                                                            NhL-N
                                                                                               O
                                  9.
                                 10.
                                 10,
                                  2
                                  4.1
                                  3.4

-------
122
             o
             LJ
             O
             z
             o
             o
                                           Nondegradable
                                             Residue
            Figure  I.  SCHEMATIC OF AEROBIC BIOOXIDATION  PROCESS
              (COD) a'=0.5

              (BOD5)a' = 0.75
a = 0.5 (COD)

a = 0.5/1.4 = 0.36(VSS)
                                         0.36 Ib

                                          VSS

                                       NEW CELLS
                         0.27-K4= 0.39
        0.23-0.36 = 0.083
           IbVSS RESIDUE
                             0.39 Ib 0,

                              USED
                           Figure  2  .

-------
       0.8
       0.6
       0.4
   K   n p
   <•>   w.t
   o
   o
   03
   U.
   O
   o:
   u.
0.2      0.4     0.6
    _F_ /mg influent BOD
    M  I mg MLVSS-day
                                                            123
                                         0.8
                 1.0
  Figure 3.  CORRELATION OF  BIODEGRADABLE FRACTION OF
            MLVSS,  x, WITH ORGANIC LOADING, F/M.
      400
      300
      200
o
V)
       100
                  BOD-
Sample Correlation
For COD or TOC
                       Shows effect of
                       variable composition
                       on organic reaction
                       rate, K
             Non- biodegradable fraction
              	V	
               I      i      l
             I
         0     10    20    30    40    50    60    70
               SOLUBLE EFFLUENT BOD OR COD, Se(mg/l)

Figure 4. DETERMINATION OF ORGANIC  REACTION  RATE
         COEFFICIENT  ( EQUATION 5 )

-------
124
1.0
              0.2
          X
          x
                                             BOD-
                                           Somple correlation
                                           for COD or TOC
                                                     I
                         0.2       0.4       0.6      0.8
                            Sr  [ Ib  organics removed   ^
                           xXtf  Vlb biodegradable VSS-dayJ
                                               1.0
         Figure 5. DETERMINATION  OF OXYGEN UTILIZATION COEFFICIENTS
                   (EQUATION  6)
                         Sample Correlation
                         indicates influence
                         influent VSS on
                         sludge sythesis
                         coefficient
                                                 Sample
                                                 Correlation  —
                                                 for COD or
                                                 TOC
                              .0.4     0.6     0.7
                             /Ib organics removed      \
                          xXv\lb biodegradable VSS-daJ

       Figure 6. DETERMINATION OF SLUDGE PRODUCTION COEFFICIENTS
                (EQUATION 7 )

-------
   140
^   120
o>
E
                                                                 125
 o
 (O

 a
 UJ
 a
 z
 UJ
 a.
 V)

 CO

 h-

 UJ
U.
u,
UJ
   100
    80
    60
    40
     20
                                                ZSV
                            ±
                                   _L
                0.2
                            0.4         0.6

                           Ib influent total  BOO
0.8
                     M
                                                                4.0
                                                               3.0
                                                               2.0
                                                               1.0
                                                                  o
                                                                  o

                                                                  Ul
                I-
                UJ
                CO

                Ul

                o
                N
                                                             1.0
                             Ib MLVSS-day

 Figure  7.  CORRELATION OF EFFLUENT SUSPENDED  SOLIDS AND  ZONE

           SETTLING  VELOCITY  WITH F/M.
 "0        0.2       0.4       0.6       0.8       1.0

              TOTAL SUSPENDED SOLIDS CONCENTRATION  (%)


Figure  8 .   BATCH SOLIDS FLUX CURVE FOR CLARIFIER  DESIGN
                                                                  1.4

-------
  RAW WASTE
 GRAVITY OIL SEPARATION!   EQUALIZATION
                                        I

                                Skimmed Oil
              RESIDUAL OIL
              SSUSPENDED
               SOLIDS RE-
                MOVAL
BIOLOGICAL BOD REMOVAL
    a FINAL
 CLARIFICATION
TREATED
EFFLUENT
                                                                                                                     CO
                                                                                                                     o
                          EQUALIZATION   f]~
                              BASIN
                                    OIL
                                  SKIMMER
                                  (Floating)
                                                                               | ^ Return
                                                                              I i    Sludge
                                      Dewatering Float Sludges            ,
      TO
         -^
   RECOVERY
OILY SLUDGES TO TREATMENT 8 SUBSEQUENT
RECOVERY OR DISPOSAL BY INCINERATION
                        Conditioning
                         Chemicals

                TO FINAL
                 DISPOSAL   -*	
                (LANDFILL,
             INCINERATION, ETC.)
                                            T   Conditioning
                          GRAVITY THICKENER  ^J^ Chemicals
                         OR AIR FLOTATION
                                         Supernatant
f THICKENER  J.^ Chemical!
.OTATION -T^f  X   I


Inn*       \       /
                                                         \^S
                                                      Allernotive I
                                              AEROBIC
                                              DIGESTION
                                               BASIN
                    SLUDGE FLOW

                   -PROCESS WASTE FLOW
                                  DEWATERING    THICKENING   STABILIZATION
                                  EXCESS BIOLOGICAL SLUDGE HANDLING FACILITIES
FIGURE 9:
SCHEMATIC  PROCESS  DESIGN FLOWSHEET FOR ACTIVATED SLUDGE SYSTEM

-------
          "A NEW PERSPECTIVE ON THE ROLE OF THE ACTIVATED SLUDGE
                      PROCESS AND ANCILLARY FACILITIES"

                                   J.  F. Grutsch
             Coordinator of Environmental Projects,  Standard Oil (Indiana)

                                   R. C. Mallatt
             Manager,  Environmental Conservation, Standard Oil  (Indiana)

      Refiners have found the activated sludge process (ASP) to be a very effective and
broadly applicable process.   The versatility of the ASP has impressed many users and
emphasis has been placed on this aspect of the  process.   However,  design and operation
of the ASP have not proceeded in a direction that is optimum for  meeting  future effluent
quality goals.  The ASP  is  generally used to achieve major reductions in  suspended and
colloidal material, as well  as dissolved substances.  When the refiner is almost wholly
dependenton  ASP to achieve end-of-pipe treatment objectives, this operating mode may
achieve the best overall reduction in contaminants.  In the future, however, the refiner
will be confronted with very restrictive effluent limits on suspended solids, BOD, COD,
ammonia, and some bioresistant materials, as well as variability limits for each parameter
as conditions  for obtaining an NPDES permit.   Under these conditions, the ASP becomes
only one of a series of process steps required to meet the objectives.  When viewed in
this context,  the ASP  can be optimized to achieve  further control of process variability
and increased removal of soluble contaminants  to concentration levels below that achiev-
able by activated carbon.  The key to optimizing the ASP in this new mode is to design
and operate the preceding end-of-pipe process facilities for maximum removal of inter-
fering colloidal and suspended contaminants.

APPLICABLE REFINERY TREATMENT TECHNOLOGY

      Best practicable control technology currently available (BPCTCA), as defined by
the federal Environmental Protection Agency (EPA) for petroleum  refineries, is based on
both in-plant and end-of-pipe control technology.   Suggested in-plant controls include:

      1.   Installation  of sour water strippers to reduce the sulfide and ammonia
         concentrations  entering the treatment plant.

      2.  Elimination of once-through barometric condenser water by using surface
         condensers or recycle systems with oily water cooling towers.

      3.  Segregation  of  sewers, so that uncontaminated storm runoff and  once-
         through cooling waters are not treated normally with the process and
         other contaminated waters.

      4.  Elimination of polluted once-through cooling water by  monitoring and
         repair of surface condensers, or by use of wet and dry recycle systems.

      EPA recognizes also that good housekeeping practices and precautionary measures

                                     127

-------
 128

taken during the turnaround of operating units are important considerations that can have a
substantial impact on waste loads discharged to end-of-pipe water treatment facilities.

      BPCTA end-of-pipe treatment technology is based upon the existing waste water
treatment processes currently used in the industry.  Although various options are possible,
as shown in Figure 1, the most sophisticated and complete end-of-pipe treatment sequences
usually make provision for equalization and storm water diversion; initial oil and solids
removal in API or parallel  plate separators; further oil and solids removal in filters, clarifiers,
or dissolved air flotation units; carbonaceous wastes removal by any of several biological
oxidation processes or combinations thereof; and a final clarification or polishing step,
consisting of chemical coagulation or destabilization  and either dissolved air flotation or
granular  media filters for phase separation.   For BATEA,  in 1983, the contaminant con-
centrations achievable by an additional activated carbon process element is an indicated
objective.

      More specifically, the principal objectives of the end-of-pipe treatment sequence for
process design considerations are as follows:

      1.   Maximize removal of soluble contaminants contributing to biochemical oxygen
          demand (BOD).

      2.   Maximize ammonia removal.

      3.   Maximize removal of contaminants contributing to chemical oxygen demand
           (COD).

      4.   Maximize final  effluent clarity.

      Achieving these primary objectives gives rise to some secondary design  objectives:

      5.   Effect the efficient removal of gross quantities of solids and oil.

      6.   Provide for hydraulic and chemical equalization of refinery wastes.

      7.   Minimize the generation of excess  biological sludges.

      8.   Minimize any odor and bulking problems from biological sludges.

      9.   Minimize IOD (Immediate Oxygen Demand) of water entering secondary
          treatment.

     10.   Minimize sulfides in the effluent.

     1 1 .   Discharge  an effluent containing dissolved oxygen and a good threshold odor
          number.

-------
                                                                                 129

SECONDARY TREATMENT IS THE KEY TO OPTIMIZING THE END-OF-PIPE
TREATMENT SEQUENCE

      If the overall treatment sequence is to be optimized, the principal objectives
listed above point to secondary treatment as the key end-of-pipe treatment element.
Various other process steps are designed to  enhance its performance.  Further,  in contrast
to prevailing practice, secondary treatment must focus on removal of only soluble con-
taminants.   This limited objective for secondary treatment requires the objectives of the
remaining elements in the treatment sequence to be defined,  establishes a basis for operat-
ing conditions that maximizes the performance of each treatment element, thereby optimiz-
ing the effectiveness of the overall treatment sequence.  Adherence to this design philos-
ophy can yield an effluent quality superior to that achievable with an add-on activated
carbon treatment element.

OPTIMIZING PRETREATMENT PAYS  OFF

      Optimizing pretreatment not only permits the ASP to be designed and operated in  an
especially attractive mode of exceptionally high sludge age as will be described later,
but major savings in energy costs and  capital costs can be made.   Consider, for example,
the indicated raw waste load in Figures 2-4 to be average data.   These quantities are
highly variable in a refinery and may be  increased five to ten fold by many situations
including,  for example,  (1) an occasion  of high pH or excessive sulfides which  contributes
to increased colloidal matter with high zeta potentials, (2) sulfonate waste from an alky-
lation  unit, which peptizes and disperses contaminants, (3) loss of catalyst fines that
remain dispersed  in water, (4) loss of  residual oils,  coke fines,  wax, clay, or lube stocks
and additives, which emulsify oil and solids and tend to escape the API separator,  (5) loss
of efficiency in the API separator by hydraulic overload, etc.   Excessively high raw waste
loads from the API separator are characterized by the fact that they consist primarily of
discontinuous phase materials.    Adequately providing for chemical and physical equali-
zation in the intermediate treatment section will provide a more consistently uniform
quality effluent over a very wide range of raw waste loads.

      Maintenance of high sludge age in the ASP requires that the feed be free from  oily
inerts that accumulate  in  the activated sludge mass and impart undesirable properties to
it that preclude operation in the high  sludge age mode.  In large and complex  refineries
the oil and grease in the raw waste load  can vary over a year's time as shown in Figure 2.
Whereas the 50% probability (median) is only 70 mg/1 oil, about 5% of the data may be
in the 600+mg/l  range due to  causes previously outlined.  Optimized ASP operation  at
very high sludge  ages has been achieved  with excellent sludge properties when  all  dis-
continuous phase oil and solids are removed.   Pretreatment can achieve these oil levels
indicated in Figure 2 when treatment  consists of chemical destabilization and filtration.

      Removing oil from  the process water effluent prior to biological treatment recovers
the oil in a sludge much  easier and less costly to handle than if the oil is commingled in
a waste activated sludge.  Waste activated sludge is costly enough to dewater but when
commingled with oil, the oil  blinds precoat filters, increases significantly the volume of
waste sludge to be dewatered, and makes the final disposition of dewatered oily sludge
more troublesome.

-------
130
      Equalizing feed quality in terms of organic loading can also be achieved by the same
pretreatment as shown for BOD/COD data in Figures 3 and 4.  Colloidal and suspended
matter contributes significantly to the total BOD/COD;and periods of storm flow which
flushes sludge and silt from  sewers, high pH,  loss of emulsifiers, coke fines,  clay fines,  etc.,
typically give  a highly variable raw waste load which is frequently underestimated because
of sampling and testing difficulties.   Removal of the  discontinuous phase contaminants can
have a large impact on reducing the total organic load and the variability in quality of
water entering secondary treatment.

      Equalizing the organic loading results in substantial savings in capital  and operating
costs, simpler operations, and better effluent quality  with less variability in  quality of
water leaving the secondary treatment facilities.   For example, data from Figures 2,  3, and
4 are as  follows:

                                                             % Probability Less
                                                             Than Indicated Value
      Contaminant             Treatment After                50       95       98
 Oil and Grease, mg/1          Primary                        70       600     800
                               Intermediate                    4         10      12

 BOD,  mg/1                    Primary                       185       380     400
                               Intermediate                   80         96     105

 COD, ma/1                    Primary                       400       980   1,400
                               Intermediate                  220       310     350

      An engineer faced with design of an ASP to operate in the nitrification mode at 2 mg
 DO/1  with uniform effluent quality faces an almost insurmountable challenge,  both econo-
 mically and performance-wise,  if he has to deal with the variability of the  raw waste  load.
 Considering that data in the preceding table represent a year of operations, they also  reflect
 (1) seasonal, (2) weekly, (3) daily, and (4) hourly cycles.   The final  design of an ASP
 based on primary effluent would be such a compromise that year-round operation would be
 unsatisfactory, yet the unit would be expensively over-designed.  If,  for example,  a guide-
 line of 1 Ib.  0,~/lb. COD is used and the 98% probability COD value  is selected as a design
 basis,  the raw waste would require four times the aeration horsepower the intermediate
 treated waste does.   Actually, it would require substantially more,  because the alpha
 (oxygen  transfer) characteristics of raw waste are not as good as the intermediate treated
waste.

      Another example of the impact of the representative data on the ASP  is recognition
that,  for the same sludge age,  the equilibrium solids level in the ASP relates directly  to
organic loading.  One critical aspect of clarifier design is the solids flux rate.   The
increased solids circulating at the higher organic loading can increase substantially the
clarification capacity required.   Thus, once the engineering and operating details of an
ASP unit are addressed, the designer soon recognizes that proper design of the  intermediate
treatment section will contribute to  lower costs, improved and simpler  operation of the
secondary treatment facilities.

-------
                                                                                   131

PROCESS CONTROL OF THE ACTIVATED SLUDGE PROCESS

      A key to designing and optimizing the operation of an ASP unit as an integral part
of an end-of-pipe treatment sequence is first, to recognize that it is only a part of the
treatment sequenceiand operation of each part should be optimized in light of what it can
do best,  or is  required to do by another part to perform  optimally; and second, to provide
for use a method of process  control that optimizes performance.

      Performance of the ASP in  terms of BOD/COD removal relates to organic loading;
i.e., the food-to-mocroorganism  (F/M) ratio.  Eckenfelder (1) has summarized concisely
the effect of organic loadings on ASP performance when treating domestic wastes in the
manner shown  in Figure 2.   These data and the work of others  (2,3) show that better BOD
removals, total oxidation, nitrification, and  less  excess sludge are achieved with lower
F/M ratios.   The consideration which most often  limits the ability  to produce a high  qua-
lity effluent is development of an activated sludge mass with poor settling characteristics
at organic loadings under .2 F/M, where increasing values of  SVI are observed.  Sludge
volume index  (SVI) is a measure of the sludge settling characteristics and is the volume
occupied by 1 gm of sludge after the  aerated  liquor  settles for  30 minutes.   While the SVI
test has shortcomings beyond the  scope of this discussion, it can be a useful process indi-
cator.   On the basis of these data, optimal ASP operating conditions are usually cited
as F/M loadings in  the  range of 0.2 to 0.5.

      Our ASP development work has shown that optimum ASP  operating conditions are
achieved by removing essentially all  the dispersed colloidal  contaminants by chemical
destabilization and filtration prior to ASP treatment.  With  removal of colloidal matter
in the feed, the SVI of the  activated sludge did not increase at sludge ages well in excess
of 35 days, but consistently measured 70-80 with  excellent flocculating and settling
properties.

      Thus, contrary to comments in the literature that operation at high sludge age
(F/M 
-------
 132

material that intensifies the zeta potential and enhances the dispersion of the biomass.
Forster's ZP-SVI  relationship in Figure 6 shows SVI to be remarkably sensitive to ZP and
it is easily perceived how the highly negative charged particles in refinery primary efflu-
ents cause the SVI to deteriorate.

      For refinery effluent waters, the impact of accumulating colloidal solids in the acti-
vated sludge mass must  not only be addressed, but dispersed phase oil and oily solids also
must be considered.   Oily materials contribute to electrokinetic effects and destabilization
of the oily colloidal system allows the agglomeration and coalescing of the oily matter into
the activated sludge mass;  this, in turn, leads to other deleterious effects.   For example,
the coalesced oil matter associating with suspended  solids can cause so little gravity
differential with  water  that the associated materials pass through  the clarifier to the effluent.
The coalescing of the oily  materials also encourages the  formation of a stable, voluminous
emulsion mass consisting of oil, water, solids and air.   Typically this mass floats on the
surface of the aeration  tank and the discontinuous emulsion phase builds  rapidly because
its surface properties encourages agglomeration of like materials.   The  emulsified mass
does not biodegrade significantly.   Dispersion, or redispersion, of emulsified material in
the aeration tank typically causes the material to enter the clarifier where it collects on
the surface.    If the clarifier  is not designed to trap floating material, the material
escapes in the effluent  causing a significant degradation in effluent quality.

      Well-settling activated sludges have been  observed to have a density of about 1.016.
When associated  with oily  matter,  it is misleading to simply calculate the final sludge
density from composition and component density data, because the oily,  emulsified mass is
much more buoyant than calculated, due to entrapped air.

ASP PROCESS CONTROL DYNAMICS

      The amount of colloidally dispersed  inert solids and oil entering the aeration tank
affects the degree of process control available to the operator.  Colloidal and suspended
matter in the influent to the ASP at quite low concentration levels can make the  process
non-operable at the optimum conditions.

      The ASP process control  is via the food-to-microorganism (F/M) ratio, or perhaps
more practically  the sludge age (SA).   The sludge age and F/M ratio are related as follows:

                 J_ = a ( A F/ A T) -  b             (Equa.  1)
                 SA           M
      where,   a =  the sludge yield coefficient,
              b  =  the endogenous rate coefficient,
              M =  the mass (Ibs) of microorganisms in the system,
             SA =  the sludge age  (days),  and
        A F /AT  =  the mass (Ibs) of food (BOD or COD) supplied per day.

      The F/M  process  control  involves measurement of the BOD  or COD load ( A F/A  T)
per day and adjustment of the sludge inventory (M) to maintain a desired ratio.   The sludge
age method of process control can provide a simple hydraulic means to achieve the same end

-------
                                                                                 133


by sludge wastage.   The sludge wastage rate to maintain an indicated sludge age is cal-
culated from the equation:

      u/ *    /KA^nN      i     £(VA+V)XA    -  QX   )    (Equa. 2)
      Wastage (MGD)      1     7_ A    c'   A         e  !     ^      '
                      t x~or y       SA3
                      \.  f    e "

      where,  X   =  sludge concentration in recycle

              X   =  sludge concentration in effluent

              XA  -  MLSS
              V .  =  volume of aeration tank, MMGD
              V   =  volume of clarifier, MMG
                c                      '
              Q   =  feed  volume,  MMGD
      The sludge age method of process control has many advantages which are discussed
in detail by Walker (5).

      The accumulation of negatively charged colloids,  inerts and oil in the activated
sludge mass is  hypothesized  to be a principal cause of poor sludge settling properties (SVI)
and dispersed solids, thereby limiting the effective SA or F/M operating range.   The
sensitivity  of the ASP to influent inert solids can be readily demonstrated.   For example,
if undesirable  materials such as heavy oils,  catalyst fines, clay,  coke fines,  metallic
sulfides, etc., escape preceding treatment at the rate of 30 mg/1,  become flocculated by
activated sludge and accumulate, their effect can be predicted.   Assuming for simplicity
of calculation that  the inerts can escape only via sludge wastage and not via the  clarifier
overflow, the  equilibrium concentration levels in the waste and recycle streams and
aeration tank can be estimated using equations (1) and (2).

      The quantity  of inert suspended materials flocculated by the activated sludge mass
at equilibrium is shown in Table 1 for two cases; one in which the cell  yield is  0.4 Ibs/lb
COD removed and the second  in which  the yield is 0.6.   For each case, equilibrium
conditions  are calculated for sludge ages from 5 to  45 days.  In columns C through J
equilibrium data are shown for only soluble  substrate (COD), and in columns K through
S the impact of the inert suspended matter accumulating  in the system for a rate equivalent
to 30 mg/1 in  the feed.

      Not  expressly illustrated by the data in Table 1 is  the desirable change in biological
population  dynamics when operating  at  the higher SA.   At low SA, bacteria predominate
in the activated sludge mass.   As SA increases, organisms which are comparatively slow
reproducers accumulate to a greater extent.   One  effect of operation at high SA is to
allow accumulation in the activated  sludge  mass of nitrifying and other organisms to
degrade the more "bio-resistant" contaminants present.   A second effect is the accumu-
lation of protozoa in the activated sludge mass.  In an ASP operating at high SA,  the
role of protozoa  is chiefly effluent clarification achieved by predation on the dispersed
bacteria; i.e.,  those tending to carry out in, and  thereby contaminate, the clarifier
effluent (6).   Bacteria within the floe are relatively immune to predation.

-------
 134

      The soluble BOD/COD/TOC remaining in the effluent from an ASP relates importantly
to the SA in the treatment of refinery effluents.   As described earlier, the COD/BOD ratio
for refinery effluents may be quite high.   As shown in Figure 7, the BOD is usually removed
quite readily by the ASP with  little attention to process details.   However, operating at
comparatively low SA's usually results in an effluent containing substantial remaining con-
tamination measured as COD.   Operating the ASP at increasingly higher SA results in
increasing COD reductions, as shown quantitatively in Figure 7.  The TOC removal curve is
similar to the COD curve.   Dickenson and Giboney (7) in a study of refinery effluent treat-
ment observed that, with sufficiently long periods of acclimation, the organisms in an ASP
can acquire the  capacity to metabolize components normally biorefractory.   A specific
example of such a material was found to be tertiary butyl alcohol, which took a 5-month
acclimation period  before degradation was obtained.   Many other such observations are
reported in the literature and have been experienced by the authors.  Thus, operating at
high SA will  continue to reduce significantly the COD, whereas the observed effect on the
BOD may be  slight.   Additionally,  we have achieved total organic carbon (TOC)  levels
below 10 mg/1, superior to  that achievable with activated carbon.

BIOLOGICAL REASONS FOR LIMITING THE  OBJECTIVE OF BIOLOGICAL TREATMENT
TO REMOVAL OF SOLUBLE CONTAMINANTS

      Only soluble organics can be absorbed through the cell membrane of the microorganism.
Above certain minimum "saturation"concentrations (typically perhaps 5-10 mg/1) of degrad-
able organics, the rate of removal is independent of degradable organics  concentration, but
directly proportional to the amount of microorganisms  present.   Thus, the rate of biodegrad-
able organics removal is determined  by the rate of metabolism of the microorganisms down to
certain minimum concentrations.   Below these minimum "saturation" concentrations, the
rate of biodegradable organics removal is determined by their rate of passage through the
cell membrane.   As the concentration gradient across the cell membrane decreases, the
rate of mass transfer decreases, and further biodegradable organics removal  under these
conditions involves increasing the reaction time.

      Large particles of biodegradable organics are removed by two possible mechanisms:
(1) the microorganisms can evolve a  mechanism for engulfing or directly passing large mole-
cules into cells where confined enzymes initiate oxidative biochemical reactions, or (2)
they excrete the necessary enzymes causing the reaction sequence to be initiated outside the
cell membrane.   These exoenzymes  are produced and controlled by mechanisms not under-
stood, and in the activated sludge process it is not possible to predict the influence of
process variables on the kinetics of colloid and large particle biodegradation .  This latter
mechanism involves bacteria, and is  believed to be the principle mechanism operative in the
activated sludge process.

STRUCTURE OF THE BACTERIAL CELL

      Figures 8 and 9 are schematics of a bacterial cell and bacteria cell walls and mem-
brane which are useful in picturing the extreme complexity of biological  oxidation.  The
food material must pass through the cell wall and the cell membrane.  The outermost surface
of most bacteria is a slimy capsule varying in thickness up to 100,000+ A .   The composition
of capsules varies with species and may consist of polymers of glucose or other sugars, amino

-------
                                                                                 135

acids and sugars/Or polypeptides.   Capsules generally consist of about 98% water, and
the capsular layer probably serves the microorganism as an osmotic barrier; i.e.,  a
mechanism for guarding against too  rapid an influx or efflux of water (8).

      The soft outer layer of the cell wall is 100-150 A° thick and consists of an  outer
layer of proteins and polysaccharides and inner layers of lipids and possible  lipopoly-
saccharides and proteins.   The acidic polysaccharides are partially responsible for the
negative electric charge on the bacteria surface.

      The rigid layer  is about  50  A  thick and is comprised mainly of heteropolymers of
peptidoglycans arranged in a coarse three-dimensional foraminous mesh which provides
rigidity and strength to the cell wall.

      These outer layers offer  no  obstacle to inward passage of molecular oxygen, minerals,
amino acids, glucose, and large organic molecules.   Cell  wastes can pass outward unob-
structed (9).

      The periplasmic space is about 100 A  thick and, in some  bacteria,  contains degrad-
ative enzymes which  break down  the larger molecules passing  through the  cell wall into
simple  sugars, amino acids and other low molecular weight fragments.   These lower mole-
cular weight fragments are transportable across the cell membrane by transport systems.

      Membranes consist almost entirely of proteins and lipids (14).  The proteins serve as
enzymes and provide  the distinctive functional properties.  The lipids provide the gross
structural properties and are a permeability barrier.

FUNCTIONS OF THE CELL MEMBRANE

      The cell membrane is about 100 A  thick and serves three  key functions: It acts as
an osmotic barrier and is impermeable to ionized substances and  nonionized  substances with
molecules larger than glycerol; it provides a housing for active transport systems,  and
provides a site for key enzymic reactions involved in energy metabolism.  Cell membranes,
therefore, are highly  selective in regard to the substances that may pass through  them,
inward or outward (10).

      Substrates may be transported through cell membranes counter to concentration
gradients.  This process is called  active transport because it requires added energy.  Active
transport is accomplished by enzyme transport  systems sometimes  called permeases. These
systems are so efficient for concentrating metabolites that the  osmotic pressure in  the cell
can reach 300 psi  (14). Like other  enzymes, permeases are specific for the  substrate
involved and are inducible.

REACTIONS WITHIN  THE CELL

      With reference  to Figure 8, transport of the substrate through the cell wall  and
membrane is only a small part  of the over-all biological system.    Once inside the cell,
the substrates are acted upon by systems of enzymes called  endoenzymes.  There are
several thousand different enzymes within the  cell of a single  bacterium (15). Endo-

-------
 136

enzymes undertake two general types of processes inside the cell; synthesis of cell compo-
nents in food reserves,  and the release of energy from food stuffs.   The energy is either
stored in special  molecules like adenosine triphosphate (ATP)  inside the cell or is immediate-
ly used for any of the active processes of  the cell.  Recently it has been discovered that
bacteria can energize the  active transport of many substances by the oxidation of various
chemical substances without  the transfer of phosphate groups (17).

       Endoenzymes do not  act individually but as parts of coordinated and sequentially
operating systems.   Whatever effects one portion of the intracellular enzyme system has
some effect on all parts.   The  activity of an enzyme is inhibited by accumulation of the
end-products of the enzyme catalyzed reaction.   In a sequentially operating enzyme system
excessive accumulation of  a  reaction product may inhibit the  reaction not only of the
enzyme manufacturing the  reaction product, but  all prior enzymes in that sequence.  This
is an important form of automatic control  called  feedback inhibition.   In the presence of
excessive amounts of end-products, not only is enzyme activity inhibited but the actual
synthesis of the enzymes themselves may be repressed.  If a cell normally synthesizing a
certain substance is artifically  supplied with that substance  from an extraneous source,not
only is activity of the enzyme inhibited,  but synthesis of some or all of the enzymes in the
production sequence for that substance is  repressed until the enzymes are needed again.
This is called  feedback repression.   Differentiation is made between inhibition of the action
of the enzymes by their end-products (feedback inhibition) and repression of the synthesis
of the enzymes themselves  by the accumulation of end-products (feedback repression) in the
enzyme sequence (Figure 10).   In a sequentially operating  system the end-product of each
enzyme can be the inducer of the next enzyme in the series and the inhibitor or represser of
the preceding enzyme, thus carrying forward the  work of the enzyme factory (18).

      In the cell, energy is derived from  the enzymic  bio-oxidation of foods, and the energy
liberated is transmitted by  high energy compounds, notably  certain organic phosphates;  or
directly as recently discovered.  The exact manner in which  energy is transferred from  low
energy level oxidized foods into high energy foods is not yet fully understood,  but it has
been suggested that the participating enzymes are perhaps hundreds in number and must have
a specific orientation to each other in the cell (18).

REACTIONS EXTERNAL TO  THE CELL

      Suspended and colloidal matter require solubilization by exoenzymes preparatory to
transport into the cell.   Many  enzymes probably exist free  in colloidal suspension in the
fluid matrix of the cell.  Some of these are secreted to the exterior of a cell.   They are
called exoenzymes.  Exoenzymes are mainly digestive in function.   By hydrolysis they
decompose  complex organic matter,  such  as proteins, cellulose, and fats,  to simple soluble
molecules of amino acids,  glucose, glycerol and  fatty acids.   These relatively small mole-
cules can be transported through the cell  membranes of many microorganisms, there to be
utilized as  food.   Among the lower fungi, algae, and bacteria, digestive enzymes are to
a large extent excreted aimlessly into the surrounding fluid where they may or  may not come
into  contact with food.  They may be wholly  dissipated by  dilution, convection currents,
or other factors (12).

-------
                                                                                 137

STRUCTURE OF ENZYMES

      Enzymes are large complex molecules v/ith molecular weights ranging from about
10,000 to more than 1,000,000 and consisting of two components;  a colloidal protein  part
called apoenzyme,  and a non-protein, noncolloidal, low molecular weight, thermostable
part called coenzyme (organic) or cofactor (inorganic).   The two components constitute
the complete active enzyme called holenzyme but  commonly referred to by the term
enzyme.   Coenzyme molecules generally carry the distinctive reactive portion  of an
enzyme.   Acting with the apoenzyme, the coenzyme brings about the specific substrate
reaction  that is characteristic of that particular enzyme.  In many coenzymes the
distinctive reactive portion is a familar vitamin; nicotinic acid or its popular derivative
niacin; otheisare vitamins of the B complex such as thiamine which is vitamin B. or
riboflavin vitamin  B^.   In fact, all vitamins that function physiologically have been
found to  act as a reactive group of one or another  coenzyme (13,  19).

ENZYME MECHANISMS

      In  most enzyme reactions, the specific apoenzyme involved appears first to attach to
the substrate at certain specific sites involving only  10-20 amino acid residues confined
to an area of surfaces perhaps  15-20 A  in diameter  (19).  These  sites represent recipro-
cally corresponding physical structure and anionic  and cationic groups in the molecule
of the substrate.    This preliminary combination appears to place  certain bonds  in the
substrate under stress.  The coenzyme, because of its appropriate molecular structure,
then combines with a part of the substrate. For example, a halogen ion or amino group,
etc.  This ion or group is then either passed by the coenzyme to another coenzyme or to
a different substrate molecule, or may be liberated as waste into the surrounding fluid.
The final result depends on the nature of the reaction being catalyzed.   The enzyme,
freed of  the altered substrate residue, is  then ready to combine with another substrate
molecule and repeat the process.  If any energy is released by the reaction, it  is partly
lost as heat and in part taken up into the cell substance  by the formation of energy-
rich compounds that contain certain types of bonds as organic phosphate bonds,  thioester
bonds, and some others in certain coenzymes.   The  energy stored in such energy-rich
compounds is later used by the cell in motility and cell  synthesis.

ENZYME STEREOCHEMISTRY

      All known enzymes are globular proteins whose active site is either a crevice,
shallow depression or a pit and,  stereochemically,  conform to the principle of the
"hydrophobic in,  hydrophilic out" protein model.  All charged, and most uncharged,
polar groups are on the surface and the great majority of the nonpolar hydrophobic  groups
are buried in the interior.  This has a profound chemical purpose in that the reactant
molecules can be taken out of the water  where water molecules keep charged molecules
apart, and into the enzyme which is made up of hydrocarbons with a  low dielectric
constant.  In the organic solvent of the  enzyme, strong  electrical forces can be brought
to bear on the reactant causing them to react rapidly (15).

      Being colloidal proteins, enzymes  are macromolecules that may consist of hundreds
of the 22-24 amino acid residues linearly linked in different arrangements.  Trypsin,

-------
 138

chymotrypsin and carboxypeptidases,  for example, contain 200-300 amino acid residues per
molecule, or between 3,000 and 5,000 atoms.   The active site involves perhaps less than
20 of the acid residues and the three dimensional structure of the entire molecule is necessary
for maintaining the stereochemical structure of the active site.   Changes in the three
dimensional structure of the protein are typically accompanied by a loss of enzymic activity.
Since each enzyme is different from all others with respect to molecular configuration,
each is unique or specific.   Each enzyme can react  only with certain particular substrates
that have a corresponding stereochemical structure.   Specificity of enzymes varies greatly
in degree (19).

      Lysozyme, the first enzyme to have its detailed molecular structure worked out, ribo-
nuclease, and carboxypeptidase A are examples of "small enzymes" of molecular weights -  -
14,600, 13,700, and 34,600 respectively.   These enzymes are ovid  shaped and carbo-
xypeptidase A has dimensions of about 40 X 44 X 52  Angstroms.   Most enzymes are  far larger
than these (16).

GENETIC INFLUENCE ON ENZYME  PRODUCTION

      Each species of living  cells has a genetically determined natural endowment of certain
functioning enzymes. These are characteristic of and constantly present in all  of the cells
in that species of cell.   Such inherited enzymes, part of the normal constitution of the cell,
are called constitutive enzymes.   In  addition, many cells possess genetic determinants or
genes for the synthesis of such enzymes which are genetically repressed.   These genetically
repressed enzymes are manufactured only when  an inducing substrate or related substance
enters the cell.   In  the presence of the inducer the represser of the appropriate synthesizing
mechanism is removed and the enzyme specific  for the  inducer substrate is synthesized.
Such enzymes are  said to be  inducible.  Therefore,  although genes determine the full
enzymic potentialities of a cell,  environmental factors in the form  of inducers, determine
just which of the latent enzymic potentialities of a cell shall appear under any given
circumstances.   The cell  is evidently  not under the necessity of synthesizing all its potential
enzymes but only those needed (11).

ENZYME MEDIATED REACTION EQUILIBRIA

      Enzymic reactions are  reversible.  The accumulation of products affects the action
of any enzyme in either direction, as would be expected  in chemical  equilibria.   Under
any set  of constant conditions,  the equilibrium  point for  an enzyme catalyzed reaction is
constant. There is a constant relationship between concentration of enzyme and concentra-
tion of substrate.  Up to the point of "saturation",  the rate of reaction increases with
increase of ratio of one  component to the other.  With a constant amount of enzyme,
increase of substrate increases rate of reaction until every molecule of enzyme is fully
saturated with substrate.   Further additions of substrate cannot increase the rate of reaction.
Conversely with a  fixed amount of substrate, rate of reaction increases with the additions
of enzyme until all molecules of substrate are in contact with enzyme.   Further additions
of enzyme do not effect the rate of reaction.   In many instances enzyme  catalyzed reactions
appear to proceed  in only one direction because the  equilibrium  is never reached.   Under
normal conditions  in the living cell, enzyme reactions are constantly pushed in this manner
toward the one  or  the other side of the reactions.  Theoretically reversible reactions cannot

-------
                                                                                  139

actually reverse when large differences in energy levels are involved, since resynthesis
cannot be brought about by the same enzymes because they cannot restore the lost energy.
To reverse the reaction  requires that work be done by other systems of enzymes that capture
new energy from other sources (11, 20).

ENZYMES SENSITIVITIES  AND INHIBITIONS

      Enzymes are sensitive to all of the various precipitating and coagulating factors that
effect proteins in general;  temperatures  above 80 C,  excessive concentrations of heavy
metals or of hydrogen or hydroxyl  ions, active chemicals like chlorine, and strong alkalis
quickly destroy all types of proteins.   In addition to being sensitive to deviations in
temperature, enzymes are  very sensitive to variations in pH, osmotic pressures (salinity),
and UV and other radiations.   Whatever affects enzymes, of course, affects  microorganisms.

      The functioning of some enzymes may be inhibited by certain non-metabolizable
substances whose molecular structure is analogous to that of a true, metabolizable substrate.
Such metabolites like inhibitory agents are called metabolite  antagonists and because  of
their molecular structure, can pre-empt the specific combining site on a particular enzyme
protein to the exclusion of the  true metabolite.  The enzyme or coenzyme is not necessar-
ily destroyed but it can no longer function.

ENZYME CHEMISTRY

      A key to recognizing the important process design responses which must  be made in
an  integrated end-of-pipe  treatment sequence in which the activated sludge process is
an  element, is recognition of the complex enzyme systems  operative  and their requirements.
Exoenzymes excretedaimlessly into the surrounding water by bacteria for hydrolysis of
colloidal and suspended solids into smaller soluble molecules so they can pass  through the
cell membranes to be utilized as food, not only may be dispersed and lost in the effluent,
but are exposed  to and  unprotected from  environmental changes in pH, salinity, and other
conditions impacting on the stereochemical conformation or properties of proteins, causing
deterioration in  their functioning.   Endoenzymes on  the other hand,  are at least partially
protected from these environmental changes by the bacterial  cell wall and membrane.
Thus, using  the biological  process for removal of only soluble contaminants, as we recom-
mend, is consistent with obtaining a stable system less influenced by  transient  changes in
environmental conditions.   An additional important consideration contributing to capacity
and process  stability is  that activated sludge units utilizing only soluble contaminants can
be  operated at very high sludge ages.   This permits  much higher biological cell inven-
tories, which in turn increases  stability, capacity,  and purification while minimizing
generation of excess biological cells.

SUMMARY

      1 . A systems optimization  of a refinery end-of-pipe treatment sequence points to
          the activated sludge process (ASP)  as the  key element.

      2.  To achieve maximum water quality, system optimization points to reversing
          the historic work horse role of the ASP; i. e.,  it should be used only for

-------
140

           the removal of essentially soluble contaminants.

      3.  Reversing the role of the ASP yields a dramatic series of beneficial effects:

           a.  The SVI characteristics of the activated sludge mass are excellent at
               very high sludge ages; i.e., over 40 days.
           b.  Process control is greatly simplified by operating at a very high sludge
               age; i.e., greater than 40 days.
           c.  Using very high sludge age for process controls eliminates the need  for
               many  process control tests.
           d.  Operating at very high sludge age produces an exemplary effluent
               very low in TOC and other contaminants.
           e.  The cell yield coefficient is remarkably low at very high sludge age.
           f.  At very high sludge age the population dynamics of the sludge mass
               improve.
           g.  Maximum ASP capacity for purification is achieved by operating  at
               high sludge ages.

      4.  Systems optimization wherein the key element  is using the ASP for removal
           of only soluble contaminants permits clear definition of the roles the  other
           elements play; i.e., colloidal and suspended matter must be essentially
           all removed in pretreatment sections.

      5.  The technology is available to handle the  new requirements made by the
           process operations in their changed roles.

      6.  Current research developments on cell membranes and enzyme systems
           support strongly this new role for the ASP.

      7.  Cell genetics, wherein inducible enzymes or mutant species are required,
           support operating the ASP at very high sludge age for maximum purification
           capacity.

      8.  Bacteria are essentially  enzyme factories.    Enzymes are sensitive to temperature,
           pH, excessive concentrations of heavy metals, oxidizing agents, salinity, UV,
           and other radiations.   Using the ASP for  removal of only soluble contaminants
           is consistent with obtaining  a stable  system  less influenced by transient  changes
           in environmental conditions  since the endoenzyme systems are at  least partially
           protected from  these environmental changes by the bacterial cell wall and
           membrane.

      9.  This recommended systems optimization saves significant energy and other
           operating and capital costs of the end-of-pipe sequence.

     10.   This recommended systems optimization minimizes the generation of solid
           wastes, and the solid wastes generated are amenable to disposition.

-------
                                                                                  141

 REFERENCES

 (1)  Eckenfelder, W. W., Manual of Treatment Processes, V. 1, Chap. 4, p. 7,
     Environmental  Science Services Corp.
 (2)  Downing, A. L.7 "Factors to be Considered in the Design of Activated Sludge Plants,"
     Advances in Water Quality Improvement, Gloyna & Eckenfelder Editors, Univ. of
     Texas Press, p. 190.
 (3)  Wuhrmann, K., "Research Developments in Regard to Concept and Base Values of the
     Activated Sludge System," Advances in Water Quality Improvement,  Gloyna &
     Eckenfelder Editors, Univ. of Texas Press,  p.  143.
 (4)  Forster, C. F., "The Surface of Activated Sludge Particles in Relation to their
     Settling Characteristics," Water Research,  V. 2, (1968) p. 767-776.
 (5)  Walker, L. F., "Hydraulically Controlling Solids Retention Time in the Activated
     Sludge Process,"   Jour. WPCF, V. 43,  No.  1, (1971) pgs. 30-39.
 (6)  Jones,  G. L., "Role of Protozoa in Waste Purification Systems," Nature,  V. 243,
     pgs. 546-547.
 (7)  Dickenson, R.  L., Giboney, J. T., "Stabilization of Refinery Waste Waters with the
     Activated Sludge Process: Determination of Design Parameters," Proceedings of the
     Industrial Waste Conference, Purdue University, 19, pgs. 294-303.
 (8)  Frobisher, et al,  Fundamentals of  Microbiology, W. B.  Sounders Company, Philadel-
     phia (1974), p. 198.
 (9)  Ibid, pp. 201,204.
(10)  Ibid, p. 204.
(11)  Ibid, p. 92.
(12)  Ibid, p. 116.
(13)  Ibid, p. 89.
(14)  Fox, C. F., "The Structure  of Cell Membranes, " Scientific American (February, 1972).
 (15)  Dickerson,  R.E., Geis, L,  " The Structure and Action of  Proteins," Harper and
     Row, New  York (1969), p. vii.
(16)  Ibid, pp. 88,96,97.
(17)  Luria, S.E., "Colicins and the Energetics  of Cell Membranes, "  Scientific American,
     (December, 1975).
 (18)  Lehninger,  A.  L., "Energy Transformation in  the Cell, "  Scientific  American,
     (May, 1960).
(19)  Finean, J.  B., Engstrom-Finean Biological Untrastructure, Academic Press, New
     York and London  (1967).
(20)  Aiba, S.,  Humphrey, A. E., Millis, N. F., Biochemical  Engineering, Academic
     Press, New York, (1965), pp. 38,39.

-------
142

 DISCUSSION

 Milton Beychok:  Jim,  what you have really said is that if you filter well ahead of the
 activated sludge unit, you can  cycle up in sludge age and the process behaves much better
 in terms of capacity, effluent quality,  and operating simplicity because it's only working
 on soluble BOD.

 Jim  Grutsch:  Yes.

 Milton Beychok:  What do you do in your  DAF (dissolved air flotation) or filter  unit that is
 different than other people so you can achieve this?

 Jim  Grutsch:  We address the water chemistry.   Essentially everything in nature  has a
 negative electrical  charge on the surface.  Colloidal particles exist because they are
 repelling each other and resist agglomeration which  would facilitate phase separation. In
 a filter, for example, the  surface of a granular media filter is also negatively charged.  If
 we just attempt to filter out negatively charged  solids with a granular media filter, a  large
 fraction of the negatively  charged colloids will  be repelled by the media and rapidly  break
 through to the effluent. Similarly with a  DAP unit,  the flotation air has a negative electri-
 cal charge.  This is the  most important reason why Wes Eckenfelder's DAF unit wouldn't do
 a very  good job clarifying activated sludge  from mixed liquor; the negatively charged fine
 particles in activated sludge repel  the negatively charged flotation bubble resulting in
 poor phase separation efficiency.

     The key to making DAF units or filters work effectively,  therefore,  is to reduce the
 electrical charge on the colloidal particles  so that they can agglomerate and be trapped in
 granular filter media, or separated by attached air bubbles.  There are some excellent
 chemicals available for  this  purpose at reasonable cost.  Determination of the most cost-
 effective chemical treatment is easily done  by comparing zeta potential  titration curves.
 The  use of zeta potential titration curves is, unfortunately, in its infancy but this tool will
 see greatly increased use \A  the coming years.

 Anonymous:  Do you encounter  any  problems from oil and grease?

 Jim  Grutsch:  The electrical charge on oil particles  is also negative and the previous
 discussion on  colloids also applies to oil and grease particles.  Granular media  filters
 should be protected from gross slugs of oil, and the equalization basin does this.  The
 higher the ratio of inert solids-to-oil in the feed to the DAF of filter unit, the lower the
 oil  in the effluent.

 Robert L. Wortman: Would you care to comment on  the effect of polyelectrolytes on keeping
 suspended solids down in the final effluent when operating with a very high degree of nitri-
 fication ?

 Jim  Grutsch:  This question is central to the main thrust of our paper and I would like to
 develop the response in some detail  so that the importance of optimizing the end-of-pipe
 sequence as described is not overlooked.

-------
                                                                             143


1 .   We have no experience using pal/electrolytes in the activated sludge unit
    clarifier at high sludge ages because the need doesn't exist when proper prefil-
    tration  is used.  The average amount of total suspended solids in the clarifier
    overflow is less than 10 mg/l , and less than 5 mg/1 from a final stage of granular
    media filtration (also not using chemicals).  This is due principally to the excellent
    SVI maintained of about 80, with no "arms and legs" in the supernatent.  A wide-
    well clarifier is used,  but it is typically fully loaded; i.e., overflow rates greater
    than 70 gpd/s.f. and solids flux  rates greater than 25 Ibs/D/s.f.  Actually, the
    design capacity of the clarifier in this service has not been determined.

2.   "A very high degree of nitrification" specified in Mr. Wortman's question equates
    to operation  at a high sludge age (or low F/M).  With reference to Figure 5 and
    Wes Eckenfelder's data, under these operating  conditions the SVI of the sludge
    increases which presumably gave rise to the question.  A reason for the deteri-
    oration of  the SVI is shown in Figure 6;  i.e.,  the electrical properties of the
    sludge are deteriorating.

    Electrical  properties (ZP) of the  sludge deteriorate because of increasing accu-
    mulation of electrically charged inerts in the activated  sludge mass at increasing
    sludge age.  This  can be  visualized better by plotting some of the data in Table 1
    as I have done in  Figures 11 and 12.  Data from Part A of Table 1  is plotted in
    Figure 11.  The curve for inert solids shows the effect of capturing only 30 mg/l
    by the sludge mass. Thus,  for this comparatively weak waste, by comparing the
    ratio  of inert solids to biological solids it is observed that at 20 days sludge age
    half the sludge is  inerts,and at sludge ages greater than the biological fraction.
    This inert fraction has a negative zeta potential, and at some ratio of inerts to
    biological  solids the capacity of the biopolymers to destabilize the solids (i.e.,
    keep  the zeta potential of the sludge mass low  enough so that fine particles are
    not repelled  and thereby dispersed)  is surpassed and the SVI deteriorates (Figure
    6) and "arms and legs"  appear in the supernatent.  Without the inerts in the
    activated sludge mass,  excellent SVI's are  achieved at very high sludge ages.
    Our objective,  therefore, is to remove completely suspended matter in the
    influent to the activated  sludge  unit.

    Figure 12 shows the mixed liquor solids characteristics for the conditions of
    Part B,  Table 1.

-------
144
BIOGRAPHY

     James F.  Grutsch is Coordinator-Environmental
Projects, Standard Oil Company (Indiana).  He holds under-
graduate and graduate degrees in chemistry from Indiana
University.  Prior to his present assignment with  Standard,
Jim served successively as Group Leader for finishing,
blending and reclamation at the Amoco Oil Whiting
refinery, and Coordinator of Waste Disposal for Amoco.
Jim taught undergraduate chemistry for six years at
Indiana.
BIOGRAPHY

     Russell C.  Mallatt is Manager-Environmental and
Energy Conservation, Standard Oil  Company (Indiana).
He holds degrees  in chemistry  from the University of
Illinois and the University of Rochester.  Prior to his
present assignment with Standard, Russ served successively
as Chief  Chemist,  Chief  Process Engineering and
Technical Services Superintendent of the Amoco Oil
refinery at Whiting,  Indiana.

-------
                                                                                                          145
        TABLE 1  "MATERIAL BALANCES AROUND ACTIVATED SLUDGE PLANT FOR VARIOUS OPERATING MODES'
             Basis:  Flow = 20 MM GPD; Total Aeration and Clarifier Tank Volumes = 10 MMG; Thickened
              Sludge From Clarifier = 8000 mg/1; No Solids Loss in Clarifier Overflow; COD = 150 mg/1
A.  Basis:  a = .4; b = .05; COD Removal = 25000 Ibs/D (150 mg/1)

     Sludge Age  Equilibrium MLVSS            Wastage Rates
                     Sludge Yield
Case
1
2
3
4
5
6
(days)
5
10
15
25
35
45
mg/1
480
800
1029
1333
1527
1662
Total Ibs
40
66
85
111
127
138
/
/
/
/
/
r
032
720
783
200
375
572
F/M
.625
.375
.292
.225
.196
.181
Ibs/D
8006
6672
5719
4448
3639
3079
gpd @ .
120
100
85
66
54
46
8% Solids
,000
,000
,716
,666
,541
,148
Ibs/lb COD
.32
.267
.229
.178
.145
.123
              MGD Clarifier   Clarifier Flux,
                    .8% Solids I bs. Sol ids/Day
1.2
2
2.57
3.33
3.82
4.16
84,868
146,784
193,693
259,365
303,352
334,884
                   30 mg/1 Inert Suspended Matter in the Incoming Water Hove the Following Effects

                    mg/1
Inerts in
Wastage, mg/1
5000
6000
7000
9000
11000
13000
MLSS
Inerts
300
600
900
1500
2100
2700
MLSS
Total
780
1400
1929
2833
3627
4239
Inerts/MLVSS
ratio
.625
.750
.875
1.125
1.375
1.625
MGD
Draw (c
1.
3.
4.
7.
9.
10.
Clarifier
i> .8% Solids
95
5
82
08
07
6
Equilibrium
MLSS, Ibs.
65,052
116,760
160,879
236,272
302,492
353,533
Clarifier Flux.
Ibs Solids/Day
142,789
274,386
399,300
639,825
879,344
1,081,810
Wastage Rates
Ibs/D gpd
13,
11,
10,
9,
8,
7,
roio
,676
,725
,451
,643
,856
@ .8% Sol ids
195,000
175,000
160,750
141,650
129,541
117,750
 B.  Basis:  a = .6; rest as above in A

      Sludge Age  Equilibrium MLVSS
Wastage Rates
Case
7
8
9
10
11
12
(days)
5
10
15
25
35
45
mg/1
720
1200
1543
2000
2291
2492
Total Ibs
60,
100,
128,
166,
191,
207,
048
080
674
800
062
859
F/M
.417
.25
.195
.15
.131
.12
Ibs/D
12010
10008
8578
6672
5459
4619
gpd @ ,
180
150
128
100
81
69
.8% Solids
,000
,000
,567
,000
,820
,230
Ibs/lb COD
.48
.4
.34
.267
.218
.185
Sludge Yield   MGD Clarifier     Clarifier Flux,
            Draw @ .8% Solids  Ibs. Solids/Day
1.8
3
3.86
5
5.73
6,23
130,905
230,184
307,045
417,000
491,622
545,145
                   30 mg/1 Inert Suspended Matter in the Incoming Water Have the Following Effects
Inerts in
Wastage, mg/1
3333
4000
4667
6000
7333
8667
MLSS
Inerts
300
600
900
1500
2100
2700
MLSS
Total
1020
1800
2443
3500
4391
5192
(nerts/MLVSS
ratio
.417
.5
.583
.750
.9167
1.083
MGD Clarifier
Draw® .8% Solids
2.55
4.5
6.11
8.75
11
13
Equilibrium
MLSS, Ibs.
85,068
150,120
203,746
291,900
366,209
433,013
Clarifier Flux.
Ibs Solids/Day
191,828
367,794
531,981
839,213
1,135,249
1,428,942
Wastage Rates
Ibs/D gpd@ .8% Sol ids
17,014 255,000
15,012 225,000
13,583 203,583
11,676 175,000
10,463 156,821
9,623 144,222

-------
146
   Optional Refinery  Treatment  Sequence
                                             FIGURE
   Treatment
Pre-or Inplant
Treatment
Primary
Treatment
   Objectives   Phenolics,S=,NH3,    Free Oil and
                RSH, F~ Acid Sludge,  Suspended Solids
                Oil Etc.,Removal a Removal
                Water Reuse  or
                Waste Equalization
   Processes    Unit Separators
                Steam Stripping
                Fuel Gas Stripping
                 Air Oxidation
                 Neutralization
                 Surge Ponds
                   API Separators
                   CPI.PPI Separators
                         Sludges
Intermediate
Treatment	
Emulsified Oil,
Suspended and
Colloidal Solids
Removal
                   Chem. Coagulation
                   SAir Flotation	
                   Chem. Coagulation
                   S Filtration	
                                      pH Control
Immediate Oxygen
Demand Reduction

Equalization of
Wastes
1
Sludges
F
Tertiary
Treatment
                                                                    Variable
                                                                    Objectives
                                Chem.Coagulation
                                Q Air Flotation
                               Chem. Coagulation
                                8 Filtrotion	

                                Activated
                                Carbon
                                        Sludges
                                     Sludges
                            FIGURE  2.  Probability  Plots  of Oil a Grease
                                         Data Before  and After  Inter-
                                         mediate Treatment
                   o
                   •s.
                   a>
                   (A
                   o
                   a>
                   w
                   O

                   03
                      1000
                       800
                       600

                       400
       200


        100
        80

        60

        40



         20


         10
         8
         6
                            _  Raw Waste Load from
                               API  Separator
O
   o
                                                  Waste Load after
                                                  Intermediate
                                                  Treatment
                         0.5 I  2  5  10 20  40  60  809095989999.5
                             %Probability less than Indicated Value

-------
                                                                                 147
                               FIGURE  3

               Probability Plots of BOD Data Before

               and After Intermediate Treatment
   400
2  300
  M
TJ

O




a  200
 x
 O

 "o
 O

 'i
 a*

 o
 o

 'o
 100


 80
     60


     50
          ORaw Waste Load from API  Separator

          •Waste Load after Intermediate Treatment
        I   1
      0.5
          1
                      10    20  30 40 50  60  70  80    90

                          % Probability less than Indicated  Value
95
98  99 99.5
                         FIGURE  4

             Probability  Plots of COD Data  Before

             and After Intermediate Treatment
    1500







    1000



    800




    600






    400
o
E
a
o>
>>
x
O

"o
u

1
0)
200
    100
      0.5  I
                ORaw Waste Load API Separator

                • Waste Load after Intermediate Treatment
                                                                      O
                                                               I	I
                                                                      I    I
                      10     20  30  40  50  60  70  80    90

                         % Probability less than Indicated Value
95    98  99 99.5

-------
148
           ioo h
            90
            80
            70
                                        FIGURE  5


                    SVI Of Activated Sludge In Unit Treating Refinery
                            Effluent Pretreated by Filtration
        BOD

      Removed
   ( from Eckenfelder)
                                                      SVI (from
                                                     Eckenfelder)
                                 _L
            I
          SVI Found with
          Prefiltration
          li     ii
                     .15   .2     .3   .4   .5  .6    .8   1.0
                          Loading,Lbs.BOD/D/LB MLVSS
                   I   l    I     I	|	I      I
                  35 25   15   10        5       3     2

                        Sludge Age,Days (a = .6  £ = .05)
                                                                     300
                                          200  o>
                                                E
                                               _3
                                                O


                                          100  I


                                          50  «
                                   1.5
                              FIGURE  6

           Relationship of Activated Sludge
           SVI with  Zeta Potential In-Situ(Forster)
           -2lr-
           -20
        >
        E
       _o

        c
        o
        a.
        a>
       N
            -19
           -18
                     I
I
I
I
I
I
                   60    80    100    120   140

                      SLUDGE  VOLUME  INDEX
                         160    180

-------
                                                                  149
     400-
        FIGURE

Soluble  BOD and  COD in ASP
Effluent  at Various  Sludge Ages
                          20       30      40
                       Sludge Age,  Days
                           FIGURE 8
Schematic of a Bacterial Cell, Its  Biochemical Activities,
and Exoenzyme Solubilization of Insoluble Substrates
    Soluble Substrates

     Soluble Substrates
     from Actions of
     Expenzymes
            r^^T	
                   ~
                          ^  Permease Enzyme
                       f'f,  ^ Transport System
                                                        Cell
                                                        Membrane

                                                    \Wcell Wall
                                    CO^ and Other
                                    Metabolic By-Products

-------
  150
FIGURE 9
Schematic of Layers Observed in Bacteria Cell Walls and
Cytoplasmic Membrane (Approximately to Scale)
 • ••••••••*-••**" »•••••••
 '«••»• • • • • • •_•• • •«. • • • • • • • .•
 ',. • •	 • • • • *.	r
 *. • • ••••••*. •••.•!«••••«•»•
 •«•••••••. i • . • • i • ?. •••••»•!•
 • » • «.•» •«. i•.••••«•.•••••"•••
                                ::•:•:•:  }c
                                	>d
• Outarmost Layer, Protaim and Polysacch«rid«s
b Lipids
c Possibly Lipopolyuccharidas
d PouibJy Protains
a Fibrous H«taropolym«rs of Paptidoglyeans
f Pariplasmic Spaca Containing Dagradaiion Enzymas
g Outar Layar of Call Mambrana
h Middla Mambrana Layar, Bimolacular Phospholipid Layar
i  Innar Layar of Call Mambran* (--25 A°)
                A  C*puil< Matirul; 200-100,000+ A°
                B  Soft L«y«r. 10O-1SO A»
                C  Rigid Liy*r; -SO A°
                D  Pcriplumic Sp*c«; ~100 A°
                E  Cytoplwnic Mwnbnrw: -100 A°
                Pr  Proteins
                P,  Polyucchwido
                En  EniynM Trantport Synwm (PtnnnMil
                                           FIGURE  10
     Enzymes   Work  in  Sequentially  Cooperating  Systems
                                Enzyme Synthesizing
                                System
                        Feedback Repression
                                                                         Feedback  Inhibition

-------
                                           151
                   Figure 11
 Character of  Mixed Liquor Solids
 with  Sludge Age
6000
4000
2000
    0
Mixed L|quor Suspended Solids (MG/L)
Flow,20 MGD
Retention Time,12Hrs.
a = 0.4  - b = 0.05
COD Rmvd.,25000 Ibs/D
Inert Solids in,
so MG/L            ..^^ /Inert Solids
                         _  . , 0 ...
                         Total Solids
                     CM ounuo v   ^fff


                     	'"
                      Biological Solids
                                  6000
4000
                                  2000
     0   5  10  15  20 25  30  35  40 45
                Sludge Age (days)
                   Figure 12.
 Character of Mixed Liquor Solids
 with Sludge Age    	
6000
4000
2000
    0
Mixed Liquor Suspended Solids (MG/L)
Flow,20 MGD                   ....•
Retention Time,12Hrs.          ..•••""
a = 0.6  - b = 0.05       ...••'"
COD Rmvd.,25000 Ibs/D ...<,LTota| QOIJHS
Inert Solids in,     ..,•"   l1.^1^
so MG/L       .••'    Bioloical Solids
                            Inert Solids
                                 6000
4000
                                  2000
     0   5  10  15  20 25  30  35 40 45
               Sludge Age (Days)

-------
                     CASE HISTORY ON BIOLOGICAL TREATMENT
                                        OF A
                          PETROLEUM REFINERY WASTEWATER

                        Charles E. Ganze, Facilities Supervisor
                                         and
                        Joe P. Teller, Deputy General Manager
                          Gulf Coast Waste Disposal Authority
                                    Houston,  Texas

Gulf Coast Waste Disposal Authority's American Facility located  in Texas City,  Texas,
provides intermediate, secondary and tertiary treatment for a 23 million gallon per day
waste stream.

       The Research and Development and Process Design work was done by Amoco Oil and
its Parent Corporation, Standard Oil of Indiana.  A number of process patents have been
applied for, with several granted and others pending.  Williams Brothers Construction has
been licensed to construct similar facilities incorporating the patented processes.

       The wastewater treated at the American  Facility is generated primarily by Amoco
Oil Company's  Texas City Refinery which currently processes approximately 350,000 barrels
of sweet crude per day.  In addition to the wastewater generated in the crude refining
processes, sulfuric acid concentrating unit,  cokers, and other refinery related process, this
treatment facility treats wastewater generated by Amoco  Chemical Company's Texas City
Plant "B", other Amoco Chemical Company processing units located at the  Refinery and all
storm water collected in the 500 plus acres inside the battery limits of both Amoco Oil and
Amoco Chemical Company's Plant "B".

       To characterize the influent wastewater, parameters listed in the wastewater dis-
charge permits are used.   The data presented represent the values actually observed over
7 months operation.
                                Influent Characteristics

                           Average            Maximum          Minimum
BOD, mg/1                   183                1,110               36
TOC, mg/1                   118                 620               30
TSS, mg/1                    234                1,908              32
O&G, rng/1                 11                  110                1
NH.-N, mg/1               39.4                  100                9
Phenols, mg/1                6.7                   85              1.2
Sulfides, mg/1                0.4                  4.0              0.07
PH                                                 11                 2

       The various waste streams have received primary treatment prior to reaching the
Gulf Coast Facility.   In many cases, streams have passed through unit oil separators for
the  removal of gross quantities of oil before being discharged to the refinery's collection
system, and then at the end of the collection system, all wastewater is processed through

                                        153

-------
154


API Separators for removing the remaining oils and the bulk of the suspended solids.  As is
generally the case with refinery effluents, the waste treatment system receives a waste
stream that varies greatly in both quantity and quality.  The process design was based on a
knowledge that the treatment system would experience a highly variable waste load and
that more stringent effluent requirements  could be expected in the future.

      To describe the waste treatment system,  it is broken into three steps:   Intermediate,
Secondary,  and Tertiary.  The intermediate step incorporates the use of physical and chem-
ical processes which are designed to protect the  secondary or biological  step.  It consists
of equalization with pH  adjustment and prefiltration.   As in many waste treatment facilities
treating industrial  wastes, equalization is considered a necessity.   It serves  several functions:
first,  it helps smooth out some of the variations in types of wastes and waste strength; second,
it helps protect the secondary facilities from sudden surges in flow; and third, it helps re-
duce  the overall treatment chemical costs by allowing the various wastestreams to mix,  and,
in many cases, neutralize each other, especially in the case of pH.  The 4-hour equaliza-
tion time  provided is considered minimal  for adequate protection to the biological system.

      Prefiltration in this system  is performed by pressure-type multimedia filters.  The
design hydraulic loading to the prefilter system is 5 GPM per square foot.  Using only
organic polyelectrolytes as filter aids, the prefilters are designed to remove the suspended
solids and oils that remain after primary treatment and removal of much of the colloidal
materials.   The logic in keeping an extremely high quality water as a feed  to the biologi-
cal step is:  first,  to reduce the organic loading  by removing the suspended solids,  colloidal
materials  and oils; and  second,  to reduce the amount of inorganic solids that are put into
the system.   When the inorganic  solids and oils  are kept out of the biological system, the
settling characteristics of the biological sludge is improved; with a better settling sludge,
a higher sludge age can  be attained; thus we then start attacking the more  difficult to treat
organic compounds and ultimately obtain  biological nitrification.

      The backwash from the  prefilters is  clarified with  the overflow recycled to the  equal-
ization basin and the underflow (sludge) mixed with digested waste-activated sludge  and
pumped to a land farm for disposal.

      The product water from the  prefilters passes into the second step of our treatment
system, the  biological system.  At this point, two aeration tanks, which operate in parallel,
provide 12 hours retention time at design  hydraulic loading.  This reasonably long retention
time improves the  treatment capabilities for handling shock loads.  Aeration is provided by
ten 150-horsepower, floating, low-speed surface aerators.   From  the aeration tanks, the
mixed liquor passes into  the secondary clarifiers.  The two 140-foot diameter,  18-foot
side-water-depth-wide feed well  clarifiers operate in parallel.   The hydraulic loading on
the secondary clarifiers  is 600 gallons per square foot per day, which is generally considered
adequate.  The sludge removal mechanism is the multiple pick-up,  sludge sucker type.
The return sludge is pumped by centrifugal,  recessed-impeller type pumps.   Waste sludge
is pumped to a gravity thickener where the underflow passes into an aerobic digester, and
the thickener overflow is recycled to the  equalization basin.

      The overflow from the  secondary clarifiers then flows to the third step of our treatment

-------
                                                                                 155

system,  the tertiary.   The tertiary step consists of polishing filters.   These final or post
filters are designed simply to remove the biological solids that escape the secondary
clarifiers.   These gravity type,  dual media filters operate at a design hydraulic loading
of 4 GPM per square foot with the backwash pumped back to the aeration tanks as the
solids are of biological origin.

      Operations began before all construction activities were complete; we have experi-
enced and are continuing to experience, the usual start-up mechanical and design problems.
Among the significant mechanical problems are electrical switch gear malfunctions, an
aerator  gear box,  backwash sump pumps, a clarifier scum pump motor, and the underdrain
nozzles on the final filters.   Design problems that we have noted to date include: no
special  provision for operating in sub-freezing weather (freezing sample lines, service
water lines, caustic lines,  and wet-leads to instrument-primary devices); and a widely
fluctuating float in the Parshall  Flume which is caused by excessive turbulence as a  result
of too short an approach  to the throat.   Even without all equipment and facilities fully
operable, a number of interesting observations have been made.  Over a three month
period,  when mechanical problems were at a minimum, we observed what we  consider
phenomenal  reductions for a secondary waste treatment facility.

                         BOD             92%
                         TOC             85%
                         Ammonia-N      88%
These removals were experienced while maintaining the following conditions:

                         MLSS            3400 rng/1
                         Sludge Age       Approximately 50 days
                         F/M             0.07 pounds BOD/pound MLSS

with no apparent deterioration in sludge settleability (Average SVI -85).

      The average performance for the first seven months of operation was as follows:

                                 REMOVALS
                 BOD Reduction                      88%
                 TOC Reduction                      77%
                 TSS Reduction                       94%
                 Oil & Grease Reduction              91%
                 Ammonia-N Reduction               51%
                 Phenols Reduction                   96%
                 Sulfides Reduction                   72%

It should  be emphasized  that much of the data collected for these calculations do not
reflect  a  stable  operation.

      We feel that by completing the remainder of the construction, clearing up all the
initial mechanical problems generally associated with start-up, and by continuing to
improve on our own operational  procedures, we can meet the conditions set forth in the
existing permits which extend through June 30,  1979.   Beyond this, we foresee even more

-------
  156
stringent requirements for BOD, TOC, and Ammonia-Nitrogen.   Provided our treatment
capabilities improve as  expected,  this facility will produce an effluent that will comply
with anticipated future  requirements.

DISCUSSION

J. J. Chavez : How much ammonia do you have in your raw influent and  in the final efflu-
ent stream and where  do you think you get the bulk of ammonia removal in your system?

Charles Ganze:  The  ammonia is removed, biologically, in the aeration tanks.  The average
influent concentration (feed to the aeration system) was about 40 ppm with a maximum of
about 100 ppm or so.

J. J. Chavez: How about the final  effluent?

Charles Ganze:  The  final effluent -  it would go into the  .05 ppm range,  very difficult to
analyze.

Anonymous:  Please discuss your choice of filters.

Charles Ganze:  Economics is one thing.  The pressure type filters are a little more expensive.
You can give them a little bit higher loading.  The multimedia does give us some advantages
on handling the colloidal  materials and the suspended solids.   Now the gravity-type filter
again is used on the effluent,  again the only  thing that we anticipated getting from that were
biological solids.

G. T.  Minnick: How do you dispose  of your  aerobically-digested sludge?

Charles Ganze:  The  sludge that we  actually waste             is a very minimal  amount.
As Jim indicated we observe very low sludge  yield,  we are operating again at 50-day
sludge age at an elevated temperature and we have a low  sludge yield.   The aerobically
digested sludge is mixed again with this back  wash from  the prefilters and  it is pumped to a
land farm area.   Now we call this a  land farm  and not a land fill.   It is  actually  tilled
into the soil.  At this facility we  have some  150 acres dedicated  to this activity.   We  just
take a tractor with a disk  out there and till it into the soil.

R. A. Farnham:  What are the calculated variability factors associated with your process?

Charles Ganze: We don't think we have enough data to try to put together any thing like
that, I only showed you some maximums and minimums that we have observed over this
relatively short period of time, which is seven months, and we haven't tried to put together
any probability type numbers.

Anonymous:  What is the loading on your secondary clarifier?

Charles Ganze: 600 gallons per day  per square foot, surface loading.

-------
                                                                                   157
Leonard Orwin:  At the beginning of your talk you mentioned that there were several
features of this plant for which you had applied for patents, I wonder if you are able to
point out any of those?

Charles Ganze;  Standard Oil, since they did the R & D and design work, has  applied
for the patents.  I'll  ask Jim Grutsch if he will come back  up here and speak on that for
just a moment.

Jim Grutsch:  More  than 20 claims are allowed  in the patent covering areas of (1) pre-
treatment, (2) the activated sludge process, and (3)  the integrated end-of-pipe treatment
sequence.    Some of  the specifics are as follows:  Pretreatment  (1) The use of a portion of
the pretreatment basin for reduction of immediate oxygen demand (IOD).  (2) The use of
certain catalysts for IOD removal. (3) The addition of colloid destabilizing chemicals follow-
ed by filtration of suspended solids and oil and grease to levels  of 10 mg/1 and  less.
Activated Sludge Process (1) Two stages of sludge treatment in  facilities for one unit; i.e.,
recycle of part of the clarifier  return sludge  to: (a) the aeration tank rundown line, and/or
(b) a thickener thence to the aeration tank rundown line, and/or (c) a thickener, aerobic
digester, thence to the aeration tank rundown line.   (2) Operating at a  sludge age greater
than 10 days.  (3) Pretreating the activated sludge unit influent to 10 mg/1 or less suspended
solids and oil and grease. (4)Using air or oxygen between stages of the activated sludge unit.
For example, turbulence in  the wide well region of the clarifier was visible in the slide.
This turbulence is intentional.   Air is educted into the buried rundown line so that the air/
water mixture is under a hydrostatic head to  improve  oxygen transfer. When the air is re-
leased in the wide well clarifier,  it provides flocculating energy.   Process.  A process
consisting of an equalization section  incorporating hydraulic equalization, pH control,
pre-aeration, colloid destabilization, and filtration  to 10 mg/1 or less of suspended solids
and oil and  grease, followed by an activated unit incorporating two stages of sludge recycle
and interstage aeration, in which the average sludge age is greater  than  10 days.
       A Tulsa based firm,  William Brothers Waste Control, is authorized to use these feat-
ures.
BIOGRAPHY

       Charles W. Ganze is Facilities Supervisor at the
Gulf Coast Waste  Disposal Authority's 40-Acre and
American Facilities in Texas City,  Texas.   He has
acquired two degrees from East Texas State  College. He
holds a B.S.  in General Science/Biology and a M.S. in
Earth Science with a minor in Biology.  Charles taught
science in the LaMarque Independent  School District
(1967-1968) and the Andrews Independent School
District (1966-1967) and was employed by Union Carbide
Corporation as a Secondary Waste Treatment Research
Worker prior to his employment with the Gulf Coast Waste
Disposal  Authority in 1973.

-------
158

BIOGRAPHY

       Joe P. Teller is Deputy General Manager in
charge of Operations at the Gulf Coast Waste Dis-
posal Authority in  Houston, Texas .   He holds a
B.S.  Degree in Civil Engineering from Texas Tech-
nological College  and  a M.S. Degree in Sanitary
Engineering from the Georgia Institute of Tech-
nology.   Mr. Teller was employed as Deputy
Director of the Texas Water Quality Board in
Austin before joining the staff of the Gulf Coast
Waste Disposal Authority in 1972.

-------
           OPTIMIZATION OF THE ACTIVATED SLUDGE PROCESS
                         THROUGH AUTOMATION

                           Mohamed A. Zeitoun
         The Dow Chemical  Company, Texas Division, Freeport, Texas

    As an introduction, I would like to go over our accomplishment's in extending fhe
application of the activated  sludge process for the treatment of brine wastewaters.

ACTIVATED SLUDGE TREATMENT OF WASTEWATERS OF HIGH SALT CONTENT

    Many petrochemical processes produce organically contaminated wastes of up to
12% NaCl .  Glycols and glycerine made by the chlorohydrin  process result in a brine
waste of 6-10% NaCl and constitute the major organic waste  load at the Texas Division
of The Dow Chemical  Company.

    Up  to 1969, operation of the activated sludge process was believed to  be possible
at salinities  less than 2% NaCl .  Although numerous bacterial species living in very high
salt concentrations were known, utilization of halophilic bacteria in a biological
oxidation process was lacking.

    Under an EPA grant (1969-1971) we started with a fill and draw procedure  to find
the minimum dilution required.  We started with sewage  sludge and slimes from ditches
and holding  ponds exposed to the wastewater from the glycol production facilities,
and at 2% NaCl concentration. Then we raised the  NaCl  concentration gradually, over
a period of 6 weeks, to  10% salt. To our surprise, no  dilution was required. Then we
operated laboratory continuous units, obtaining 80-88%  TOD  removal at a  residence
time of  12 hours, a good settling sludge, but a turbid effluent. A  chemical flocculation
step was required to clarify the effluent and  maintain low suspended solids.

    We then built a  mini-plant on a skid for a flow rate of 0.5 to I . 0 gpm and fed it
continuously  from the production facilities.  We then learned  the necessity of good
equalization.  The mini-plant (Figure  I) is essentially equalization-aeration-bio-settl ing-
flocculation and a final settler.

    At  that  point in  time we believed  that activated sludge can be extended to the
treatment of brines up to  12% salt content.   The operation was checked in a pilot plant
(150 gpm) using commercial size equipment.   Its operation continued for over two years
where several  sludge handling  methods were  tested and it served as a technology center
for the design of several treatment  facilities  around the world. We spent about $1 million
on this testing facility.

    In the Texas Division at Freeport,  Texas, we utilized existing holding  ponds and
built a mammoth aerated lagoon system using submerged  Kenix mixers.   It is treating
20,000 gpm  in 45 acres of aerated  basin at a capital  cost of about $5.5 million.  Now
we are adding air flotation,  settling tanks and pressure filters  to remove  the suspended
solids  from the aerated lagoon  effluent at a capital cost of $7 million.

                                     159

-------
 160


     At Stade near Hamburg, Germany, we builf an activated sludge planf to  treat a mixed
 effluent of glycol, glycerine, and other minor streams,  which were tested in mini-plant
 operation.  This plant has been in operation for  about 2 years.  It is designed  for 9,500
 gpm at a cost of $7 million.  Another plant built at Sarnia,  Canada, has been in operation
 for about 4  months, and  is treating 3,000 gpm at a capital cost of $4.5 million.

     The Louisiana Division at Plaquamine is building a  Unox activated sludge plant at a
 cost of $8.5 million for treating 8,500 gpm and  is scheduled to start early 1977.  A total
 of $33 million has been invested  in plants treating 60 million gallons per day, (Table I),
 of brine wastes all  over the world.  All this resulted  from an initial study sponsored by an
 EPA grant costing  the taxpayer $150,000.  This is the story of  the best  spent $150,000 by
 EPA.

 AUTOMATION OF THE ACTIVATED SLUDGE PROCESS

     The problems  of automated control of activated sludge plants lie in selecting parameters
 for control and developing sensors for these parameters.  The common practice of monitoring
 plant operations with random daily or weekly analysis is inaccurate and clearly inappropriate
 for control and optimization.  Parameters to control the process  must be measured on a con-
 tinuous basis or must be obtained rapidly enough on a repetitive basis so that updated data
 are available for use within the time          constraints of  the  process.

     The objectives of the reported study were to develop on-line control systems that will
 (a) add nutrients (N&P)  in proportion to  the total carbon in  the  feed, (b) control  the sludge
 recycle rate by an F/M signal that is measured as the ratio of  the total carbon in  the feed
 to that in the mixed liquor, (c) control the chemical flocculant addition at  the end of the
 process in proportion to the turbidity of the bio-settler overflow, (d) divert the feed to a
 holding pond  if a  toxicant is detected in the feed by an instrument "Biological  Inhibitor
 Detector",  that measures the oxygen uptake of the biota on  repetitive  basis.

     The existing,  skid-mounted mini-plant was  renovated and new electrical circuits for
 the instrumentation were added.  The control panel (Figure 2) included three 2-pen recorders.
 The influent flow,  set  manual'y at the board,  and the pH of the influent were  both recorded.
 The sludge recycle was manually controlled and  recorded.  The alum flow rate and the
 turbidity of the final effluent were both recorded and controlled at the board.  The pH of
 the final effluent was both recorded and  indicated on the board.

 Influent pH  Control
     The influent pH control system (Figure 3) was designed for fast response in a well
mixed system.  By coupling the pH feedback loop around the recycle feed pump,  the loop
capacity, mixing time and measurement lags were kept to a  minimum.  The pump also
served as the feed pump to the aeration basin and provided the feed at the  proper pH
independent of the volume and pH of the influent in the equalization tank. The pH
control  system had a high and  low limit switch that  closed the feed valve if the pH  got
out of control and automatically opened the valve as the control range was regained.

-------
                                                                                161
Food to Microorganisms (F/M) and Nutrients Addition Controls

    The F/M control system is shown schematically in Figure 4, together with the
nutrients addition control system. An Ionics Model 1212 Total Carbon Analyzer was used
to measure the total carbon in the feed and that of a homogenized and diluted mixed
liquor sample from the  aeration basin.  To achieve this, the sample  handling system of
the  carbon analyzer was modified to accommodate two sample streams,  one a continuous
flow and the other a batch sample.  A schematic of the  sample handling system is shown
in Figure 5.  A continuous sample of the  feed passes through a 3 micron Cuno(R)  filter,
through a valved rotameter and through a two-way solenoid valve (A).   It then passes
through the Analyzer sample injection slide valve to a 3-way solenoid  valve (B)  and on
to drain.  A 1:1 dilution of the mixed liquor sample aids in the  homogenizetion step
required to obtain a representative sample.  Two timers  operate sequentially to first
operate pumps to provide the aliquot and next operate the Virtis(R) 45 homogenizer.
Upon completion of the homogenization time, another sample pump  empties the homo-
genizer contents into the stand-pipe above  solenoid (B), which allows deaeration of the
sample before analysis. All three solenoid valves are energized at this  moment;  two-way
solenoid (A) is closed and the drain connections of valves  (B) and (C) are closed. The
homogenized sample passing through the stand-pipe above  valve  (B)  flows by gravity
through the sample injection valve and into the stand-pipe above solenoid valve  (C)
until the stand-pipe levels approach equilibrium.  Flow in this direction flushes the
prior feed  sample from  the injection valve.  Upon injection of the homogenized sample,
all of the solenoid valves are de-energized, restoring the  flow of the continuous  feed
sample stream through  the system.  Meanwhile,the batch sample  in both stand-pipes is
drained.

    The output from the Total Carbon Analyzer is converted to  a pneumatic F/M signal
with a 0 to 1 range.  The signal is fed to a recorder-controller which adjusts an output
pneumatic signal proportional  to the offset  between the  desired  and  actual F/M valve.
The output signal controls the  percent energized time of a 60-second cycle timer, that
operates a  normally closed solenoid valve allowing the  flow of the recycled sludge  to
waste.  The system also includes a high signal selector and regulator that is set to
designate a maximum waste flow rate independent of the F/M controller.  This safety
device prevents bacteria washout in case of control system failure

Nutrients Addition Control: Constant nutrients addition, based on flow rate, will
occasionally result in excess or insufficient nutrients addition.   Insufficient addition
decreases bacterial viability and decreases the organic  carbon removal efficiency.
Excess nutrients addition,  besides being economically unsound,  can cause downstream
algae blooms and subsequent pollution problems.  High  nutrients residual may also
prevent the recycle of industrial treated effluents.

    The activated sludge mini-plant was operated with  varying  flow of nutrients to the
aeration basin based upon the  measurement of the total  carbon in the feed.  Ammonia
and phosphorus concentrations, in the biosettler overflow, were analyzed  routinely to
determine  the optimum C:N:P ratio, for the waste water being treated.  Figure 6 shows
the  relationship between ammonia consumption  and microbial growth.  Although the

-------
162

experimental data reflects the ammonia consumption for cellular growth plus the ammonia
absorbed onto the bio-floe, the experimental  value of 0.133 compares closely with the
theoretical  value of 0.15 according  to cell formula Cj-
     In order to determine the minimum nutrients  requirement, the ratio of nutrients
addition to  feed total carbon concentration was progressively lowered until analysis of the
bio-settler overflow showed a very low concentration of N and  P.  The results shown in
Figure 7 indicate a large adsorptive capacity of the mixed liquor bio-floe for values of
N :C greater than 0.2.  The C:N:P ratio of 100:9:0.54 was found  to be optimum for the 8%
salt containing wastewater.  Using this proportionally the nutrients control system assured
microbial viability and the effluent concentration was maintained below 5 ppm ammonia
and I  ppm phosphorus.

F/M Control System:  A transient or unsteady state model of the completely mixed
activated sludge process is developed to simulate the dynamics of  the F/M control system.
The Monod  kinetic equations are incorporated into the substrate and microorganisms
material balance equations around  the aeration basin.  The substrate degradation is assumed
to occur only  in the aeration basin.

    The scheme of F/M proportional control  is inserted into the model as a function of the
wasted sludge flow rate.

    W = A  (F/M)  + B,  O - W - W
            v    '     '              max.
     where     W - wasted sludge  flow rate
               A and B are constants  related to a specified gain.

A maximum  waste  flow  rate, W      is imposed to prevent complete cell  washout in case
the control  system should fail.  The' control is related to the recycle flow by:

    T=  R+ W
    R  =  T - [A (F/M) + B]

    where     T  = total biosettler underflow rate
               R  = recycle flow rate

The dynamics of the unsteady state bio-settler operation are computed in  terms of a con-
stant compaction  ratio, X /X.  There are obvious limitations in  using a constant ratio  to
describe the unsteady state performance of a  biological settler,  however it was adequate
for the predicfionof the  bacteria concentration in the aeration basin.   Figure 8 shows that
the sludge compaction ratio remained relatively constant during a  50% increase and
decrease in  feed concentration.

    The  Monod type kinetic equations were found adequate in simulating the dynamics
of the activated sludge  treatment of propylene glycol waste water because of four basic
reasons:

-------
                                                                                 163
    I.  The Monod model was developed to describe the behavior of pure cultures.
        In the fresh water glycol system, 80% of the microorganisms present were of
        one species,  Pseudomonas.  In the brine  (6-8% NaCl) glycol system 90%
        of the culture was identified as one halophilic bacterium.

    2.  The propylene glycol represented the only soluble carbon source in the feed.

    3.  The carbon energy source was rate limiting.  Deficient nutrients (N&P) were
        added by the  nutrients control system.

    4.  The pH and temperature were regulated for the  optimum rate of biological  growth

Sludge growth coefficients and substrate removal  kinetic constraints were determined for
freshwater glycol feed and saltwater glycol feed.  For the freshwater biota,  the yield
coefficient Y = 0.21 gm VSS/gm TOD removed, and the decay coefficient k . = 0.06
day   , while for the saltwater bacteria, these values were 0.133 and 0.02 respectively.
The substrate utilization  rate for the freshwater system is best described by:

    dF -  k xS]
    dT
     where k = 5.0 day

       and K = 117  mg/l as TOD

 For the saltwater bacteria, K  is very large compared to S.. (substrate concentration in the
 aeration basin),  and the rate equation reduces to:

     dF = k x  $
     dT       '

     where k = 0.0132 (mg/l TOD)'1 day''
 Testing of the F/M Control System: The mini-plant test runs were started with  the F/M
 controller on manual,  i.e., constant recycle and sludge waste flow rates were set.  The
 system was allowed to approach a steady-state conditbn,  in terms of reactor microorgan-
 ism concentration, substrate removal efficiency, and bio-settler characteristics such as
 sludge concentration and sludge blanket height.  The system was then put on automatic
 control and a step  increase in  the concentration of the feed was made by adding an
 aliquot of industrial grade propylene glycol  to the equalization tank.

     An automatic  Total Carbon analysis was made every six minutes during these runs,
 with alternate analysis of feed sample and homogenized mixed liquor sample.  To reduce
 the volume of data, each F and M concentration plotted in this section  is represented as
 the average of five total carbon analyses.  Waste, recycle and F/M data are also hourly
averaged values.

     A blank  F/M test  in which no control is affected shows the dynamics of cell growth
under continuous flow conditions and  is used for comparison for F/M  control test system

-------
164


 response.   Figure 9 shows the mixed liquor total carbon and F/M response to a 47% increase
 in loading.  The solid lines are the predicted responses using fhe transient state computer
 model.  A very good fit is obtained for the increase in mixed liquor total carbon.  Because
 of a non-linear calibration of the total carbon instrument above 50% of scale, data for F/M
 signal lie below the computer prediction of F/M, till  the mixed liquor total carbon values
 reach the same  non-linear range as the feed values, then the predicted curve agrees with
 the experimental data.

     The results of one of the control tests made with freshwater gIycoI feed are presented
 in Figures 10 and II.  The initial conditions of this test run  compare closely with those
 used  in the blank test (Figure 9).  The model  prediction of  the  increase in mixed liquor
 total carbon (t = 0 to t = 27.5  hours) fits the data well.  The F/M signal data  lie above
 the predicted F/M curve for the first several hours following the step increase and below
 the predicted curve following the step decrease, because an average value of feed con-
 centration is used in  the simulation, rather than  individual  data points.  A good represen-
 tation of recycle and wastage response  is obtained with a fast F/M  response (Figure 11).
 The prediction of the transient biological state after the  step decrease  in feed concentra-
 tion (t = 27.5 hours)  is not accurate because of an inadequate representation of the bio-
 settler dynamics.  However, the final steady  state computer prediction ( t =50 hours) fits
 the data well.

     In order to have a meaningful basis of comparison for the F/M  control tests,  it is
 necessary to have a blank run with identical initial conditions.  The initial  steady state
 must be the same in terms of feed, recycle  and waste flow rates, substrate concentration,
 reactor and bio-settler underflow microorganisms concentrations, temperature, and
 physiological conditions of the  biota.  While this is difficult to achieve experimentally,
 a valid comparison can be made by using the  mathematical  modeling which predicts the
 transient accumulation of microorganisms.

     The computer simulated curves in Figures  12 and 13 show the differences in response
 characteristics for the freshwater propylene glycol system with  F/M control versus the
 system with no control.  For the case of an 8.5-hour hydraulic  residence time with an
 initial recycle ratio of 0.55 it takes 17 hours  for the controlled system  F/M to return  to
 within 20% of the set point after the step increase in loading (Figure 12). This represents
 a 63% reduction in response time.  Figure 13  represents the system having a  6.95-hour
 hydraulic residence time and an initial recycle ratio of 0.3.  The F/M ratio returned
 to within 20% of the set point  in 10 hours,  less than a third of the time required for the
 system without control.  The larger mass of waste  sludge available for recycle allowed a
 69% reduction in response time.

     One of the most important achievements of this control system  is maintaining  the
 effluent quality.  Figure 14 shows the relationship between  substrate concentration in the
 aeration basin for the blank and the F/M control simulation of  Figure 13.  The control
 enables a fast return to the desired effluent quality.

     Figure  15 shows the simulation of an F/M blank test and an F/M control test for the
 brine propylene glycol  waste water system.  Low volume  of waste sludge is available for

-------
                                                                                165
additional recycle, due fo the lower yielding halophilic bacteria,  which results in the
rather  long response time of 33 hours with F/M control.  Manual adjustment of the recycle
and waste flows would be adequate for loading control because of the long  response times
involved. On-line  F/M control for high salt systems does not seem justifiable unless an
aerated sludge storage or stabilization tank is available to supply additional recycle in
periods of high loading.

Chemical Flocculation Control System

     The overflow of the bio-settler is too turbid,  especially in case of saline wastewaters,
and needs additional clarification.  A chemical flocculation control  system was used to
remove the suspended and colloidal  matter from both  the fresh water and the highly
saline  biologically treated effluents.

     The schematic in Figure 16 illustrates the flocculation controls and process flows.
The proportional  control was based on a  turbidity measurement.  Aluminum  sulfate was
used as the flocculant, and the pH was controlled (proportional-integral) at the level
corresponding to  the optimum for alum flocculation.  Both feed forward and feed back
modes  of control  were  tested.  A Hach Surface Scatter Turbidimeter,  No. 2426,
operated better than other types tried because it has no contact  between the sample and
the optical surface.

     The optimum pH for alum flocculation was determined by a  series of jar tests at
various pH values,  on  samples having the same initial turbidity.  The optimum alum
dosage was found to increase above  pH  6 and below pH 5.  At a pH above 7.0,  the
settled floe was more fluffy and had a higher subsidence line.

     A  stoichiometric relationship  was found between  the bio-settler overflow turbidity
and the optimum  alum dosage, at constant pH (Figure 17).   Comparing point 5 with point
6 in  this graph illustrates the effect  that the physiological  condition of the bacteria has
on the  alum requirement for optimum flocculation.  Data point 5 represents a series of
jar tests on a sample of bio-settler overflow whose initial turbidity was increased from
25 JCU to 44JCU by adding aeration basin mixed  liquor.   Thus the sample contained
larger  bio-colloids of older microorganisms.  On the other hand, data point 6, where
the higher optimum alum dosage was required,  represents a  case of finely dispersed
microorganisms, as the sample was collected during a period of very high growth rate
of microorganisms in the aeration basin.

    Because of the lag time of several hours associated with the flocculator and final
settler, control stability was a major problem in feed  back control.  The feed forward
mode of proportional control resulted in a high quality effluent and stability was not a
problem.

     The feed  forward control was  tested on both the propylene glycol fresh water system
and the glycol-salt waste  water.   Typical performance is given in Tables 2  and 3
respectively.  An average of 98% of the suspended solids were removed by  the floe-

-------
166
culaHon step in the fresh water system and 85% were removed in the salt system.  The
chemical flocculation step increases process efficiency in terms of TOD removal.  For
the fresh water system the overall process efficiency increases from 90.9%  TOD removal
to 93.4%.   The turbidity of the final  fresh water effluent was reduced  to 2 to 3  JCU
whereas the bio-settler overflow turbidity was typically 15 to 25 JCU.   The bio-settler
overflow turbidity for the saline waste water system was reduced by 85%,  from  35  to 80
JCU to 5 to 10 JCU.

Biological  Inhibitor Detector

     The objective was to develop an  upstream sensing device for toxic loads or  spills to
an activated sludge process.  If the presence of a toxin in the feed to the  plant  can be
detected rapidly enough on a repetitive basis, the feed could be diverted  to a holding
pond until  the toxin  is identified and  eliminated, or bled at a low rate to  the treatment
plant.

     Oxygen uptake  rates, both Warburg and oxygen electrodes, have been used to
measure the effect of toxins and inhibitors on activated sludge.  Tests were conducted to
determine an oxygen uptake measurement cycle that would be sensitive to the presence
of a toxic substance  in the feed to an activated sludge plant  and that could be automated
to rapidly detect the toxic effect on a repetitive basis.

     The oxygen uptake rate and the total oxygen consumed by propylene glycol standard
solutions containing  various concentrations of  toxic substances were measured as 10 ml of
the sample  was introduced  in 500 ml of activated sludge  under controlled conditions of
mixing, aeration, and temperature.

     The sensitivity and reproducibility of the  total  oxygen consumed, as determed from the
area under the oxygen uptake curve,  was found superior  to that of the oxygen uptake rate,
as determined from the initial slope of the oxygen concentration curve  when the aeration
was cut off.  The rate of oxygen uptake was more dependent on the concentration of  the
viable biomass than the total oxygen consumed.  The automation of a measurement cycle
utilizing the area under an oxygen uptake curve was simpler  since it did not require  cutting
off the air  every time a test sample was added.

     In order to utilize the area under the oxygen uptake curve to  monitor the toxic effects
of a feed to an activated sludge plant, it is necessary to  distinguish between the changes
due to  the  varying concentrations of the  substrate in the  feed and the presence of toxic
substances.  A  comparison of the oxygen uptakes of two standard propylene glycol  samples
before  and  after the  exposure of the bacteria to a feed sample was the measurement cycle
adopted to  measure the toxic effects of the feed.  The oxygen uptake of the first standard
established a reference activity of the bacteria sample,  which was then exposed to the
feed to  the process for long enough time  to degrade the substrate in the  feed.  Then the
same bacteria sample received an equal volume of the standard and its  oxygen uptake,
compared to that of the first standard,  was expressed as an activity ratio in response  to the
feed sample.

-------
                                                                                 167
    Figure 18  is a schematic  illustration of the measurement cycle (60 minutes), when
the feed contained no toxic substance.  The activity ratio would be approximately
equal  to 1.0.  Figure 19 shows a measurement cycle when the feed contained a toxic
substance  resulting in an activity ratio much  less than 1.0.

    Several models of an instrument based on the automated measurement cycle of three
oxygen uptake curves were built in-house and used extensively in screening the bio-
degradability and toxicityof industrial waste waters,  organic and inorganic substances.
The instrument can be regarded as a differential respirometer, and has the added feature
of measuring a BOD value of the  feed in less than  one hour.  The feed BOD value,
calculated from the ratio of the feed oxygen  uptake to that of the first standard multiplied
by the BOD of the standard,  compares very well with  BOD,- for values of 100 to 800 mg/l
(Table 4).                                              3

PROCESS  DESIGN CONSIDERATIONS -  CONTROLLED ACTIVATED SLUDGE

    The development of on-line control systems for the activated sludge  treatment process
opens up many possibilities for flexible and innovative process design.

    Figure 20 illustrates a conceptual design of an activated sludge  treatment system
including  the  F/M control system and the  biological inhibition detection (BID) system.

    Equalization dampens fluctuations in  the organic  content, in flow,  in pH,  and in
other  physical-chemical  characteristics of the feed, thus it is parr of the F/M control
system. The equalization tank is also partof the feed forward inhibition detection
control system.  It provides a time delay,  together with dilution of a  possibly toxic
feed during the biological activity analysis period. A toxic feed would be diverted to
a holding  basin by the  BID control system. After determination of the chemical nature
of the toxin, it could either be bled back into the system or treated by other means.
Pre-treatment could include chelation of toxic metals and sorption of toxic organics.

    The F/M  control system, based on flow measurement and organics concentration,
includes an aerated waste stabilization tank which supplies additional microorganisms
for recycle in  periods of high loading.  The stabilization tank may be located in the
sludge waste line as shown in Figure 20, or it could be located  in the sludge recycle
line; depending on which site is more adequate in  terms of microorganisms viability and
effects on bio-flocculation.

    If the feed sample for F/M control  is  taken before the equalization tank^ F/M
control will be more anticipatory in nature with very  short response time.  The  control
system could then include a mini-computer which would be programmed to  calculate
the hydraulic effect that the  equalization will have on the feed  flow and composition
and would affect  F/M control on  this  anticipatory basis.   Hence, extra sludge recycle
could  begin before the higher loadings affect the aerated biomass.

-------
168


     Another available control system that needs to be incorporated into the plant design
is an aeration  basin oxygenation system that optimizes oxygen allocations and controls the
oxygen supply to  the stabilization tank.

     Nutrients and pH control would be provided when  required.  Turbidity control would
be provided for designs that employ flocculation as the tertiary treatment for colloidal and
suspended solids removal.

     In conclusion, future plant designs have to incorporate both control systems and
predictive models in order to optimize the operation of the activated sludge process to be
able to meet the more closely regulated effluent limitations.  Both design  and operation
of sophisticated plants depend  upon the availability of  reliable sensors and the instrumen-
tation  necessary  to control the processes.

REFERENCES

     Details of the research reported in this paper are given in two reports published by
U.S. Environmental  Protection Agency and are for sale by the Superintendent of Documents,
U.S. Government Printing Office,  Washington, D.C.  20402.

(I)  "Treatment of Wastewater from the Production of Polyhydric Organics, " 12020 EEQ
     10/71 Water Pollution Control Research Series (1971), includes 18 references.

(2)   "Optimizing  a Petrochemical Waste Bio-Oxidation System through Automation,"
     EPA-66-/2-75-02I Environmental Protection Technology  Series (June  1975), includes
     61 references.

ACKNOWLEDGEMENTS

     The  work  reported was performed at the Texas Division of The Dow Chemical Company
by the  author as project director, and N. J. Biscan, J. H.  Culp, H. C.  Behrens, W. D.
Spears, and R. W. Murray.  W.  F.  Mcllhenny was the  Project Manager.

     These projects were partially supported by a grant  from the Water Quality Office of
the Environmental Protection Agency.  Appreciation  is expressed  to the  personnel of
Roberts. Kerr Environmental Research Laboratory, Ada, Oklahoma, for their cooperation
and assistance.

DISCUSSION

Morris  Wiley :  Was this a calcium or a sodium brine?

Mohamed Zeitoun: Mainly sodium chloride.

Morris  Wiley:  Have you tried treating this in combination  with other organic waste?

-------
                                                                                  169
Mohamed  Zeitoun:  Yes.

Morris Wiley:  Is that desirable or undesirable?

Mohamed  Zeitoun: That depends on the same factors governing the  possible  combination
of fresh wastewaters.  Our aerated  lagoon treats a mixture of organic waste that were
tested  in the laboratory and  determined  to be compatible and treatable.  Some organic
wastes were found to be non-toxic to the system,  but don't degrade in the presence of
the main substrate which  is glycol and glycerine.

Morris Wiley:  Do you notice either positive or negative synergism  between  the  waste-
waters, by that I  mean you get enhanced activity or decreased activity by mixing the
brine waste and other types of waste?

Mohamed  Zeitoun:  Do you mean by other type of waste, that which is  less  saline?

Morris Wiley:  Yes.

Mohamed  Zeitoun:  Micro-biological tests showed  that the optimum salinity of this
halophitic bacteria is 6-7% NaCl .

Morris Wiley:  Do you have problems with salinity swings causing osmotic pressure
difficulties on the organisms?

Mohamed  Zeitoun: Gradual changes of salinities up to  10% and down to  4% have very
slight change of activity.  Practically,  this is not  a problem.

BIOGRAPHY

     Mohamed A.  Zeitoun is  a senior research
specialist  with the Dow Chemical Company, Texas
Division,  Freeport, Texas.  He holds the following
degrees in  Chemical Engineering:  B.Sc. Cairo
University, Cairo,  Egypt (1950),  M.S. and Ph.D.
(1958)  Texas A&M University.  He taught at Cairo
University and Texas A&M.   Mohamed has  been
conducting research work in  desalination and
industrial  waste water treatment since 1962 and  has
been with Dow since  1966.

-------
170
                  TABLE!  BIOLOGICAL OXIDATION PLANTS TREATING BRINE WASTES
                                 IN THE DOW CHEMICAL COMPANY
Process
Activated Sludge
Pilot Plant
Aerated Lagoon
A e ra t io n
Sludge Handling
Activated Sludge
Activated Sludge
Unox Activated
Sludge
Location
Freeport, Texas
Freeport, Texas
Stade, Germany
Sarnia, Canada
Plaquamine, LA
Design Flow
Rate, gpm
150
20,000
9,500
3,000
?,500
Capital Cost
Million S
1.0
5.5
7.0
7.0
4.5
8.5
Remarks
Incl. 2 years operation
Start-up late ''976


Start-up early 1977
          TOTAL
60 MGD
$33.5 million
                         TABLE 2 ALUM FLOCCULATICN SYSTEM PERFORMANCE*
                                   GLYCOL-FRESH WATER SYSTEM
                                    Effluent Before
                                 Chemical Flocculation
                                   Effluent After
                               Chemical Flocculation
Day of
Operation
1
2
3
4
5
6
7
8
9
10
MLSS
mg/1
45
22
20
20
11
36
52
86
34
55
TOD Removal
%
88.8
95.5
93.5
89.3
92.0
95.2
84.!
90.7
88.3
91.7
MLSS
mg/1
4
0
0
0
0
1
0
0
0
0
TOD Removal
%
89.7
96.0
95.8
93.0
93.9
95.4
87.7
94.9
94.0
93.9
        * proportional control with optimum alum dosage according to Figure 17.

          effluent turbidity maintained a t 2 to 3  JCU

-------
                                                                      171
         TABLES ALUM FLOCCULATION SYSTEM PERFORMANCE*
                  GLYCOL-SALT WASTE WATER SYSTEM

                       Effluent Before                  Effluent After
                   Chemical Flocculation           Chemical Flocculation
  Day of            MLSS   TOD Rerroval            MLSS   TOD Removal
Operation            mg/1        %                mg/1         %

    I                  91       76.2                 24       91.9
    2                 164      78.9                 17       91.9
    3                  52      63.2                  0       65.9
    4                  43      66.6                  3
    5                  23      90.3                  0       90.8
    6                 214      75.8                 18       96.3
    7                  66      71.5                  o         -
    8                  43      69.0                 19
    9                  52      75.4                 21       80.3
  10                  40      78.4                 18       81.0

  *  alum dosage at optimum (10% greater than doses in Figure 17)
               TABLE 4 COMPARISON OF INSTRUMENTAL
                               BOD AND BOD

          BOD,-                              Instrumental  BOD*
          mg/1                                   mg/1	

            190                                    197
            536                                    482
            588                                    565
            815                                    839

   *  Standard propylene glycol of 400 mg/1 BOD,- used for the  instrument
     calibration .

-------
  172
                                            Figure  1

                                    ACTIVATED SLUDGE MINI PLANT.,

                                          Flow Diagram
Influent
    Equalization
    Acid
                           Nutrients
                   i—«	1
Feed
Pump
                                 Aeration Basin
                                   Sludge Recycle
                                                          Bio-Settler   F1 peculator.   Final Settler
                                                                  Alum
                                                               Caustic
                                                                        Waste  Sludge
                                      Figure  2.

                            ACTIVATED  SLUDGE  PILOT PLANT
                                       Side View
                   Acid
                   Tank
                            PHV-1
                              PHT-1
                         PHE-1
                                        Instrument
                                       Panel
                                       ooo
                                      AIC-2
                                       000
                                                  Nutrients
                                                    Tank
                                                            AT-2
                              FV-1
                     P-1
                                                  P-2

                                                <3£>
                                                i      f 1 i

-------
                               Figure  3.

                     INFLUENT FLOW AND  pH  CONTROLS
                                                                            173
    Acid
    Addi ti on
To          -*
Equalization
From
Equalizati on
                                                                  To
                                                                  Aeration
                                                                  Basin
                                        Ft-1
                              Figure  4.

                     ACTIVATED SLUDGE  PILOT  PLANT
                   Nutrients Control and F/M Control
                                                       F/M  Signal
            Total Carbon
              Analyzer
                                  Dilution &
                                  Homogen izer
                                                                   /
                                                                   *•

                                                                   i o-Settler
          P-l
        Feed  Pump
                                                                 Sludge  Recycle:
                                            SIudge

-------
174
                                   Figure  5.


                            F/M  CONTROL  SAMPLING  SYSTEM
Solenold-C
1
Feed
Sample
TCA . 	 £
Slide [J 	 i 4
Valve \
3 '
Homogeni zer
— t n
T Mixed
(c9* 	 Liquor
' 	 -^ ~r c=* Sample
^ — ) Solenoid-B
^ L__ Dilution
1 (c-% 	 Water
Drain i=a
> 1
/ O-V Solenoid-A
Drain ^
W Rotameter
n


                  Filter
       -a

       ja
       O-
       E
                                  Figure  6.


           QJ      AMMONIA CONSUMPTION BY MICROORGANISMS'
           0,6
           0,5
           0,4
           0,3
       I   0,2
       •<
           0,1
           0,0
                                          Slope = 0.133
             ' 0,0    1,0      2,0      3,0      4,0     5,0

                  [•Hero organ ism Growth,  Ib  vss/day

-------
                               Figure  7

                   DETERMINATION OF AMMONIA  REQUIREMENT
                                                                           175
           50
                                               o o
         E
         o.
         ex
           40
         o
         E
           30
           20
           10
                      10         .20       .30        .40

                        Nitrogen to Carbon Ratio in Feed
                                                 .50
                          .60
Feed
Total
Carbon ,
mg/1

Mixed
Liquor
Total
Carbon ,
mg/1

w w w
500
400
900
700
500
                                Figure  3.
          UNSTEADY STATE VARIATION OF THE SLUDGE COMPACTION RATIO
                              (DATA FROM RUN C)

             600 r
SI udge
Compact!on
Ratio,
3,Or

2,0

1,0

0,0
0
                          10
                       20
30
50
                                     Time ,  hours

-------
              Figure  9 .
   F/M BLANK TEST - BACTERIA GROWTH
           AND F/M RESPONSE
     (Constant Recycle Flow) Run A
                                                                                   Figure 1C.

                                                               F/M CONTROL TEST - BACTERIA GROWTH AND  F/M  RESPONSE
                                                                                      RUN C
                 Feed Flow            1938 ml/min
                 Sludge Recycle Flow  1070 ml/min
                 Sludge Waste Flow      85 ml/min    Feed
                                                     Total
                                                     Carbon,  5QQ
                                                                                  Feed  Flow           1938  ml/min
                                                                                  Initial  Recycle  Ratio  0.55
   600

^  50°

E-  400
o

   300

   800

^  700

*•  600

   500

   0,8

1  0.7
en

"  0,6
                         Feed
Mixed L i_q u o r^    °o
  oo      	
             00
                            O  Experimental
                            	 Computed
                  00
                    00
_  o
                                                     mg/1
                           Mixed    90°
                           Liquor   _nn
                           Total    700
                           Carbon,
                           mg/1     500
                                                                                         . O.°ooo°
                                                                                                          00
                                                                  000
                                                                                        O Experimental
                                                                                        — Computed
                                                    F/M
                                                     Signal'   0,8 —
     \     \
                                                    \	\
                                                        20       30

                                                        Time , hours
         10   15   20   25   30   35   40   45   50

                    Time, hours

-------
                         Figure  11.

        F/M CONTROL TEST -  RECYCLE AND WASTE RESPONSE
                            RUN C
                    Feed
                    Total
                    Carbon
                    mg/1
 F/M
 Si gnal
        1200 r
Recycl e
Waste
FT ow,
m 1 / mi n
                                     000
                                        Po°
 O  Experimental

—— Computed
                    Mixed
                    Liquor
                    Total
                    Carbon,
                    mg/1
                                                            F/M
                             20      30
                            Time, hours
                                                                                   Figure 12.

                                                                        RESPONSE CHARACTERISTICS WITH AND
                                                                        WITHOUT F/M CONTROL - LOW LOADING
                    Recycle
                    Fl ow,
                    ml/mi n
 600

 500

 400



1000

 900

 800

 700

 600

 0,9

 0,8

 0,7

 0,6

 0,5

1200

1100

1000
                                                                                        	Control Simulation
                                                                                        	Blank Simulation
  Feed Flow              1938 ml/min
_Init1al  Recycle Ratio  0.55
  Initial  Loading        1.5 gm TCp/gm TCM«da;
                                                                                     I
                                                     I
                                 I
                                    I
                                                                                   10       20       30      40
                                                                                 Time Since  Step  Change,  hours
                                                                            50

-------
            Figure 13.

 RESPONSE CHARACTERISTICS WITH AND
WITHOUT F/M CONTROL - HIGH LOADING
                     Figure 14.

COMPARISON OF AERATION BASIN SUBSTRATE CONCENTRATION
               AND WITHOUT F/M CONTROL
     CO
WITH
1
reed
Total
Carbon ,
rng/1

Mixed
Li quo r
Total
Carbon ,
m n t 1
mg/ I







F/M





Recycl e
Fl ow ,
ml/min


600


500
400

1000
900
800
\J W \J
700
600
—


— r. j. i r • 1 i •
^~~~~ Control oimulation
	 Blank Simulation i n

0 9
Feed Flow 2325 ml/min F/M
r~ Initial Recycle Ratio 0.3 Signal no
Initial Loading 2 . 1 gm TCF/gm TCM-day U,«
^^ 0,7
^^^ w | /
Digital Computer Simulation
Feed Flow 2325 ml/min
Initial Recycle
Ratio 0.3
X. """"'—- -^, ___ Blank
^^^"--..^ "~
.S ___ — — — """ F/M Control
- xC— " ""
=-^
/u
1,0

0,9

0,8

0,7
/"\ r"
60
Substrate
— ^ Cone. , _
V Total 50
	 \ -*^_ Carbon,
N. — — 	 	 	 mg/l 40
_ \^_ 	 :=i. — __
1 ^^^^^^ 30
°'6 1 20
900 r- 10 32,5


800
700


J.U
—
	 | ^^^^ °

~~ X"-**.
^•v ^^
J ^W x,^
1 ^^. "^"^
— 1 ^k ^"^.^^Blank
/ ^X. ~~"~~~^ 	
- 1 X^^^
I ^*^*«^
~~^ F/M Control
—

—

1 1 1 1 |
0 10 20 30 40 50
1 1 I | 1 Time Since Step Change, hours
0 10 20 30 40 50
          Time  Since  Step  Change,  hours

-------
                                                                                 179
Feed
Total
Carbon,
mg/1
Mixed
Liquor
Total
Carbon,
rog/1
F/M
                       Figure 15.

             RESPONSE CHARACTERISTICS WITH
                AND WITHOUT F/M CONTROL
 800

 700

 600

 500

WOO

1200

1000

 800

 0,9

 0,8

 0.7

 0.6

 0,5
        1200
Recycle
How,   1100
m 1 / m i n
        1000
                          Salt-Propylene Glycol Systen
	  Control Simulation
	  Blank Simulation
                             Feed Flow  1938 ml/min
                 Initial Recycle Ratio  0.55
                        Initial Loading  1.5  gm TCp

                                        gm
                            33
                       20      40      60      80     100
                    Time Since Step  Change, hours
                         Figure  16.

                    FLOCCULATION CONTROLS
                Al urn
         Bio-Settler     '
         (50 gallons)
                                       Caustic
                                           Floe-Settler
                                           (50 gallons)
                                                                      Final
                                                                      Effluent
                                                                            >
                               Chemi cal
                               Flocculator
                              (10 gal Ions)
                                                           'Waste
                                                           SIudge
                      7)Feed  Forward Turbidity Measurement

                         Feed  Back Turbidity Measurement

-------
 180
               Figure  17.

         OPTIMUM ALUM DOSE AS  A
      FUNCTION OF INITIAL TURBIDITY
             0,4r-
      OJ t.
      CO OJ
      O +->
     O T-

      E \
      13 t^
     r— 01
     
L.
J_
O)
Q-
90


80

70

60
o
Z  50
(0
c
m  Li(]
o  "u
c
o

S  30
CD
X
o
                  10
                           20          30          40
                                  Time,  minutes
50
60

-------
                                   Figure 19.

                          BIOLOGICAL INHIBITOR  DETECTOR
                SCHEMATIC OF MEASUREMENT CYCLE  WITH  TOXIN  IN FEED
                                                                               181
°  90 r-
   80  -
„  70  -
c
01
£  60  n
I  50
C
a)  i|Q
o  '^
c
o
u

£  50
en
X
                 10
20          30
      Time, minutes
                                     Figure  20.

                      CONTROLLED ACTIVATED  SLUDGE  TREATMENT
                Control  Unit
Feed
                                                             Control  Unit
                                                                          Final
                                                                          Effluent
                                                          Aerated
                                                          Stabi1i zati on
                                                          Tank
                            Recovery
                               or
                            Disposal
                    SI udge
                   Handli ng

-------
  THE NATIONAL PETROLEUM REFINERY WASTEWATER CHARACTERIZATION STUDY

                 Leon H. Myers, FredM. Pfeffer, and Marvin  L. Wood
                        U.S.  Environmental  Protection Agency
                   Robert S. Kerr Environmental  Research Laboratory
                                Ada, Oklahoma 74820

 ABSTRACT

        The EPA and API jointly conducted an  investigation to provide a verifiable data
 base  under actual  operating conditions in the treatment of petroleum  refinery wastewater.
 This data base consisted of raw waste loadings  and activated sludge treatment efficiencies.
 Industrial participation in the raw waste load phase accounted for 84 percent of the U.S.
 petroleum refining  capacity.

 INTRODUCTION

        In 1965, with the passage of amendments to the Federal  Water Pollution Control
 Act,  the API Committee on Air and Water Conservation instigated a project to  secure
 representative data on wastewater treatment and control practices  in domestic refineries
 of the petroleum industry (1). The response to  this project represented 93 percent of the
 industry's crude processing capacity.   From this response, a profile of wastewater effluent
 quality was  prepared reflecting the various types of operation in 1967.

        The  Refuse Act Permit Program and the  impending legislation which became the
 Federal Water Pollution Control Act Amendments of 1972 (2) necessitated that the Envir-
 onmental  Protection Agency acquire a scientific characterization of various wastewater
 streams and  a factual  determination of the efficiencies of available wastewater treatment
 processes  so as to ensure protection of the environment from any  adverse effects caused by
 wastewater discharged under a permit system.   In  January of  1971, a series of  contracts
 was let by EPA's Office of Water Programs to assemble available information and to utilize
 consultant expertise as a base from  which effluent limitations might be prepared.   At that
 time, the prevailing concept was to determine  "standard raw waste loads" coming from
 the various unit processes, apply the optimum  contaminant removal efficiencies afforded
 by available wastewater treatment processes, and then determine allowable effluent char-
 acteristics.   There were limitations,  however, in  the usefulness and adequacy  of the
 available and current technical data.  The API offered to cooperate with  EPA in con-
 ducting a survey of the petroleum refining industry to update the 1967 data.   Following
 Agency acceptance of the offer,  a planning committee was formed consisting of represent-
 atives from EPA's Office of Research  and Monitoring,  EPA's Water Quality and Refuse
 Act Permit Section, State water quality offices in  Texas and California,  the API, and the
 consulting field.

        The  committee agreed that the refining classification scheme used in the 1967 API
survey should be incorporated into the plan of a joint API/EPA study.  This classification
assumes that the extent of processing complexity of oil refining would have a parallel
impact on the character and quantity  of contaminants entering a given refinery wastewater

                                     183

-------
  184

system.  The five refinery classifications were and are:
        A.  Crude topping (Atmospheric and vacuum distillation)
        B.   Topping and catalytic cracking
        C.  Topping,  cracking, and petrochemicals
        D.  Integrated (Topping, cracking,  and lube oil processing)
        E.   Integrated plus petrochemicals
Applying this fundamental refinery classification scheme to the nation's refineries, Class A
represents about 5 percent of the domestic crude capacity, while Classes B, C,  D, and E
account for  36, 23, 26, and 10 percent,  respectively (3).

        The  survey objectives agreed to by the committee were:
        1.   To determine the refinery raw waste load;
        2.   To verify  analytical results from representative refineries;
        3.   To determine the contaminant removal efficiency of selected refinery waste-
        water treatment systems; and,
        4.   To evaluate and update interim  reference guides for the petroleum  refining
        industry.
The  results of the raw waste load survey have  been reported by the API (4).   This presenta-
tion  today concerns the second and third objectives:  analytical validation  and treatment
efficiency.   A draft  report on  the study is under consideration for publication (5).

ANALYTICAL VERIFICATION  SURVEY

        For  the purpose of validating data obtained in  the questionnaire portion of the raw
waste load survey, EPA conducted a study of  17 of the 150 refineries which submitted the
questionnaires.   Selection of refineries to be monitored by EPA was made by the joint
API/EPA Committee from  the list of participating refineries by use of a random number table.
Code numbers for plant identification were  used to insure confidentiality.   Validation
consisted of a statistical comparison of results obtained when refinery and EPA personnel
conducted thorough monitoring and sample-splitting  programs  at each of the 17 refinery
locations.    Responsibilities during the study  were shared.   The refineries' personnel
documented flows and operating conditions, collected, composited, split, and preserved
the samples  for independent laboratory analyses within 24  hours of sampling under observa-
tion  by EPA personnel.

        Refinery Selection.  Table 1 shows the number  of refineries by class and crude cap-
acity which were monitored by EPA.  Also included  are the questionnaires received from
each refinery class, showing that for purposes of validation, EPA  monitored 10 percent of
the respondees in each class.   Nominally,  a  10 percent random sampling of a total popu-
lation constitutes a statistically sufficient sample, and the results of many polls or surveys
are based  on much less data than such a sampling.   Thus,  the survey proposed by API in
actuality approximated a  study of the entire population, i.e., the petroleum refining
industry.  This conformity to strict scientific and statistical procedures continued during
survey sampling,  analysis, and interpretation  of the  data.

        Analytical Procedures .  Selection and implementation of  laboratory methods are
critical decisions for  any  survey.   Analytical methods must be reliable; that is, they should

-------
                                                                                  185


have widespread usage among water quality laboratories, compensate for interferences in
samples, and be precise and accurate.   In addition, the methods must conform to the
equipment, skills, and analytical time limitations of the laboratory and  the survey plan.
The procedure for each method must be clearly described,  including references and any
modifications necessary because of  interferences.

       The joint API/EPA task group selected the 22 parameters  for the  study as shown in
Table 2.  A complete wastewater characterization of any  industrial effluent would require
analyses of every known wastewater parameter.   The wastewater characteristics evaluated
during this study do not reflect all known parameters but do represent the major constituents
expected in this industry's wastewater.

        In addition to these 22 parameters, EPA's sample splits were subjected to further
analyses.  A comparison was  made of the hexane and freon gravimetric  procedures for
"oil and grease"; and additional parameters were measured: color, turbidity, volatile
suspended solids, specific conductance,  nitrite nitrogen, and acute fish  toxicity.   All
participating refinery laboratories were instructed to adhere to the EPA's approved methods
(6), as modified slightly in 1972; copies  of these methods were distributed accordingly.

       Analytical Quality Control.  A host of determinate errors inherent to any  survey of
this nature dictated the incorporation of  an analytical quality control program.   Beginning
with the planning phase, emphasis was placed on such laboratory variables as selection and
sensitivity of instruments, glassware cleaning, and reagents.  Control of performance was
accomplished by following the EPA's approved methods of analytical quality  control devel-
oped  at the Robert S. Kerr Environmental Research  Laboratory (RSKERL) in Ada,  Oklahoma.
A systematic program of accuracy and precision  testing was implemented.  Initially, a
sample of API separator effluent was examined by each of the methods in order to ascertain
problem  interferences as evidenced by comparison  of duplicate results for precision and
spike  recovery results for accuracy.

        Factors  such  as the number of samples received and time limitations for completion
of analyses precluded duplicating and spiking each sample.   The laboratory  quality control
program  involved running a blank, standard, sample duplicate, and spiked sample —for
every eight samples received.   Success in duplication and spike  recovery indicated that
the results obtained for the corresponding eight-sample set were correct with a high degree
of confidence.   Failure presented two alternatives: rerun the set of eight samples, or dis-
regard the results entirely.

       The EPA's quality control results  are summarized  in Table 3 for the more important
parameters.   In all but three  instances our Laboratory met the frequency requirement of
 ^ 1.0 per eight-sample set.    With  the exception of sulfide,  we achieved s 95  percent
accuracy (or spike recovery).

       Results and Interpretation.    In  anticipation of a monumental  task of interpreting
the field and analytical data generated during the studies, it was decided to utilize RSKERL's
Computer and Statistical  Section for data accumulation.    Verified analytical data were
keypunched and machine verified by Computer Section personnel.  Uniform  rules  for

-------
  186

rounding off decimal  points and not reporting zero values were established for the data.
Negative values were reported as negative values.   Toxicities encountered in BOD,- analy-
tical  data were treated uniformly by selection of the highest consistent values.  Statistical
techniques for  data evaluation, interlaboratory and  intralaboratory deviation, frequency
distribution of  parameters, quality control regression analysis, student "t" testing, and
plotting were paramount to the studies.

        Distribution analyses were performed on  the  raw waste loads per 1,000 barrels of
crude throughout for each  of the 22 parameters and each of the five refining classes.  Median
values were selected  as the best measure of central tendency on all parameters because of the
highly skewed  frequency distribution  (Figure 1) which is a common occurance  in  water quality
data.   The difference between  the average (62) and median (37) in  this figure is quite appar-
ent.

        Comparison of the  22 parametric results from  17 refineries with the EPA results was
made.  A summary of the  difference  in median values (mg/1) between EPA and API analyses
and the student "t" values is shown in Table 4.   The "t"  value relates to statistically sig-
nificant differences of analytical results at the 95 percent level.  When the calculated
absolute value  of  "t" at the 95 percent confidence level is ^2.0, the medians are signifi-
cantly different.  The parameters for which this occurred are  oil and grease,  phenolics,
sulfide, total chromium, ammonia nitrogen, cyanide, iron, copper, lead, and zinc.  Note
that BOD was a border line case.   This observation of value differences between the same
sample points is further justification for a strict quality control program.

ACTIVATED SLUDGE TREATMENT EFFICIENCY SURVEY

       The activated sludge treatment process had previously been recognized by the joint
API/EPA committee to be representative of the best  level of treatment currently used by
the refining industry.   Consequently, this treatment system concept was selected for the
efficiency studies.

       Unlike  the net raw waste load survey where  the sampling was done by refinery per-
sonnel, the activated sludge efficiency survey required all sampling to  be accomplished by
EPA staff,  with scheduled  analysis at  RSKERL  in Ada.  Field  sampling work was  planned to
extend over fourteen 24-hour periods  as compared to a one-day  period (24 hours) for the net
raw waste load survey; therefore,  the  EPA manpower impact would be somewhat extensive
from both the field and the laboratory standpoint.    The efficiency study was managed and
coordinated by EPA's RSKERL, with aid in actual field sampling provided by EPA Region VI
representatives from Surveillance and Analysis and Enforcement  staffs.

       Refinery Selection. In the selection of the participating refineries, significant
consideration was  given to the choosing of those refineries with  treatment systems of ade-
quate design.    Through discussions with the respective refinery  personnel concerning their
operating  and control procedures,  a selection of treatment systems was made to meet this
criteria.    In order to study the maximum  number of  refineries with the EPA personnel avail-
able and still provide data in a quantifiable manner, five activated  sludge systems were
chosen from a list  of 14 existing facilities provided by the API.

-------
                                                                                    187

       Refinery location selection was made so that distribution would approximate the
geographic distribution of the refining industry as shown in Figure 2.   Hence, two were
located in EPA Region VI  (southwest); two  in Region V  (Great Lakes); and one in Region X
(west coast).   An  important aspect of a national study  of this nature was to receive the
samples at the laboratory as rapidly as possible; therefore, consideration was given to the
relation between refinery  location and airline schedules.  Since not all classes of refineries
use activated sludge as their biological treatment system, it was decided to study three Class
B and two Class C  systems.   The initial plans included  a specific Class D refinery; however,
limited time and funds forced the EPA to eliminate  this  plant from the five-refinery  list.
The parent refining company voluntarily submitted  10-day survey data collected and analyzed
in a manner identical to the five-plant  investigation for use as an addendum to the final
report.

       Efficiency  values were calculated from data obtained at two points in the treatment
process: primary settling effluent and final  clarifier effluent.   No  accounting was made for
once through  cooling, intake, and bypass waters, for domestic wastes, nor for the finishing
systems following   final clarification of the activated  sludge treatment process.  This was
because the study objective was activated  sludge treatment data; not discharge data.

       The systems sampled for activated sludge efficiency employed separation or treatment
ranging from API separators through equalization and intermediate treatment steps to final
detention ponds.  Schematic flow diagrams of the treatment systems are shown in Figure 3.
In addition to  primary separation, the systems included  various combinations of the following
individual processes: activated sludge,  trickling filters, air flotation, chemical coagulation,
aeration  basins, and clarification.   Further, most plant systems included sludge ponds,
detention ponds, and lagoons.   This diversity of treatment allowed the evaluation of treat-
ment steps other than the activated sludge  process.

       Daily samples were composited on a two-hour basis; sample preservation, fixing,
analyses, etc.,  were conducted as in the RWL procedures.  Sample splitting between the
EPA and  the refiner was done on an irregular basis, at the discretion of the refiner.

Results   The data base for the efficiency survey was obtained by analyzing 70 total  samples
from the five refineries.   A  summary of the range of influent and effluent concentrations
and the median treatment efficiencies is shown in Table 5.

       Median effluent loads from the  five systems studied and the standard deviation of
loading expressed as percent of the median are shown in Table 6.   Although the refinery
daily crude throughput and gallons of water discharged  remained nearly constant it was
observed that the pollutant load varied with daily samples,  reflecting apparent operational
differences within  each  refinery from one day to the next.

       A summary of the median final clarifier loads for 13 parameters is found in  Table 7.

CONCLUSION

       This study represents  the first industry-wide investigation of its type attempted by

-------
 188

the EPA or its predecessor agencies.   It resulted in a valid data base of raw waste loading
and contaminant removal efficiencies, and it was planned and conducted with the cooperation
of the petroleum refining industry.   There were problems encountered:  interlaboratory
analytical variability and interferences in analytical methods, particularly cyanide, sulfide,
and oil and grease.

       The survey supported our opinion  that several needs deserve further consideration.
Variability and interferences in methods require assessment and resolution through extensive
round-robin testing.   Also needed are the identification of major organics treatable by and
refractory to  the activated sludge process; the measurement of the fate and effects of these
refractories; the study of metals adsorption on sludges; and additional characterization,
especially of unit process streams such as catalytic cracking and sour waters.

       We in EPA at the RSKERL would like to recognize the technical cooperation of the
refining industry in our research  endeavors and  the many courtesies extended to us.   We
look forward  to a continued coordinated approach to solving the many problems remaining
in the management of refinery wastewaters.

REFERENCES

(1) "1967 Domestic Refinery Effluent Profile," Committee for Air and Water Conservation,
    API,  1968.
(2) "Federal Water Pollution Control Act of 1972,"  Public Law  92-500, 92nd Congress
    S-2770, October  18,  1972.
(3) "Annual  Refining Report,"  The Oil & Gas Journal 70:13,  136, March 27, 1972.
(4) "Petroleum Industry Raw Waste Load  Survey," Committee on Environmental Affairs, API,
    December,  1972.
(5) "National Petroleum Refining Wastewater Characterization  Studies," U.S. EPA Project
    No.  12050  CNI (Report in preparation).
(6) "Methods for Chemical Analysis of Water and Wastes,  1971," U.S.  EPA Report 16020"-
    07/71.
(7) "Laboratory Quality  Control Manual," 2nd Ed., 1972, U.S.  EPA, Ada,  Oklahoma.

-------
                                                                              189
DISCUSSION

J. J. Chavez:   I understand that cationic polyelectrolytes are widely used flotation
schemes in  the United States.   To your knowledge, have there been studies or surveys
of the toxic effects of using polyelectrolytes?

Fred M. Pfeffer:  I am not aware of any such studies.

Robert Farnham:  I assume that the data you collected would allow you to calculate vari-
abilities in the activated sludge process.

Fred M. Pfeffer:  Our data at each of the five plants  which were studied for treatment
efficiency were collected for  14 continuous days and  included measurements of flow and
27 parameters.   The raw data depicted variations  in flow and concentrations as  it was
collected.   Mathematically,  we have expressed the effluent variability (Table  6) in  terms
of the median  loadings from the final clarifier and the standard deviation of these  loadings
expressed as a percentage of the median.

Robert Farnham:  Did you take one sample  from each plant,  or more than one?

Fred M. Pfeffer:  Each sample was a two-hour composite.  We collected 14 samples from
four of the  plants studies,  and 13 samples  from the remaining plant.

BIOGRAPHIES
Leon H. Myers holds a BS in chemistry/biology from Southwestern
Oklahoma State University and a MS  in sanitary science from
Oklahoma University.   He is  currently Chief,  Industrial Section
of the Source Management Branch at the Robert S.  Kerr Environ-
mental Research Laboratory, Ada, Oklahoma.
Fred M. Pfeffer holds the BA and MS degrees in chemistry
from the University of Cincinnati.  He is currently a Research
Chemist at the  EPA's Robert S.  Kerr Environmental  Research
Laboratory at Ada, Oklahoma.
                                                                        •  |


-------
190
  Marvin L. Wood holds a BS in chemical engineering from the
  University of Arkansas and an MS in  instrumental science from
  the Graduate Institute of Technology of the University of
  Arkansas at Little Rock.   He is currently Deputy Director
  and Chief of the Source Management Branch of the EPA's
  Robert S. Kerr Environmental Research  Laboratory at Ada,
  Oklahoma.

-------
                                                                 191
      TABLE 1  "REFINERIES MONITORED BY EPA"
                                         REFINERIES MONITORED
REFINERY
CLASS
A
B
C
D
E
QUESTIONNAIRES
RECEIVED
14
79
31
12
14
NO.
2
8
3
2
2
CRUDE CAPACITY
BPSD
13,200
424,400
388,500
397,300
702,000
                 150
        17
1,925,400
        TABLE 2  "ANALYTICAL PARAMETERS"
BOD
COD
TOC
PHENOL
SUSPENDED SOLIDS
DISSOLVED SOLIDS
PH
ALKALINITY
ACIDITY
OIL & GREASE
T. PHOSPHORUS
AMMONIA N
KJELDAHL N
NITRATE N
CHLORIDE
CYANIDE
SULFIDE
T. CHROMIUM
IRON
COPPER
LEAD
ZINC
        TABLE 3  "QUALITY CONTROL RESULTS"

             FREQUENCY PER 8 SAMPLES

BOD,-
COD3
OIL & GREASE
PHENOL
TSS
SULFIDE
AMMONIA N
KJELDAHL N
CYANIDE
PH
BLANK
1.8
2.9
0.7
1.9
-
0.8
3.3
1.6
1.7
-
STANDARD
0.7
1.7
-
3.0
-
2.3
9.8
1.9
5.2
-
DUPLICATE
4.9
2.4
-
2.8
3.0
2.2
1.6
1.9
1.7
0.7
SPIKE
1.8
-
1.9

3.3
1.2
1.7
2.7

                        AVG.  SPIKE
                        RECOVERY %
                                                              98

                                                             98

                                                             85
                                                             98
                                                             100
                                                             98

-------
192
                            TABLE 4  "COMPARISON OF EPA & API DATA1
BOD
COD
OIL & GREASE
PHENOL
TSS
SULFIDE
AMMONIA N
CYANIDE
T.  CHROMIUM
IRON
COPPER
LEAD
ZINC
pH (Units)
FLOW (GPM)
MEDIAN DIFFERENCE
  (EPA MINUS API)
       1.0
       0.5
       8.4
       0.02
     -10.0
       0.5
       0.2
       0.02
       0.03
       0.28
       0.02
       0.04
       0.04
       0.0
       0.0
"t" STATISTIC
    1.9
    0.4
    3.9
    4.0
    0.4
    5.6
    3.8
    3.8
    4.8
    5.1
    4.7
    7.4
    3.4
    0.4
    1.3
                                                                                    SIGNIFICANT
                                                                                    DIFFERENCE
YES
YES

YES
YES
YES
YES
YES
YES
YES
YES
                               TABLES "TREATMENT EFFICIENCY1
INFLUENT (m

BODC
S
COD
OIL & GREASE
PHENOL
TSS
SULFIDE
pH (units)
AMMONIA N
KJELDAHL N
CYANIDE
T. CHROMIUM
IRON
COPPER
LEAD
ZINC
MAX.
288.0

583.0
89.0
16.0
261.0
43.0
11.5
160.0
194.0
2.5
4.25
44.0
0.88
1.73
1.23
MIN.
27.0

93.0
8.0
0.26
15.0
0.10
6.2
6.0
8.0
0.02
0.03
0.50
0.02
0.00
0.07
a/i)
MEDIAN
85.0

213.0
29.0
3.4
36.0
2.9
8.8
12.0
17.0
0.2
1.43
1.02
0.14
0.10
0.22
EFFLUENT (mg/1)
MAX.
53.0

201.0
37.0
4.2
83.0
1.7
8.7
124.0
134.0
0.03
1.45
0.50
0.11
1.04
1.84
MIN.
5.0

46.0
4.0
0.01
5.0
0.0
5.0
0.05
3.0
0.02
0.02
0.18
0.02
0.10
0.04
MEDIAN
8.0

70oO
11.0
0.01
25.0
0.3
7.3
11.0
15.0
0.1
0.26
0.78
0.05
0.11
0.16
MEDIAN
REMOVAL (%)
89.2

54.6
60.0
99.6
23.8
90.0
—
16.2
17.2
30.9
58.2
37.1
57.7
0.0
22.0

-------
                                                                                      193
                  TABLE 6  "STANDARD DEVIATIONS OF EFFLUENT LOADINGS'
BOD,
COD°
OIL & GREASE
PHENOL
TSS
SULFIDE
AMMONIA N
KJELDAHL N
CYANIDE
T. CHROME
IRON
COPPER
LEAD
ZINC
MEDIAN LOADINGS
   (lb/1000 bbl)
        O
       23.0
        2.9
        0.006
        7.9
        0.10
        5.7
        7.6
        0.04
        0.083
        0.330
        0.015
        0.046
        0.059
STANDARD DEVIATION
   OF LOADINGS
   (% of median)

       55
       21
       41
        1100
       44
       97
       31
       25
       24
       29
       29
       26
       20
       55
                    TABLE 7 "EFFLUENT LOADING FROM FINAL CLARIFIER"
BOD,
COD
OIL & GREASE
PHENOL
TSS
SULFIDE
AMMONIA N

CYANIDE
T. CHROME
IRON
COPPER
LEAD
ZINC
                                               MEDIAN LOADING    (lb/1000 bbl)
9973
5.5
42.0
9.2
0.006
17.0
0.10
4.1
0.06
0.025
0.480
0.032
0.397
0.086
2115
3.1
22.0
2.8
0.006
4.0
0.05
17.0
0.01
0.078
0.148
0.005
0.018
0.027
0288
3.9
49.0
7.1
0.006
14.0
0.32
7.5
_
0.136
0.330
0.015
0.064
0.036
6512
4.0
16.0
2.9
0.003
7.9
0.13
0.1
-
0.085
0.083
0.014
0.030
0.059
6693
1.5
23.0
2.4
0.002
6.4
0.04
5.8
-
0.210
0.650
0.030
0.046
0.089

-------
194
    50
    40
  UJ

  O 30
  ui
  cc
  u.

  UJ

  p 20
  <
  _i
  UJ
  tr
     10
     0
fO
I

1
Q
UJ
    -60 -30  0  30  60  90 120  150  180  210  240 270 300 330  360 390 420

                        BOD5  RAW WASTE  LOADING



  Fig.  I.    RELATIVE  FREQUENCY  DISTRIBUTION OF BOD5 RAW

             WASTE LOADING  FOR CLASS  B  REFINERY
        10
                    8
                    .•  ••
    F.g. 2.   DISTRIBUTION  OF THE PETROLEUM REFINING INDUSTRY

-------
REFINERY NO. 9973

    Once Through Cooling Water-
    Process Water
      2400 gpm
REFINERY NO. 2115
•>• API Separator
      4hr.
  •*• Equal ization •
       7.5hr.
 •Aeration Basin-
      I2hr.
                                                                         -*-Final Clarifier
                                                                                2.9 hr.
                                    (S)
                                              •1200 gpm
                                                                                  I
                                                      (S)
    Once Through Cooling Water
    Oily Water
     3000 gpm


REFINERY NO. 0288

    Clean Water

    Oily Water
     6600 gpm


REFINERY NO. 6512
    Separator
    .57 hr.
-^-Primary Clarifier
      .65hr.
                                  (S)
 Trickling Rlter—r^- Aeration Basin—»-Final Clarifier
   .23hr.     I     2.7hr.            3.5hr
            (S)      ^—900 gpm	'
                                                                               (S)
il 	 	
^ A? i ocpuruTor
i 2. 5 hr.

0.5hr.
fc
"• MCI UIIUM LJUSIII •
I3,hr- (s)
^ C 	 PRnnnnm.

*• Final Clarifier • ^
2. 5 hr.
     All Waste Water
       2270 gpm
 REFINERY NO. 6693
 API Separator
   Equalization
     48 hr.
Chemical Coagulation
       6hr.
                                                  (S)
                                                                            •»- Activated Sludge-
                                                                                    12 hr.
                                                    (S)
     All Waste Water
      620gpm
          Separator
          2.5 hr.
                                         (S)
         *- Aeration Basin
              48 hr
                                                  100 gpm-
         •*• Final Clarifier
               2hr.
         	   I
                                                           (S)
                                                            (S)
                                                                      ( S) Denotes Sample
                                                                              Point
        Fig. 3.    REFINERY WASTE WATER  SYSTEMS CHOSEN  FOR  THE  ACTIVATED
                                        SLUDGE  EFFICIENCY STUDY
                                                                                     •o
                                                                                     Ul

-------
196

-------
          SESSION IV







 "SLUDGE MANAGEMENT"






Chairman





Milton Beychok




Consulting Engineer, Irvine, California







Speakers
Carl E. Adams,  Jr.




"Sludge Handling Methodology For Refinery Sludges"







Jacoby A. Scher




"Processing of Waste Oily Sludges"







C.B. Kincannon




"Oily Waste Disposal by Soil Cultivation"







R.L. Huddleston




"The Disposal of Oily Wastes by Land Farming"
                 197

-------
198

-------
        "SLUDGE  HANDLING METHODOLOGY FOR  REFINERY SLUDGES"

                          Carl  E. Adams, Jr., President

                                       and

                                 Robert  M.  Stein,
                         Director of Operational Services

               Associated Water and Air Resources Engineers,  Inc.
                              Nashville, Tennessee

SOURCES OF REFINERY SLUDGES

      There are  several  major sources of  sludges generated from normal refinery
operations.   These sludges  can  be classified  as storage tank bottoms,  bottoms from
API separators,  crude  desalting sludges,  catalytic  solids,  spent  clays, coking fines,
and solids  from  the  utilities and the biological wastewater treatment system.  The
basic  source  of  many  of the  solids  is the crude oils which contain materials that
originate either  in the well or  result during  transportation  and storage of the crude
oil.   The solids generally associated with the  crude oil  include iron rust,  rust
sulfides, clay, sand,  salt crystals,  wax,  and paraffin.   These  solids will settle  out
either in the  tank bottoms  or gravity API separators.   A delineation of types and
characteristics of  refinery solids  are given in Tables 1  and 2.   Estimates  of quan-
tities  of some of these sludges are  given  in  Table  3.   The waste  sludge from API
separators plus usual secondary  treatment  facilities  will  contain  approximately 1,500
to 2,000 Ib.  of dry solids  per million  gallons  of effluent treated.

      Tank Bottoms.   A summary of the  tank  bottom sludge characteristics  is pre-
sented in  Table  4.  The tank bottoms  can be  divided into two  basic types  of solids:
leaded and non-leaded.  The leaded solids are generated from  those processes which
have  been  in contact  with  tetra-ethyl  lead,  e.g., gasoline and related products or
the slop oil.  These solids  must  be handled  carefully because of their contact with
the tetra-ethyl lead (TEL).   TEL  sludge from storage tanks will  contain about 0.12
percent by weight TEL or 120 mg/1.   A  procedure recommended for handling TEL
sludges  is as  follow  (9):

                 Manually remove  the  sludge  and  transport it  to  a  drying
             bed area where it should be spread in a  layer of  3-in. thick-
             ness  or less.  The sludge layer should be "weatered" for at
             least 4 weeks at ambient temperatures of above  32°F (if the
             temperature falls below 32°F for  part of the time, the drying
             period  should  be correspondingly  increased).  Protective cloth-
             ing should  be used during the transporting and spreading of the
             sludge  but  face masks should not be  necessary.  At  the end of
             4 weeks, the  sludge should  have  weathered  down  to a TEL  con-
             tent  of 20  ppm  wt.  which Ball  states to be  a  non-hazardous  level.
                                     199

-------
200

      The non-leaded solids originate from those processes which have  not  been in
contact with tetra ethyl  lead.   These sludges result from the crude  oil storage, lube
oil, specialty oils and oil  intermediates,  gas  oils, furnace oils,  and injection gas oil
or from asphalt.   These solids  range  in  concentration  from  about 10 to 30  percent
solids by  weight, averaging about  20 percent.  The principal method  of  removing this
sludge from the tanks is  to  add water to transfer by pump or vacuum truck.   There  are
some  problems  experienced in the removal  of  the  residual waxy crude tank  bottoms
because this material  is extremely  viscous and difficult to handle.   It  has  been found
that the  heat addition minimizes pumping  transfer problems.

      Crude  Desalting Sludges.  The  crude desalting process is  another source of solids.
Crude desalting is employed  to  separate the salts  from the  oil by the  use of an
emulsifier and  settling tank.  Sludges produced from  this process include  free and
emulsified oils and  suspended solids.  These solids  are usually discharged directly  to
the treatment system in  the main waste  flow.

      Catalytic Solids-   The catalytic processes may or may not provide  a mojor  sludge
problem^   The noble  metal  catalysts  used  in  reforming and hydroprocessing are usually
extremely valuable  and  should  be completely  recovered and reprocessed.   The  non-
noble catalysts can  be recovered and sold  to  a chemical processor for  metal  salts
recovery.   The remainder of the catalysts are  not  usually .reprocessed, since they are
inert, but they can be disposed of in a landfill operation.   Many times  the  catalytic
fines  which are discharged from fluid beds catalytic crackers can be collected in  the
electrostatic precipitator system.  These fines might have value  as a sludge conditioner.

      Spent Clays.     Clay or acid  treatment  can  be  used to remove  color framing
and other undesirable materials  from  solvent refining  and dewaxed  lube oil stock.
The acid  treatment  produces  high  concentrations of suspended solids, sludges  and stable
oil  emulsions.   The clay treatment results in a sludge which contains  clay, fuel oil,
and emulsified oils.

      Coking  Fines.     The coking operation  produces a sludge  which  consists  of
finely divided  coke particles and  waxy  materials.  These sludges can  be  land-filled
or may possible have  some reclamation value.

      Utilities  and Wastewater  Treatment Sludges*     The other major sources of
sludge are from the utilities  and wastewater treatment system.  The utilities  solids are
usually  sludges generated  from  the addition  of lime or alum to treat raw  water for use
in the plant.  Since  these solids are  inert and  have  no heat content,  they should be
dewatered  and landfilled.   Sludges generated by the wastewater  treatment  fcailities
include oily sludges from  API separators and  air flotation systems and  biological
sludges  from trickling filter and activated sludge processes.   These solids can be
dewatered  and incinerated  with other combustible  solids, or disposed of in combina-
tion with  the utilities and  other inorganic  solids.

      The materials  from  a refinery which are  readily combustible (See Table 1) in-
clude the waxy bottoms, oil sludges, coke fines,  and waxy tailings.   The non-com-

-------
                                                                                  201
bustible sludges are sand, rust,  silt, TEL-sludge, salt,  spent catalyst, and  lime
sludge.  The  excess biological sludge, although not readily combustible, can be
dewatered  to  a combustible  state.

      Both  in-plant recovery and reclamation of amenable solids combined with final
sludge handling and ultimate disposal  are required  in order to implement a  successful
sludge management program  in  a refinery.   In-plant management  consists of  tight
operational control, preventive  maintenance for leak control  and the location of
separators  at critical points  to  capture materials before they become  contaminated  and,
thus uneconomical  to  recover.   Sludge handling methods  consist of gravity or air
flotation thickening, dewatering by vacuum, pressure or centrifugation,  and final
disposal by incineration, landfill,  landfarming,  and barging to sea.

IN-PLANT CONTROL,  RECOVERY, AND  RECLAMATION

      An important element  in  a refinery sludge handling program  should be a good
in-house control program.   This program  should  include the  reduction of oil  leakage
by  preventive  maintenance of pipe  lines  and equipment and the repairing of leaks as
soon as possible after detection.  The  refinery should  avoid discharges of oils by the
necessary  elimination  of the  water from process  units,  proper dispostion of solids  and
spent chemicals, careful handling of emulsions and awareness  by all personnel  that
waste  treatment begins at the process  unit.   It has been  reported that a refinery can
expect total  oil losses of 0.1 to 2.0  percent by volume of throughput;  however,
losses  in excess of 0.4  percent  indicate the need  for improved housekeeping.

      The  control  system should be designed to allow for good  housekeeping  and
should be  equipped with adequate  sampling  and monitoring  facilities to indicate
problems immediately.   The  system design should provide  facilities  for cleaning  up
after charging  new catalysts  so  that materials  do not have to  be flushed into the
sewers.  Adequate installation  of both oil-water separators  and coke-water  separators
at crucial  locations is important to a  good housekeeping  program.

      Proper attention should be given to the  prevention  of  the formation of oil
emulsions  or,  when these emulsions  do exist, the system should be  capable  of isolation
and  separate treatment of these materials.   API separators for removal of floating oils
should be  located as close to the source as possible.

      Acidification of caustic wastes with sulfuric acid breaks some emulsions for oil
recovery.  Acid wastes  may  be reused as a source  of  fuel or  to produce by-products,
such as oils,  tars,  asphalts,  resins, fatty acids and  chemicals.  Many refineries  sell
or trade these  spent acids and  acid sludges  to acid  manufacturers.   Sometimes the
spent acids are sold to  other industries, where  they may  be used directly or fortified
with acid  of full strength.   Hydrochloric acid alkalation  and  spent aluminum chloride
sludges are sometimes  burned in special incinerators.  Additionally, the  burning  of
oil  sludges and salt-bearing sludges in power station furnaces  has been  practiced.
However,   air pollution  considerations  must be considered  in these  cases.

      A well-planned  slop oil and  oil  solids program is also important to the solids
management program.   An emulsion-breaking system can be  used to recover  oils which

-------
202

can  be reused within the refinery-  The water from the emulsion  breaking steps can  be
sent to the wastewater treatment plant, and  the  oily solids  sent to the sludge recovery
system.  A summary of the  slop-oil emulsion composition is presented in  Table 5.

SLUDGE HANDLING AND DISPOSAL

      There are basically three  types  of sludges which  must be handled by a refinery
wastewater treatment and sludge handling system.  These are:  recoverable oils,  oily
sludges, and  excess biological sludges.  Figure 1  indicates  the sludges from a typical
refinery waste treatment  system.  Although many refineries  handle the oily sludges in
combinatior. ,
-------
                                                                                  203
tion is  highly  variable  from  day to day.   Sometimes  tank bottoms contain heavy
emulsions which require special  treatment,  and often the sludge and  skimmings from
air flotation systems can present special difficulties  in  handling.   In  many cases,  the
skimmings from air  flotation  are sent  to a skimmed oil  tank  from  which they  can be
taken to landfill  or pumped  to an oil recovery unit for subsequent  reuse  of the  oils in
the refinery.  Chemical treatment is  usually  required for oil recovery

      Thickening.    Gravity thickening  methods  are successfully utilized for oily
sludges.  The  performance of  gravity-type  thickeners varies  considerably  depending on
the nature of  the oily  sludges being  treated.  Solids loading rates  in the range of 5
to 30 Ib/sq  ft-day  are  reported.   The  range  of thickened solids concentrations is from
3  to 10 percent.   Normally,  3  to 7  percent solids can be obtained from a gravity
thickener.

      Studies which have been conducted on gravity  thickening of  oil sludges indicate
that there  is no significant correlation  of surface  solids loading and solids recovery
(10).   Surface  hydraulic loadings above 400  gal/sq ft-day appeared to cause signifi-
cantly  decreased  solids recovery; however,  there  is a wide scatter  of data.   The
effluent from the gravity thickener would be unsatisfactory for direct release  due to
the high oil,  solids and COD concentration.  Basically, solids and oil recovery are
low from gravity thickeners.

      Another  alternative for thickening of oily  sludges is by use of  dissolved air
flotation.   A  number of studies  have been  conducted on dissolved air flotation and
the results of  some  bench-scale  evaluations with oily sludges are  presented in Table 8.
Flotation thickening of some  types of oily  sludges  was  unsuccessful.  Tests with heavy
particles indicated  some dispersion after setting for 10  minutes.   Thus, removal  of
heavy  particles by  partial  gravity settling should  improve air flotation thickening.

      Facilities for  chemical  addition should  be  included in any flotation thickener
for use  when necessary.  The results  of tests indicated  that  dissolved air  flotation
would  be successful for thickening of froth  flotation sludges, but  flotation thickening
of other refinery  oily sludges  is not practical.

      Dewatering.     Dewatering alternatives for  oil  sludges include  centrifugation,
vacuum filtration, and  pressure filtration.
      Centrifugation
      There  have been  a number  of reports  on the use  of centrifugal  dewatering for
refinery sludges.  A schematic of a system  that  is being utilized  for  a number of  oily
sludges  is presented in  Figure 3.  This  consists of a two-stage system whereby  the  first
stage is used to dewater the  thickened  sludge and the  second stage is utilized to
separate oil  from the centrate.   In the first stage, a horizontal  solid bowl centrifudge
is  employed  and  in  the second stage  a  disc-type  cetrifuge  is utilized.   In  this process
the sludge is heated to 180  to 200°F prior to centrifugation.

      The results  of tests on an oily  sludge with a Sharpie's  P600 centrifuge  are pre-
sented   in Figure 4.   Note that the total  suspended solids recovery is relatively flat

-------
 204

as compared to the volatile suspended solids recovery.   The volatile  suspended solids
are indicative of waxes and  heavy tars.  Consequently,  most of the heavy organic
materials go into the oil  phase  at high feed rates.   Since it  is desired to separate
most of  the solids  from  the oil,  centrifuge selection should be based  on the relatively
flat  portions of the recovery curve to  maintain  a high solids recovery in  the  cake.
The  results  of  another test for  dewatering oily sludges on a solid  bowl  centrifuge is
presented in Figure 5.

      After using the solid bowl centrifuge  as a first-stage, a  second-stage disc-type
centrifuge should be utilized„   The function of a second-stage nozzle-type centrifuge
is to remove the fine solids from the  solids-stabilized emulsions and thus  separate the
oil phase from the solids.  The  results of a test using the disc-type centrifuge are
shown  in Figure  6 and  indicate  that successful  operation  was obtained for all feed
rates studied.

      The performance of a full-scale  centrifuge system treating an oily sludge is
presented in Table 9.   An internal  inspection of the  scroll centrifuge  following 9
months  of operation  revealed that the  hard  surface  coating on  the internal flights had
worn away  exposing  the stainless steel.   Layered deposits  0.5  in. thick of carbon
fines and grit  were found inside the scroll (8).

      The maximum period that  the  disc  centrifuge  has been on-stream without plug-
ging has been 25 days.   The downtime required for removal  of the disc stack  and
cleaning can  extend to three days.

      The results of  centrifuge testing  and experience can be  summarized  as follows:
      1 .  A vertical  solid bowl  centrifuge is  not recommended for  dewatering  most
          oily wastes.
      2.  A horizontal  solid bowl  centrifuge followed by a high-speed nozzle or disc
          centrifuge is  best suited  for dewatering  mixtures of contaminated  API
          bottoms,  sludge decant pit  material and truck  bottoms.
      3.  A super decanter centrifuge dewatering oily sludges  is  anticipated to recover
          75  to  90 percent of the solids  in the  cake  when charged with  heated  oily
          sludges.   The oily phase  will  consist of 1 to 5 percent solids and the cake
          will contain approximately  50 to 60  percent solids.
      4.  A nozzle  ejector centrifuge separates  95  to 98 percent of the feed oil  in
          the  oily phase  and 2  to 5 percent in  the ringdam nozzle and nozzle water.
          Thirty to 50 percent  of the  nozzle  ejector  feed solids will  be removed with
          the  oily phase  with  the  remainder being  in the  two  water phases.

      Vacuum  Filtration
      A  significant  number of refineries use vacuum  filtration for dewatering oily
sludges.   If  properly  implemented,  vacuum  filtration of oil sludges  renders the  solids
suitable  for direct landfill or subsequent incineration.   In  order to  accommodate oily
solids, a precoat vacuum filter  should be used and the incoming  solids should  be
heated to temperatures greater than 170 F.   Many variables affect  the performance  of
filtration processes; however, the relative filterability  of various  sludges and estimates

-------
                                                                                    205
of vacuum filter loading rates can  be made  using standard  Buchner funnel and  filter
leaf tests.

      The results of  some filter resistance  tests (7)  using oily sludges are presented in
Table 10.   These tests  indicate that  oily sludges quickly blind the filter cloth  when  a
precoat was  not used.  These  data  indicate  that heating  of oily  sludges to 165°F
decreased specific  resistance and  improved the characteristics  of  the  cake.  The effect
of chemical  addition on filterability  is also  indicated in  Table 10.   The addition of
lime and  ferric chloride did not improve the filterability of oil  sludges.

      A vacuum filtration filter leaf  test was  conducted on the oily sludges in  the same
study.  The  design  filter loading  rate for  precoat vacuum  filters  for various types of
oil sludges  is presented  in Table  11.  Using an operating  temperature of 160 to  180°F
improves the  vacuum filter  performance  in terms  of oil recovery, filtration loading,
and cake solids concentration.  Several noteworthy observations can  be made  con-
cerning the effect of clay addition on vacuum filtration performance.  First, the  oil
recovery decreases with increased clay addition.  Secondly,  the effective  filtration
rate decreases with  clay addition.   However, the effective filtration rate is  based on
solids, excluding the clay, therefore, the total solids filtration  rate  may  not be
decreasing as rapidly as it  might seem.   The  major conclusions regarding  vacuum  fil-
tration of oil sludges are:
      1.   Increased  feed temperature  greatly improves vacuum solids  performance.
      2.  Addition  of spent clay  decreases oil recovery and solids  filtration rate
          on a clay-free basis.
      3.  The measured filtration loadings of  0.8 to 3.0 Ib/sq ft-hr  were required
          for oil recovery.

      Pressure Filtration
      A fixed plate, high  pressure  filter press may be used to dewater many  types of
oily sludges.  Sludge  is applied to a cloth  covered filter plate  at  high pressure  and
water and oil  are passed through  the  cloth media while solids are  retained.  Para-
meters which are important in evaluating  the  process are feed temperature,  lime
addition,  spent clay addition, cycle  time, cake characteristics,  and filtrate  character-
istics.

      The results of  filter press tests  on  an oily sludge are presented in Table  12.
The feed  to  the pressure filter consisted of a  base  mixture  of  refinery oily sludges
which  contained API separator bottoms,  sludge decant pit material, and tank bottoms.
These tests  indicated that heating the feed was  required to obtain  satisfactory  filtra-
tion.  Effective breaking of solids-stablilized emulsions  is obtained.   Cake solids in
excess of 50  percent can be obtained on  pressure filtration with  oil  concentrations in
the range of  5  to 20 percent.

      The filter press cake  solids  concentration  correlated with average  filtrate temp-
erature as  shown in  Figure  7.   It is  apparent that  filtration temperature has a  dramatic
effect on  cake  solids and cake oil  concentrations at constant  cycle times.  The
addition  of lime improved  filtration  somewhat, although the effect  is  not  conclusive;

-------
 206


however, experience at other refineries indicates that  a definite increase in filtration
efficiency can be  obtained  by adding  lime.   A  significant decrease  in  cake oil content
can  be obtained by washing the cake  with  hot water.   The  filtrate is usually  readily
separable into an oil  and water phase.  Test results  indicate that the oil will  probably
require processing  through the  high chloride  slop system.   The water phase  has not been
found  to  create any  problems in the wastewater  system.

       Disposal.  The ultimate disposal  of  oily sludges can  be by barging to sea,  land-
fill, land farming, and incineration  with  landfill of  the ash.  However, sea disposal  of
oily materials  is viewed  as  a short-term alternative and will eventually be  eliminated
as an  option.   The disposal  of  the oily sludges  on soil  is acceptable  if it  can  be
shown  that  such  disposal will not contaminate groundwaters or contribute to contami-
nated  storm runoff, and will not create a potential seepage  problem.  A proper land
farming operation using soil  bacteria for degradation  of oils  would satisfy the above
requirements.  The utilization  of a lined  landfill with  leachate treatment would also
meet these  requirements.

       Land  farming of  oily  sludges  has been  successfully practiced by refineries where
sufficient land area is available for  proper  decomposition  of the  oil-containing  solids.
Land farming  involves  spreading the  sludge  in 4  to 6-in.  layers, allowing  the  sludge  to
dry  about one week,  adding nutrients, and  then  discing the  sludge  into the soil.  A
Texas  refinery land farms crude oil tank bottoms  containing high molecular weight, oil-
containing olefinic or  aromatic  components  and waxy raffinate containing high para-
finic components.  Decomposition  rates were found to  be  approximately 0.5  Ib/mo-cu
ft without nitrogen and phosphorus  addition  and  1 .0  Ib/mo-cu  ft with nutrient
addition.  Disposal costs were estimated at  $7 per barrel  of oil  including the  cost of
nutriel addition.   The species  of microorganisms  which were  found to remove the oils
were pseudomonas, flavobacterium, nocardia, corynebacterium, and artherobacterium.

       Incineration  of oily sludges and  landfilling of the ash  should provide an
acceptable  means of final disposal.  Important factors  in disposal by incineration
sould  include the following:
       1.  Selection and  cost of the  incineration  equipment.
      2.  Operating costs.
      3.  Potential air problems by particulates, sulfur dioxide and other pollutants.

      Table 13  presents the  results  of an economic evaluation of flow and  optimum
solids  concentration to incineration.   The fuel costs  associated with evaporation of
excess water  in an incineration system are presented  in Figure 8.  Oil  and volatile
matter in the contaminated  sludges can be destroyed  by incineration.  The type of
incinerator which is most applicable to a  particular waste  depends on the nature of
the waste.  Sludges and solids  of  high solids contents  are  conventionally handled by
multiple-hearth incinerators.

      A  fluid bed incinerator operates  with  a bed  of ash and sand particles fluidized
by forced air feed.  The  fluidized bed incinerator is best  suited for feeds  that are
partially  liquid so  that the  incinerator can  be  fed  by pumps and screw  conveyers.

-------
                                                                                    207


While  feeding  dry solids to a fluid bed incinerator is possible,  the multi-hearth
incinerator will be  more economical if  most of the feed is  in the  form  of  cake of non-
pumpable solids.
      Rabb indicates "suitable"  incinerators as follows:
          2-10wt.% solids feed  - fluid sand  bed
          5-70 wt.  % solids feed  - rotary  kilm
         40-70 wt. % solids feed - multiple-hearth

      Biological  Sludges.      Basically, the sludge handling methodology employed
for processing  biological sludges is similar to that previously described for  oily
sludges.  However, due  to the  putrescible  potential of  biological  sludge,  stabilization
is required prior to disposal by  sea or land.  The basic  process alternatives are
depicted in Figure 9 with  the most feasible  alternatives  having  been summarized in
Figure 1 .  Consequently,  the process sequency for biological sludge handling consists
of stabilization,  thickening, dewatering,  and final disposal.  Oftentimes,  the
thickening step will  precede stabilization in order to reduce the stabilization costs
which  are  highly dependent on  hydraulic  flow.   Detailed design procedures and
examples for these processes are given elsewhere  (1) and will not  be  described  herein.

      Biological  Sludge Stabilization.      Stabilization  of excess biological  solids is
required because  of the degradable nature of the microbial  sludge.  There  are three
basic methods  of  sludge stabilization which are technically  available:   aerobic
digestion,  anaerobic digestion,  and wet air oxidation.   Because of the  complexity of
anaerobic digestion and the problems experienced  in operating the process, this
method is not  being  utilized in new refinery sludge stabilization systems.  However,
there are several  older treatment  systems which have  used anaerobic digestion.   There
have been considerable operational and materials problems experienced  with  wet air
oxidation and, to date,  this method has not received acceptance in the refinery
industry.

      The most common method  of stabilizing water biological sludge  in  the  refining
industry is by  utilization of aerobic digestion.  Aerobic digestion  is used chiefly to
stabilize the sludge and  render  it  suitable for  land disposal.  Additionally, the
overall sludge  quantity may be  somewhat  reduced  and the dewatering characteristics
improved slightly.  Aerobic digestion is an extension of aerobic biological oxidation
in the absence of an  external food supply.  Thus,  the microbes  digest each  other due
to endogenous  respiration  and autooxidation.   A simplified  summary of the  aerobic
conversion of organic materials  into cellular materials and the subsequent breakdown
into by-products and residue is  shown in Figure 10.

      The results  of aerobic digestion on various biological  sludges are  shown in
Figure ll.  Approximately 10 to  15 days detention are  required on an  annual average
to achieve 50  to  60 percent reduction of the volatile content of the  sludge.   This
reduction represents about  80 to 90 percent  reduction of the degradable content and
is generally suitable for  land  application.   Tests  on refinery sludges (3) have
indicated that  about 0.5 to 1 .2 Ib of oxygen  are required  per  Ib  of VSS destroyed
per day.

-------
 208

       Thickening  of  Biological Sludges.      There appear to be three alternatives by
 which refineries can concentrate  biological sludges.  These  are flotation thickening,
 gravity thickening, and  centrifugal thickening.  The primary objective of a thickner
 is to reduce sludge volume by removing the bulk water and concentrating the sludge,
 thereby increasing the system effectiveness  and lowering the cost of subsequent dewater-
 ing processes
       Gravity  Thickening
       Gravity  thickeners utilize  gravity sedimentation and compression for sludge con-
 centration.  The  key design parameters are hydraulic and  solids loading rate.   If the
 surface loading rate  is excessive,  poor solids recovery will result.   If the solids  load-
 ing (Ib solids applied/sq ft-day) is too great,  then there  will  also  be poor solids
 recovery and low underflow  solids concentration.  Solids  loading rates for gravity
 thickening of waste  refinery sludges range from 2 to 15  Ib/sq  ft-day.  Biological
 refinery sludges can  be concentrated by gravity  to a concentration  of 2  to 4 percent
 solids by weight.

       Gravity  thickeners do  not  require much operator attention and will perform fairly
 consistently provided the influent hydraulic flow and solids loading do not vary  sub-
 stantially.   In extremely warm climates,  gravity thickeners may generate obnoxious
 biological  odors  if they precede  the stabilization process.

       Centrifugation
       Centrifugation  has  been utilized  sparingly for thickening waste  biological sludges
 in the refinery industry.   There  is little information  available  on the application of
 disc  centrifuges for thickening of refinery biological sludges.  Basket centrifuges  are
 capable of thickening waste  activated  sludge  to  levels of  5  to  6 percent concentration'
 with  80 to 95  percent  solids capture.   One refinery has reported using basket  centri-
 fuge  for thickening to  a  concentration  of 8 percent.  However, excessive maintenance
 was experienced  because of vibrational problems.

       Air Flotation
       A third  method of thickening  is by  air flotation whereby  air bubbles are released
 from  solution and  attach  themselves to  the sludge floes.   The air-solids mixture than
 rises  to  the surface of the  basin  where  it  concentrates and is removed.   The primary
 variables with regard to flotation  thickening are  recycle  rate,  feed solids concentration,
 air-to-solids  ratio, and solids and  hydraulic rates.   Pressures between 50 and 60 psi
 are commonly employed for flotation thickening.   Recycle  rates  on  the order of  100  to
 500 percent, solids loadings  of 2  Ib/sq  ft-hr,  and  hydraulic  loading rates (including
 recycle) in the range of  1  to 4 gpm/sq ft are  common  of flotation thickeners.

      Results  of tests  on refinery  sludges indicate that these sludges are quite amenable
to flotation thickening  (3).   These results are  summarized  in Table  14 and Figure 11A.
These  data  indicate that the  optimum pressure  was about  50 to 55 psig and  that  polymer
polymer  addition was not justified.  The correlation of these data in  Figure  12  shows
the optimum air-to-sol ids ratio to be about  0.01  Ib/lb  to  obtain a float  solids  of 3.5
percent.

-------
                                                                                     209


      Dewatering of Biological  Sludges.
      Vacuum  Filtration
      Vacuum  filtration is one of the most  common methods  used to dewater waste-
water sludges.  Water is  removed  from  the  sludge under an  applied vacuum through  a
porous media which  retains the solids.   Chemicals are usually required in order to
decrease  the specific resistance of  the  sludge,  thereby allowing higher sludge loading
rates (Ib/sq ft-hr)  and, thus, smaller, more economical filters.  The results of several
tests on  refinery wastewaters  indicate that  ferric  chloride or a  combination of lime and
ferric chloride usually provides the optimum coagulant from  an economical  and per-
formance standpoint.   The optimum ferric chloride dosage  usually  ranges  from  200 to
400 Ib/ton  of  dry  solids.   Tests conducted  on waste activated sludge indicated that
there is  not significant advantage  of  a  precoat filter based on  overall economics.
Tests indicate  that vacuum  filtration of  refinery biological sludges  is usually able to
obtain solids concentrations in the  range of 10 to 16 percent at filter loading rates
ranging  from 1  to  5 Ib/sq ft-hr.

      The results of  filter leaf tests on  both a  raw  and digested refinery  waste
activated sludge are presented in Table  15.  This table  indicates  that the digested
sludge presented better dewatering  characteristics than the undigested sludge.

      Cenrrifugation
      Basket centrifuges have been  found to be the best centrifugal  method of con-
centfatu^j waste biological solids.  The tendency for solid  particles to separate  from
the water portion  of the  slurry  due to a slight  difference in density is accelerated by
the application of centrifugal force.   In fact,  by spinning  the  slurry at appoximately
1,500 rpm, a  centrifugal  force of over  1,000 times  that of  natural gravity is
extended and the  particulate  matter is  concentrated  on the external wall of the
revolving basket or bowl.  A schematic of  the internal components of the basket
centrifuge is presented in Figure  13.

      There are a  number of applications of utilizing basket centrifuges  in  the refining
industry.  Normally, the  centrifuges  provide  a cake solids  ranging from 8 to 18  per-
cent concentration with biological  sludges.   A correlation of the  cake concentration
and solids recovery and cake solids and distance  from  the  wall is  presented in Figure
14.   It should be  indicated  that there  have been some problems in the utilization of
the basket centrifuge due to mechanical vibrations of  the  centrifuge.

      Pressure  Filtration
      Pressure  filtration is another alternative for the dewatering of the  biological
solids.   This process usually can obtain  a solids concentration approaching  50 percent
with biological sludges, thus  producing the  driest cake for  disposal.  The pressure
levels for the  pressure  filtration systems  range from  50 to 225  psig.   Normally,  it has
been  found  that  the  pressure filtration system requires  higher chemical dosages than
for the vacuum filter and centrifugation system; however, these  produce a  higher
solids concentration,   In  order to obtain  an acceptable cake,  the manufacturers
recommend a pressure specific resistance of less than  3 x 101
cm

-------
 210


        Heaf Content of Sludge.     The heat content of various samples of dewatered
secondary refinery sludge  has been determined in a bomb calorimeter (3).  The average
fuel  value for the undigested secondary sludge was approximately 8,000 BTU's per Ib
of dry solids.  This value  is within the  expected  range of secondary waste activated
sludge with  70 percent voatile content.  Approximately 6,000 BTU's per Ib of dry
solids were found after aerobic  digestion which indicates that part of the vlatile con-
tent  matter is destroyed by aerobic digestion.  It is obvious that stabilization would
not be desirable if incineration were the means of final disposal.

        Ultimate Disposal.     The alternatives  for ultimate disposal of the biological
sludge are similar to those for the waste oily sludge with the expection of land
farming.  In many cases,  a  thick slurry of activated sludge can be placed in 4 to 8
in.  layers and utilized as a limited nutrient and soil  conditioner.  In this case, additional
water is removed by  natural  evaproation and infiltration into the underlying soil.
Following initial periods of drying, the sludge layer can be disced to encourage the
activity of aerobic soil bacteria.   The  soil bacteria working  in conjunction with
plant growth will further stabilize  the waste solids and transform them into a hemocyte
material suitable for agricultural use as a soil conditioner.


SUMMARY

A design summary table for the  thickening and dewatering alternatives of oily sludges
is presented  in Table 16.   A summary on biological  sludges is presented  in Table 17.

REFERENCES

 (1)    Adams, Carl  E., Jr. and W. Wesley Eckenfelder,  Jr.  Process Design Techniques
        for Industrial Waste Treatment.   Nashville, Tenn.:Enviro Press,  1974.

 (2)   Adams, Carl  E., Jr., W.  Wesley Eckenfelder, Jr., and Gary M. Davis.
       'Characterization and  Biological Treatability Investigations of a  Refinery Water
        Discharge."  Unpublished  report for Ashland Oil, Incorporated,  March 1972.

 (3)    Adams, Carl  E.,Jr.,  W. Wesley Eckenfelder, Jr., and Robert C. Lasater.
        "Fina I Process Design for Treatment of a  Combined Industrial  Wastewater."
        Unpublished report for Lower Neches Valley Authority, December 1972.

 (4)    Beychok, M.R.  Aqueous Wastes from Petroleum and Petrochemical Plants.
        London:  John Wiley &  Sons, 1967.

 ('))    Cross, F.L. and J.R.Lawson.  "A New Petroleum  Refinery." American
       Institute of Chemical  Engineering Symposium Series,  Vol. 70, No. 136, p.812.

 (6)   Engineering-Science, Inc.   "Wastewater Treatment Facil ities and  Ballst Handling
       Systems."  Unpublished report for the Suntide Refining Company, December 1972.

-------
                                                                                 211
  (7)    Ford, Davis, L.  "A Preliminary Engineering  Study for Wasrewafer Treatment and
        Pollution Abatement."  Unpublished report- for The Sun Oil Company,  Sept. 1971.

  (8)    McCrodden, B.A.  "Treatment of Refinery Wastewater Using  FilfraHon and
        Carbon Adsorption." Proceedings of the 29th  Industrial  Waste Conference,
        Purdue  Univeristy,  1974.

  (9)    Peoples, R.F., P.  Krishnan,,  and R.N.  Simonsen.  "Nonbiological Treatment
        of Refinery  Wastewater. " Water Pollution Control Federation Journal, Vol.44,
        No.  11  (November 1972), pp. 2120-2128.

 (10)    Sebesta, Ed C. and David L.  Ford.  "A Preliminary Engineering Study  for Solid
        Waste Disposal and Pollution Abatement." Unpublished report for  The  Sun Oil
        Company,  February 1972.

 (11)    "Session on Sludges and Spent Clays."  18th Mid-Year Meeting, American
        Petroleum Institute, 1964.

 (12)    "Treatment of Recovered Oil Emulsions." Disposal of Refinery Wastes.  New
        York:  American  Petroleum Institute, 1969.
DISCUSSION

John Smith :  You spoke of the rules-of-thumb reference to air oxygen input rates
and sludge concentrations on bio-sludges.  Does this also refer to those sludges
generating on the rotating bio-disk units or have you found a difference?

Carl Adams:  We have not conducted much sludge handling on bio-disk sludge,
although we have performed about six pilot studies with  the bio-disk.  One was
on a very large scale and  we are presently evaluating  the bio-disk at Cranston,  Rhode
Island on a fairly large scale. We are not far enough  along into the study to
comment on sludge handling.  I would assume that the sludges dewater as wel I or
better than conventional activated sludge.   I say this because the material sluffs
off in chunks.  Most of the wastes we have  evaluated have been high-strength,
difficult-to-settle wastes using the activated sludge process.  Consequently, we
examined the bio-disk in order to establish  a fixed medium.   Thus,  these
wastes are bad examples for me to quote as  representative of bio-disk performance.
However, with the pulp and paper industry  the sludge  from a bio-disk settled
very well.   We did pilot a large 3-meter unit, with a pulp and paper waste,
and observed  very good  settling.

John Smith:  The work we have done to-date indicates the settling  is quite
good but the oxygen input required for digestion is considerably higher?

W.C. McCarthy:  From a  leaching point of view, what  criteria are used  to decide

-------
212


if a treated oily sludge would be suitable for landfill?

Carl Adams:   I can gef ouf of this one easily by referring this question to the next
apeaker.  Genrally, you are required to contruct a monitoring well if you can get
permission from the authorities.   The easy way  for the regulatory authorities is to
require you to line the area and treat the runoff.  If you do not want to do that, you
generally have to  convince the authorities that  you do not have heavy metals which will
leach out.  You can genrally do this; however, ash from incinerators, boilers and so on
contain quite a bit of heavy metals from the coal that was burned.  We usually argue that
the rain water won't leach these  metals because pH is sufficiently high that you will not
get leaching .  The only way  you can resolve it  if there is an argument is generally
to construct monitoring wells.   If a leaching problem then   is observed to develop, you
may have to line the pond  and treat the storm water runoff from the landfill.  We  argue
against it generally and hope  we are not filling the pond with anything that would cause
leaching and problems.

-------
                                                                                  213
BIOGRAPHY

   Carl E.  Adams, Jr. is President of
Associated  Water and Air Resources
Engineers,  Inc.  (AWARE) located in
Nashville,  Tennessee, and President
of AWARE  Engineering in Houston,
Texas, and has consulted to over
200 United States and foreign indus-
tries.  He holds the following degrees:
B.E.,  Civil Engineering, Vanderbilt
University; M.S. , Environmental
Engineering, University of Texas.
He is a professional engineer, re-
gistered in  18 states.  Dr. Adams
serves  as an Adjunct Professor in
the Department of Environmental
Water  and  Resources Engineering
of Vanderbilt University.  He holds
membership in more than six pro -
fessional societies, has authored and
presented more than 50 technical
papers, and edited and co-authored
three books dealing with industrial
wastewater treatment technology.
   Robert M. Stein is Director of
Operational Services for Associated
Water and Air Resources Engineers,
Inc.  (AWARE) of Nashville, Ten-
nessee.  He holds the following
degrees:  B.A., Applied Science,
Memphis State  University,  Memphis,
Tennessee; and M.S. , Environmental
Engineering, Vanderbilt University
Nashville, Tennessee.  He is a pro-
fessional engineer, registered in the
State of Tennessee.  Mr. Stein was
associated for several years with the
State of New York in the Department
of Transportation.

-------
TABLE 1 TYPE OF REFINERY SOLID WASTE
                                                                    214
Source
Crude Oil Storage
Product Storage
Crude Processing

Catalytic Cracking
Catalytic Reforming
Alkylation
Coking

Feedwater Treatment
Wastewater Pollution Control
API Separators
Flotation
Separators
Secondary
Clarifiers
Sludge Thickeners

Combustible Non-Combustible
Waxy Bottoms Sand, Rust, Silt
TEL Sludge
Sand, Rust, Silt,
Salt
Catalyst
Catalyst
Sludge Corrosion Products
Coke, Waxy
Ta i 1 i ngs
Lime Sludge

Slop Oils Sand, Silt

Float Solids


Lime Sludge
TABLE 2
Biodegradable















Bio Sludge


CHARACTERISTICS OF REFINERY SOLIDS WASTES (REF.5)


Waste Type
API Separator Sludge

Tank Bottoms
Chemica1 Treatment Sludge
Air Flotation Froth
Precpat Vacuum Filter
Sludges
Biological Treatment Sludges
Raw
Mechanical 'y Thickened
Centrifuged

Vacuum Fi'tered
Screw Pressed
Waf<=r Treatment Sludge

Typical Composition, Percent
Oil or Volatile Inert S
Hydrocarbon Water So'ids So'ids
15 66 6 13

48 40 4 8
5 90 5
22 75 3
22 29 49


0 98 1.5 0.5
0 94 4 2
0 85 10 5

0 75 15 10
0 40 40
00 95 5



Characteristics
F'uid slurry of oil,
water and sand
Oil -water mixture
Slightly viscous Puid.
Thick,oily fluid
Temperatures


Water Consistency
Thick, but pumpab'e
Viscous-peanut butter
consistency
Wet crumb'y solid.
Intact, solid cake.
Pumpable Puid, some-
times ge'atinous

-------
         Waste Types
                                           TABLE 3
                        ESTIMATES OF REFINERY SOLID WASTE QUANITIES
                                      Unit Loads
API Separator Sludge
Chemical Treatment
   (API Separator Effluent)
Biological Sludges

Water  Treatment Sludge
   A.  Lime Soda Ash

   B.  Ion Exchange

Office Wastes

Cafeteria
                                200 mg/1 Suspended Solids
                                50 mg/1 Suspended Solids
                                        Removed Only
                                0.7 Ib Dry Solids per Ib
                                        BOD Removed
                                2 parts Dry Sludge per
                                        1  part Hardness Removed
                                0.4 Ib Salt per 1,000 Grains
                                        Hardness
                                1 .0 cu yd per Employee
                                        per  month
                                0.6 ''3 per Meal
 Water
 Oil
 Volatile Solids
 Ash
 Total
                                           TABLE 4
                  SUMMARY OF TANK BOTTOMS SLUDGE CHARACTERISTICS (REF.ll)
                                      ANALYSIS  (PERCENT BY WEIGHT)
                             Average                       Maximum
                                                    Minimum
39.4
47.9
4.4
8.3
100.0
98.0
95.0
13.0
55.3

5.0
1.3
0.0
0.0

Oil
Water
Sediment
                                          TABLE 5
                       COMPOSITION OF SLOP-OIL EMULSION (REF- 11)
                              Maximum
                                                         Percent by Weight
                                                           Minimum
80.9
81.0
 8.0
13.2
18.0
11.1
                                                     Average
40.2
54.8
 5.0

-------
216
                                          TABLE 6
                       OPERATING CONDITIONS FOR PRECOAT FILTERS
                    (Treating Slop-Oil Emulsions, Separator Bottom Sediment,
                                  and Acid Oil )
  Precoat Medium = Medium-grade diatomaceous earth
Temperature, F 180
Vacuum, in. of mercury 18 to 20
Knife-cut depth, in. per revolution 0.0040
Drum speed, min per revolution 3 to 4
TABLE 7
COMPOSITION OF API SEPARATOR BOTTOMS (REF.ll)
Analysis as Taken from the Separator (Percent by Weight)
Average Maximum
Water 82.4 98.3
Oil 10.9 20.7
Volatile Solids 1.4 2.3
Ash 5.3 15.8
Total 100.0
TABLE 8
BENCH SCALE FLOTATION THICKENING (REF. 10)
Untreated Water Float Solids Sludge
Sludge TSS Phase TSS Volume Solids Volume
Sludge Mixture (mg/1 ) (mg/1) (ml) (%) (ml)
WEMCO Float Solids 730 73 15 4.4 0


WEMCO Float Solids 1,136 96 30 3.5 0
WEMCO Float Solids 420 66 10 5.5 Trace


WEMCO Float Solids 2.,260 830 40 5.2 Trace

Contaminated API Btms = 10% — — 100 — 100
Tank Bottoms = 20%
Decant Pit Sludge = 10%
Water = 60%
Contaminated API Btms = 20% — — 100 -- Trace
Water = 80%
Tank Bottoms =50%
Water =50% ~~ "" "" '7U








Minimum
61.2
1.1
0.5
0.4



Reco-
very
(%) Comments
90 Water Phase
COD = 270 mg/1
BOD, =140 mg/1
o
92
84 Some of the float
was dispersed
after 10 min.
93 Some of the float
was dispersed
after: 10 min.
Solids dispersed
rapidly.
Solids dispersed
during 10 min.
Most of solids sank.
Float dispersed
rapidly
= "  -r,r,H;t-;0ns:  1 )  100 percent Recycled Based on Sludge Volume.  2.) 50 psig Saturation Pressure

-------
                                      TABLE 9
                      FULL-SCALE CENTRIFUGE  PERFORMANCE
                                                                                                 217
Percent
Solids
Scroll Centrifuge Feed 1 .6
Scroll Centrifuge Centrote 0.7
Scroll Centrifuge Sludge 59
Disc Centrifuge Water Discharge 0.01
Disc Centrifuge Solids Discharge 0.1
Scroll Centrifuge = 30 gpm
Disc Centrifuge = 30 gpm
TABLE 10
FILTERABILITY OF OILY SLUDG ES (REF.7)
Special Test Specific Resistance
Test Sample Conditions sec /gm x 10
Composite of Plant Oily Sludges- D , .... - „
T. . . . ' Precoated Filter 0.35
Thickened
Composite of Plant Oily Sludges- Precoated Filter- _
Percent
Oil
6.9
5.9
2.5
0.5
0.1


Comments
Recovered
mg/1 of oi
Recovered
36,900
1 from sludg
40,000
     Thickened

Thickened Oily Sludge

Composite of Plant Oily
     Thickened

Composite of Plant Oily
     Thickened
Sludges-


Sludges-
Composite of Plant Oily Sludges
     Thickened
Cloth Media

Precoated Filter-
Heated to 165° F

Precoated Filter-0.67 mg
Spent Water Treating Lime/
1 .0 mg of oily solids heated
to  165 F

Precoated Filter-2,600 mg/1
of  FeCI, Heated to 165°F
       •J
0.19
                                          0.48
mg/1 of oil from sludges

Blinded Filter Badly

Recovered 74,000 mg/1
of oil  from sludge


Recovered 85,000 mg/1
of oil  from sludge


Recovered 31,000 mg/1
of oil  from sludge
                                           TABLE 11
                     AVERAGES OF VACUUM FILTER RATES AND PERCENT OF
                                    FILTER  TIME REQUIRED
Service
Slop-oil emulsion
Separator sediment
Flocculation sludge
Acid Oil
Pre coating
Downtime
Average Charge
to Filters (Barrels/day)
427
377
627
129
—
—
Average Filter Rate
(gal/hr-sq ft)
1.7
2.8
8.6
2.4
--
—
Filter Time
Required (%)
37.8
20.1
10.7
7.8
8.5
15.1

-------
                       TABLE  12



SUMMARY OF RESULTS FROM FILTER PRESS PILOT UNIT (REF. 10)
a
Run3
No. Feed Mixture
6

7

8

9

10

11



12

13

14

15

16


Base

Base

Base

Base

Base

Base



Base

Base

Mixture0

Mixture5

Mixture

Mixtureb

Mixture5

Mixture5



Mixture5

Mixture5

Non-Contaminated
API
Separator Sludge
Thickened WEMCO - 78%
Base
Base


Mixture5 » 22%
Mixture0


Admix
Added
Clay &
0.6 Ib/lb D.S.
Clay @
0.6 Ib/lb D.S.
Clay @
0.6 Ib/lb D.S.
Clay @
0.6 Ib/lb D.S.
Clay @
0.6 Ib/lb D.S.
Clay @
0.56 Ib/lb D.S.


Clay @
0.56 Ib/lb D.S.
Clay 
-------
                                                                                                   219
                                           TABLE 13
               INCINERATOR CONSIDERATIONS OF FLOW AND SOLIDS CONCENTRATION (REF. 4)
 Effluent flow (gpm)
2 wt.  % lean sludge (gpm)
Optimum Sludge Concentration
to incinerator   (wt.  %)
50
100
200
500

1,500
1
2
4
10

30
3
7
17
50
b
7
-------
220
                                        TABLE 15

                            FILTER LEAF TESTS RESULTS (REF.3)
                                                           Time
                                               Cake
Sludge Run No.
Raw° lb
2C
3C
4C
5C
6C
7C
8C
9°
ioc
Digesfed 1
b
2
3b
b
4
b
5
b
6
b
7
b
8
9b
Feed Solids
(%)
1.04
1.04
1.48
1.48
1.48
1.48
1.48
1.48
2.29
2.44
1.73

1.73
1.73

1.73

1.73

1.73

2.74

3.19
2.18
Vacuum
(in. Hg)
15
15
15
10
18
15
10
15
15
15
15

18
10

15

15

15

15

15
15
Form
(min)
2.00
2.00
2.00
2.00
2.00
1.00
1.50
3.00
2.00
2.00
2.00

2.00
2.00

1.00

3.00

1.50

2.00

2.00
2.00
Dry
(min)
3.00
3.00
3.00
3.00
3.00
1.50
2.25
4.50
3.00
3,00
3.00

3.00
3,00

1.50

4.50

2.25

3.00

3.00
3.00
Solids Thickness
(%) ( in.)
8,9
9.3
9.2
8.6
7.9
7.2
7.5
7.9
7.1
7.0
13.1

13.6
12.1

11.5

12.3

10.8

11.0

11.0
15.9
1/64
1/8
1/8
1/8
3/16
3/16
3/16
1/4
5/16
3/8
3/16

1/4
5/16

1/4

7/16

3/8

7/16

7/16
3/16
Loading
(Ib/sq fr-hr)
0.16
0.73
0.96
0.92
0.99
2.31
1.54
0.88
2.05
1.98
2.24

2.57
3.13

5.08

3.00

4.62

4.63

4.65
1.95
 Chemical dosage = 20% FeCl  • 6H,O +5% CaO.

b
 Media No.  POPR-873, polypropylene from Eimco Envirorech


°Media No.  POPR-929, polypropylene from Eimco Envirorech

d.
 Chemical Dosage - 15% FeCl,
6H2O +5% CaO

-------
                                                                                                     221
                                            TABLE 16
                SUMMARY FOR THE THICKENING AND DEWATERING ALTERNATIVE
                                        OF OILY SLUDGES
Type of
Sludge
Oily
Sludge
Process
Thickening
Average
Performance

Design
Parameters

Comments

                Gravity


                Flotation



                Dewatering

             Centrifugal-ion
                   3-7 percent solids
                   50% recovery

                   84-93% solids recovery
                   3.5-5.5% float solids
                          5-30 Ib/sq ft-day


                          Untreated sludge 420-2,200
                          mg/l TSS, 100% recycle
                          50 psig saturation pressure
                   75-90% solids recovery when Flow rate 50-350 gal/hr
                   charged with heated oil sludges
                   Oily phase wil I consist of
                   1-5% solids and cake will
                   contain 50-60% solids.
                   Nozzle ejector centrifuge
                   separates 95-98% of oil
                   in oily phase and  2-5% in the
                   nozzle and nozzle water.
                   30-50% of nozzle ejector
                   feed solids will be removed
                   with oily phase.
 Low solids and oil
 recovery

 Sucessfu! for froth
 flotation sludges
 but impractical for
 other oily  sludges
                                                      A vertical solid
                                                      bowl type not re-
                                                      commended for de-
                                                      watering most oily
                                                      sludges.
Oily
Sludge
(Cont'd)
Vacuum
Filtration
                Pressure
                         Avg .  Filter Rates gal/hr-
                         ft  slop-oil emulsion-1 .7
                         separator sediment-2.8
                         flocculation sludge-8.6
                         acid oil-2 .4
                         Filter Time Required,per
                         cent,slop-oil emulsion-
                         37.8 separator sediment-
                         20.1, flocculation sludge-
                         10.7, acidoil-7.8
                         precoating-8.5, downtime-
                         15.1


5-20% oil, 40-70% solids 2-hr cycle time, feed  con-
cake                     fents 12-38.5% TSS,6-23%
                         oil. Temperature of feed
                         58-180° F.
To dewateroily sludges,
a precoat vacuum filter
should be used and solids
should be  heated above
170°F.  Sol ids should be
suitable for landfill
following vacuum filtra-
tion.  Increased tempera-
ture for feed  improves
psrformance.  Addition of
Spent clays decreases oil
recovery and sol ids F. H .
rate
Heating of feed required
for satisfactory filtration.
Lime or spent caustic added.

-------
 222
                                             TABLE 17
                              SUMMARY OF BIOLOGICAL SLUDGES
Type of
Sludge Process
Biological
Sludges
Average
Performance

Design
Parameter

Comments
Similar to oily sludges
but should be stabilized
due to putrescible nature
Biological
Sludges
                Thickening

                Gravity
                Centrifugation
                Flotation
                Dewatering
                Vacuum
                Fi I (ration
Centrifugation
                Pressure
                Filtration
2-4% solids con-
centration by
weight

5-6% with  basket cen-
trifuges, 80-95% solids
capture

Cake solids 2.6-4.0%
Solids concentration
10-16%
8-18% cake solids
Solids recovery 20-90%
50% solids
                                             Loading rate 2-15
                                             Ib/ft   day
                                             Feed rate 6-200 gpm
                                            50-60psi, 100-500%
                                            recycle.  Solids loading
                                            216 sq ft-ht hydraulic
                                            1 -4 gpm/ft

                                            Loading rate  1-5 Ib/sq
                                            ft-hr
                          Feed rate 4-90 gpm
                                             Pressure levels 50-255
If surface loading rate
is excessive poor
solids recovery results

There is little informa-
tion avaiable
Sludges quite amenable to
flotation thickening
Chemical  usually re-
quired to decrease
specific resistance.
FeCLor lime and FeCL
usually. Optimum FeCL
200-4001 b/ton of solids
Basket type found to be
best.  There have been
problems due to mechanical
vibrations in the centrifuge

Normal ly requires
higher chemical dos-
ages than for vacuum
filters or centrifuges.

-------
 Refinery
Waste wafer
                                                                     EXCESS
                                                                 BIOLOGICAL  SLUDGES
           FIG. I. SLUDGE  HANDLING AND DISPOSAL OPTIONS FOR REFINERY SLUDGES
                                                                                              CO

-------
224
        INFLUENT Jv
         SLUDGE
                      TOP LIQUID
                      ILEVEL
                       EMULSION
                       WATER
                         INTERFACE
                                             EXCESS
                                             WATER
                                                      TO API  SEPARATOR
                                                 -WATER DRAW PIPE
                                                   SLUDGE  AND  EMULSION
                                                   LEVELS  DETERMINED BY
                                                   PERIODIC SAMPLING
                                          V
                                          .STEAM ON OUTSIDE OF  TANK

                                            SLUDGE
                                          TO SLUDGE STORAGE
                                          TO  SLUDGE STORAGE
            FIG. 2. PRIMARY OILY SLUDGE  DECANT TANK
                            FIRST STAGE
                            CENTRIFUGE
                          (Horizontal Solid Bowl)
                            SOLIDS
                                                FINE SCREEN
                                                      RECOVERED OIL
                                                           WATER
                                                           SOLIDS
                                         SECOND STAGE
                                         CENTRIFUGE
                                         (Disk-Type)
             FIG  3. TYPICAL OIL RECOVERY AND SOLIDS
                    REMOVEL  SYSTEM FOR OILY SLUDGE

-------
     CHARGE MIXTURE


     CONTAMINATED  API  SEPARATOR  BOTTOMS

     SLUDGE DECANT PIT  SLUDGE 	

     398  TANK  BOTTOMS 	

     209  TANK  BOTTOMS 	

     253  TANK  BOTTOMS 	
                                                     28

                                                     36



                                                     12
                                                                  225
          ID
          o
          tr
          <
          a:
          o
          LU
          >£

          <
          O
          a:
          LU
          >
          o
          o
          LU
          a:

          co
          o

          _i
          o
          CO
             100
              80
          60
              40
          20
                               Volatile Suspended
                                  (oil free)

                                I	
                           _Total Suspended

                           Solids (oil free)
                           Recommended
                             Design
                                        \
                          I         2         3

                      CHARGE RATE  TO P600 (gpm)


                   FIG. 4. SOLIDS  RECOVERY FROM

                          OILY  SLUDGE  ( REF 10  )
   100
o^

>-
cc

CO
LU
O
o:
LU
Q.
    90
9   80

o
CO
70
    6Q.
     0
                                                O.A 925 G S

                                                • A 2100 G'S
                                                            70
         50      100      150     200     250


                  SLUDGE FEED RATE, (gals./hour)
                                                   300
          FIG. 5. PERFORMANCE OF PILOT SOLID-BOWL
                 CENTRIFUGE (REF. 9  )
                                                            60
                                                        50
                                                         40
  30
350
     CO
     a


     d
     CO

     LU

     3
     O
                                                                LU
                                                                O

-------
CHARGE MIXTURE !S CENTRATE FROM
SHARPLES P600 CENTRIFUGE
i
3E SOLIDS RECOVERED SOLIDS RECOVERED IN
N OIL PHASE (%) NOZZLE WATER (%)
-f» CD OD
0 0 00 - M C
CC
X
o
20
(
CENTRIFUGE: T-
SH

• 	
- TOTAL
~ oru in
oUL 1 U
AV U L A
SOLID


1. TUBULAR BOWL
<\RPLES NOZLJECTC


. SUSPENDED
S (OIL FREE
FILE SUSPEN
S (OIL FREE
A 	 - 	 —
A
•
•
,
J
^

DED
'-) A
___ 	 	 	

>



'

0

O P
• P
° £
• T
P
A P
A N
0 N
^ s
CO
Q
_J
O
OT 55
_J
<
1-
0
1-
UJ
^ 50
O
CO
CO
UJ
a:
°- 45
tr
UJ
i-
D 500 1,000 1,500 U.
DISK CENTRIFUGE FEED RATE (cc/min) 4O
H ADJUSTED WITH LIME, ONE HOUR RUN
H ADJUSTED WITH LIME, TWO HOUR RUN
H ADJUSTED WITH LIME, ONE HOUR FILTRATION,
NE HOUR HOT WATER WASH
HICKENED WEMCO FLOAT SOLIDS ADDED,
H ADJUSTED WITH LIME, TWO HOUR RUN
H ADJUSTED WITH SPENT CAUSTIC, TWO HOUR RUN
0 pH ADJUSTMENT, TWO HOUR RUN
0 pH ADJUSTMENT, TWO HOUR RUN, FEED MIXTURE
IMULATED PROCESSING ALL OF THE SPENT CLAY













A

0A





•
A
•


<

\

cj
)



•



o
CN
CM
FIG. 6. SOLIDS RECOVERY  FROM
      A SECOND - STAGE DISK
      CENTRIFUGE (REF. 10 )
60    70   80    90    100   110    120
   AVERAGE FILTRATE TEMPERATURE  (°F)
  FIG. 7.  EFFECT  OF FILTER PRESS
         OPERATING CONDITIONS ON
         CAKE  SOLIDS (REF. 10 )

-------
       BASIS
UJ
H-
2  OC
UJ  <
r™  til

i  J:
o
   •w-
o:
UJ  u.


I;

o  o


UJ  (0
13  ID
O  O
   X
(O  h-
o
o
UJ
D
U.
       350
       300
       250
       200
        150
  Z    100
             FUEL  OIL COST= $ 3.60 /  F.O.E.  BBL

             GROSS HEATING VALUE OF OIL =  18,500  BTU / Ib     227

             DENSITY  OF FUEL OIL =  7.35 Ibs /gal.

             STACK  TEMPERATURE = 800°F AND EXCESS  AIR =  100%

             GROSS HEAT REQUIRED  FOR WATER EVAPORATION = 2400BTU/lb
              2    4    6    8    10

                SOLIDS PRODUCTION
12   14  16   18

  Ibs  DRY SOLIDS

    YEAR
                                                     20
          0      10,000   20,000  30,000  40.00O  50,000

                SOLIDS  PRODUCTION  ( Ibs DRY SOLIDS / DAY )

         FIG. 8.  INCINERATOR  FUEL COSTS  DUE

                  TO  SLUDGE  WATER CONTENT

                  ALONE ( REF. 10 )

-------
  228
    STABILIZATION
THICKENING
DEWATERING
DISPOSAL
                    FIG. 9. BIOLOGICAL SLUDGE HANDLING
                           ALTERNATIVES
                          -STANDARD ACT SLUDGE	



                                    REMOVED
                                     WASTE I
                                   ACTIVATED I
                                     SLUDGE (ACTIVATED
                                         SLUDGE SOLIDS
                                      t   I 	'
                    SOLIDS DESTROYEE
                     IN AEROBIC
                      DIGESTER
                                           RESIDUAL1 SOLIDSj
                                                   --—
                                            FOR DISPOSAL-
                                        OXIDATION
                                        ENDOGENOUS
                                        RESPIRATION
                            TIME  OF AERATION •
FIG. 10. SCHEMATIC DEPICTION OF AEROBIC  BIOLOGICAL TREATMENT

-------
                                                         229
   Q
   LJ
   cr
   CO
   to
            Bio-chemical
            Mixed Pulp 8 Paper Waste
            Refinery
          A  Mixed Domestic Sewage
          T  Textile 8 Domestic Waste
            2   4
6   8    10   12
AERATION TIME (days)
        FIG. 11. TYPICAL VOLATILE SOLIDS
               REDUCTIONS BY AEROBIC DIGESTION
1000
                            Recycle = 230%
                               SS  = 0.9%
    0    0.20    0.40   0.60    0.80    1.00    1.20    1.40
                         TIME (min)
 FIG. 12.   RISE OF  SLUDGE  INTERFACE WITH
            TIME

-------
                                                          NO
                                                          CO
                                                          O
  0.040
  0.032
  0.024

CO
co 0.016
  0.008
     0
      0
.0      2.0      3.0

 PERCENT FLOAT SOLIDS
4.0
          FIG. 13. DETERMINATION OF
                  OPTIMUM AIR/SOLIDS
                  RATIO (REF. 3 )

-------
                                           231
                                FEED RECYCLE
                                 VALVES
SKIM DIVERSION
  VALVE
                                         FEED FROM
                                       AEROBIC DIGESTER
                                         RECYCLE TO
                                       AEROBIC DIGESTER
                            UNLOADER
                            ASSEMBLY -v
                              T _£._f__f ^* J J f^^f f^,f £ J f J J f f
                                           CURB
                                           BOWL
                                         OR BASKET
   FIG. 14. INTERNAL COMPONENTS OF BASKET
         CENTRIFUGE

-------
232
         FIGURE A.
      MEAN CAKE SOLIDS

      CONCENTRATION

          (WT. %)
                   20"
                   15"
                   10
                    5 +
                    0    20   40   60    80   100


                        MEAN SOLIDS RECOVERY (%)
         FIGURE  B-
                 KNIFE
                 CAKE  SOFT SKIM CAKE
     CAKE SOLIDS  15

         (WT. %)
                     12345    6

                     DISTANCE FROM BOWL WALL (IN.)
          FIG. 15. PERFORMANCE OF BASKET
                  CENTRIFUGE ON REFINERY
                  BIOLOGICAL SOLIDS (REF  6 )

-------
                      PROCESSING OF WASTE  OILY SLUDGES

                 Jacoby A. Scber,  Principal Environmental Engineer
               Fluor Engineers  and Constructors,  Inc.,  Houston, Texas

INTRODUCTION

      Petroleum crude oil, fuels,  and lubricants become waste  components when  emul-
sified with water,  solids, and/or debris and are potential  contaminants  of  surface and
subsurface  waters.   The volume of oily waste  is potentially large due to  large quan-
tities of petroleum oils  handled and used.   Sources of the waste are from  spillage of
crude oil and refined products  and from petroleum refinery operations.  This paper
deals with the refinery  operation, but the findings  are  expected to be applicable for
most petroleum sludges.  The title of this paper needs some clarification because the
use of the word  "sludges" is all-inclusive-oils, emulsions and  sludges will  all be
treated  as  an umbrella topic.

SOURCES

      Almost all  operations  of  the petroleum industry, including exploring,  producing
(extracting), storing,  transporting,  and refining of crude oil and the storing,  distribut-
ing, and handling  of products  are potential  sources of oily sludges.

      Accidental spills of crude oil  and petroleum products during  the  handling,
storing,  and transporting operations  are the  principal  cause for the formation of oily
sludges  in  large  quantities.   At refineries, accidents  seldom cause oily  sludge pro-
duction.   Common sources are  incoming crude oil,  ship  ballast water,  tank and
vessel cleanings,  oil-water  separators,  and numerous  miscellaneous sources  such  as
sewer boxes  and  emulsion breaking facilities (demulsifier).  The quantity of oil  sludge
for disposal at a refinery does  not depend only upon  the nature of the  crude  and
processing  units.   Oily sludge  formation  can be minimized  by prudent  operating
practices,  sensitive attitudes and suitable  control methods.   Generally, most of the
oily sludges  accumulate  in oil-water separators and in tank bottoms.

      Crude  oil shipped to the  refinery contains emulsified  material which  is commonly
referred  to as bulk,  sediment,  and water  (BS&W).  The  BS&W  concentration in  the
crude oil ranges  from about 0.01  to 0.1  percent by volume (%v).  This BS&W
fraction  contains about  30%v oil, 50%v  water,  10%v carbonaceous matter, and  in-
organic  salts equivalent to about  10%v.   The  carbonaceous material  which  is the part
of BS&W that could  eventually become oily sludge  for  disposal, amounts to about
0.005 percent  of the incoming  crude oil  to  the  refinery.   However, in the refining
steps, much  of this carbonaceous material is utilized  into  fuels.  The quantity of
waste sludge from  crude  oil is  small  for the following reasons:   1)  Crude  oil
receipts  at a refinery are often stored in  tanks for  a  few days  before being fed to
processing  units.   During quiescent  periods,  most of the BS&W material settles out  as
tank bottoms and  is removed during  tank  cleanings  at three to  five-year intervals.
Fortunately,  tank bottoms have been in  demand  for oil  and wax recoveries by
independent  operators and for road bed additives.  When crude tank bottoms must be
handled  at the refinery, a concentration  step  is possible with  conventional  slop oil

                                       233

-------
 234

demulsifying facilities.   The BS&W  emulsion is  broken and some of the carbonaceous
material is  included  with the recovered oil  phase and  some with the water phase.   The
recovered oil phase  is fed to a refining  unit and the  water phase  flows into an oil-
water separator.  2)  In the event a  settling  time is  not possible, the crude oil con-
taining  BS&W material is fed to a crude distilling unit.   The BS&W material  is
removed in  the  pretreatment (heater and desalter) section  of  the distilling  unit.  The
oil  fraction of the BS&W remains in  the crude.   Water fraction  in  the desalter
effluent flows into the feedwater stream  of  an  oil-water separator.   Emulsified
materials in  the  oil-water separator will either  enter  the recovered oil stream,  settle
with the bottom  sludge  inside the separator  box,  or remain in the  water phase flowing
to secondary treating facilities.

      Ship ballast water-handling facilities  vary at  different  refineries  as  does  the
quantity and quality of both the water  and  oil  phases  of the incoming ballast water.
Ballast  water may be fresh  water,  salt water, or a  mixture of the  two.  Salt water is
usually  clean with respect to suspended solids,  whereas fresh water of  the  surface
category, such  as that obtained  from  a  large  river, may contain large amounts  of silt
that will become deposited in  the tanks  of  vessels.   In addition to silt,  ballast water
may contain finely  divided rust and precipitated chemicals used  in  tank cleaning.
The  oil  contained in  ballast water will  range from gasoline to  heavy cracked fuel oils.

      Ballast water  that  is discharged  from gasoline  or light-distillate oil  tanks  will
contain  little oil until almost all the  water has  been  pumped out.   On the other hand,
ballast  water that is discharged from a  tank that  contained heavy  fuel  oil  may  have
oil  dispersed throughout  the  entire volume.   This is also true of the emulsion from
tank cleaning.

      Tank  bottoms  obtained during  cleaning of tanks  vary  in composition  and the
residual oil content  can  usually  be  recovered or at  least reduced with emulsion
breaking facilities.    However, the real bottoms  and final  washings  from a  tank  are
often oily sludges not suitable for feed  to a demulsified unit and are potential  feed
for  the  soil cultivation  process.

      Oily  sludge from  oil-water separator and  holding pond  cleanings is  usually low
in oil  content.   The  sludge  cleanings from  the  water  box or pond  usually  contains
about two percent carbonaceous  material.   Oil  box cleanings which are  mostly oil
can  be  fed  to the demulsifying  unit,  but sometimes  mixtures  of  oil,  straw,  grass, and
dirt  make this material  suitable only for disposal.

      Solids  and oily waste at the demulsifying unit usually accumulate in tanks which
must be cleaned occasionally,  and these cleanings cannot  be further improved  with
additional  treatment  in  demulsification facilities.

      Process  unit shutdowns include  complete cleaning ofa piece of equipment  and/or
vessel before  maintenance  work begins.  In the  shutting down of a process unit, oil
is returned  to storage.   Immediately after shutdown  and where  possible, the  equipment
is washed with  water which  removes residual traces of oil  and  carbonaceous material.
This  material  is  sent to an  oil sewer which  connects to an oil-water separator.

-------
                                                                                  235
         Many refineries will  require additional  oil and suspended solids removal by
pretreatment  processes in order to meet process requirements of subsequent  treatment
stages.   The quantity of additional oily sludge produced  can be estimated  based on
material  balances for the  suspended  solids removal process such as air  flotation.   The
additional oil solids  production as shown  in  Table I  was estimated based upon  median
TSS removals  by the  air flotation  process at median  flow rates.   Median suspended
solids removal for all classes of refineries was indicated to be  15.3 pounds per thousand
barrels of crude.  The total oily sludge production was estimated based on  the sum  of
the oily  sludge  production without additional oil and TSS  removal equipment and  the
oily sludge production anticipated for dissolved  air flotation.   Total oily sludge pro-
duction as a function or refinery size is  indicated in Figure I.

         Stating  the  sludge production in another fashion, it can be expected  that the
waste sludge from API separators  plus the  usual  secondary treatment  facilities  will con-
tain 1,500-6,000 pounds of dry solids per million gallons of effluent treated.  The  wet
sludge composition will  be approximately  1-3  wt.%  dry solids, 10-30  wt.  %  oil, and
65-90 wt. % water.

         On  an average sludge composition  basis of  2 wt. % dry solids, the flow or
accumulation  of  wet  sludge will  therefore by 75,000-300,000  pounds per million
gallons of effluent,  or about 0.9-3.6 vol. % of the effluent (Rabb  states the  range  to
be  1.5-4.0 vol. % of the effluent).

         Weston,  Merman, and De Mann report  1,500 pounds  of  dry solids  per million
gallons of effluent from API separation plus  chemical coagulation.   They also  report a
range of  120-710 for API separation  along.

         For  an existing refinery, the quantities and  compositions of oily sludge are
attainable by records and measurement.   For the design of a new grass roots facility,
these items must be estimated.  As  long  as  reasonable allowances are  made for con-
servative design,  the exact compositions  are not very critical.   It is highly unlikely
that any such estimate would  be  duplicated  precisely in real  life.

         The compositions shown  in Tables II-V  are from  the API Manual on Dispoal of
Refinery  Wastes, Volume VI,  Solid Wastes,  1963.

         There are other oily  sludges generated  in refineries which  may be  classified  as
oily but  usually  require special  consideration in handling and  disposal.  Typical among
these are sludges generated in  Tetraethyl  Lead  Gasoline Blending Operations and
Hydrofluoric  Acid Alkylation Processing.   Oily  sludges generated from these and other
exotic operations must of  necessity be segregated and handled  separately because  of
their  toxicity and other environmental hazards.

TRANSPORT  AND HANDLING

         Precaution should be  taken  in any  wastewater collection and  treatment plant
design to minimize pumping where both  oil  and  water are present together.  One of
the most  difficult processes to run successfully is the breaking  of oil-water emulsions.

-------
236

 Care should be taken  during the design phase  of  a project to minimize pumping  before
 the  one constituent (oil  or  water)  is removed by gravity or other  separation.

          If at all  feasible,  process wastewaters containing  oil should  drain by gravity
 into  the API or other  type  oil-water separators.   This  will preclude the  possibility of
 creating excessive amounts  of  emulsion.   Oily water,  if it must be pumped at all,
 should  be pumped utilizing  screw type lift pumps  or low RPM open-impeller centri-
 fugal pumps.  Any mixing of chemicals with oily water  should  be accomplished using
 the  lowest RPM mixing speed which will  perform  the task.  An example  of this con-
 cept is the  chemical  addition  of alum  and/or polymer  before dissolved air flotation.

         Oily sludges  from  separators,  tanks  and other sources are frequently quite
 viscous and can contain sand, gravel and  other large particles.   Screw pumps or
 positive displacement pumps are  to be  preferred for this  type of service.

 PROCESSING METHODOLOGY

         Oily sludge  processing techniques include  emulsion breaking,  gravity, separa-
 tion, centrifugation,  filtration, and incineration as shown  in Figure 2.

         Emulsion  Breaking

         Refinery emulsions  may  be defined as  intimate,  two-phase mixtures of oil and
 water with one phase  dispersed as  minute globules in  the other  phase.  The minute
 globules are stabilized by an interfacial  film or stabilizing agent  such that the
 globules do not coalesce and do not respond to gravity settling.

         Emulsions usually are  complex,  and emulsion technology  is,  at best,  only
 quasi-scientific.   Generalizations  are risky,  and  trial-and-error experimentation is
 essential.   For a systematic approach to  emulsion  problems, emulsions may be
 characterized  by:

         a.   Dispersed phase,  as oil-in water  (O-W) or  water-in-oil  (W-O).
         b.   Dual O-W  and W-O  emulsions.
         c.   Stability,  as measured by change with time of the dispersed phase
              particle size analysis  and/or  a  time-separation study.
         d.   Type of interfacial  film:
              Anionic - acid stable or  acid sensitive.
              Cationic  - alkali stable or  alkali sensitive.
              Nonionic - hot-alkali stable or hot-alkali sensitive.
         e.   Degree of success in  emulsion breaking according  to process, e.g.,
              physical methods  - heating,  distillation,  centrifuging, or filtration.

         Improper treatment of either O-W or W-O emulsions may invert the  emulsion
 to the opposite type.

         The most satisfactory  theory of emulsions,  which can also account for dual
 emulsions,  inversion  of phases,  and effects of  a wide range of  emulsifiying agents,

-------
                                                                                  237

is  one which  emphasizes the nature of the interface between  oil and  water.  An
interfacial film,  which  is a  complex mixture  of dissolved and colloidal matter and
suspended solids,  generally  is understood to exist.  Such an interfacial film may be
highly viscous,  gelatinous,  and/or possess an  electrical charge.  Some experimenters
consider the interfacial  film as  a third  phase.  Generally,  the  stability of  an emulsion
is  not a function  of the degree of dispersion; rather, it  is a function of the stabilizing
agents and  properties of the interfacial  film,  such as film viscosity and  electrical
charges.  This is  illustrated  by  the following  major factors  which affect emulsion
stability - four out of five  of these  factors being directly related  to the interfacial
film:

         a.  A coherent, mechanically strong interfacial  film.
         b.  An oriented unimolecular layer of emulsifying agent or tightly
             packed solid particles at  the  interface.
         c.  Viscosity of the continuous phase and  of the emulsion.
         d.   Nature and amount of emulsifying agents.
         e.  Physical storage or historical conditions, such  as temperature,
             agitation,  and  dilution.

         Properties  of emulsions of refinery waste  control  are  best  approached by
understanding what  generally contributes to the stability of an emulsion and,  therefore,
what  can be done to cause an  emulsion  to become  unstable and separate or "break".
Treating from this point of  view requires an  understanding of  causative  emulsifying and
stabilizing agents.

         Substances which cause or stabilize emulsions may be classified as  ionic or
nonionic and,  generally,  are of a colloidal nature  or consist of solid  particles which
are surface active.   Hydrated  gels or gums,  soaps,  sulfonated oils, asphaltic residues,
waxes, silt, finely  divided  coke, and  sanitary sewage solids typify the range of
emulsifying agents encountered.   Colloidal matter may be further classified  as hydro-
philic (waterside  active) or hydrophobic  (oilside active).

         The  important  ionic emulsifiers are chemically active because of the active
carboxyl, sulfonic,  amine,  of  hydroxyl  groups.  Such compounds,  singly or  in com-
bination with electrolytes,  may readily result in a  stable emulsion.  When  there is no
chemical  reaction between  the  electrolyte  and emulsifying agent,  the  electrical  charge
that may exist on the dispersed globules in an emulsion  has  little  significance in
regard to stability or to type of emulsion.

         The following  rules generally  apply for  emulsion formation:

         a.  O-W  emulsions are formed  when soaps are  colloidal in the water
             phase  (hydrophilic).
         b.  W-O  emulsions are formed  when soaps are  salted or precipitated
             from the aqueous phase.
         c.  Soaps  partially salted out  may form  and stabilize either  type  emulsion.
         d.  Electrolytes favoring O-W emulsions are salts  of univalent  cations and
             sdlts of di- and trivalent onions.

-------
238

         e.   Electrolytes  favoring W-O emulsions are salts of di- and  trivalent cations.

         One approach  to emulsion  breaking is  to counter-balance emulsifying  agents of
one  type of emulsion (W-O or O-W) with one  of the opposite type.

         Methods for emulsion  breaking can be  grouped  into  physical,  electrical  and
chemical.   Each  is briefly described below.  Frequently, the system used in any
particular  refinery will  include more than one  combination of these methods.

         Physical Methods
              Heating
              The process  of heating  emulsions to obtain a break  is the simplest and
most  economical approach available; the use of heat is  nearly always advantageous.
Heat markedly reduces  the viscosity  of the  oil  phase, melts  interfacial waxy films,
and,  in  some instances, weakens the interfacial  films of various  emulsifying agents.
Coalescence and  separation of the  oil  and water phases can  then take  place.

         For W-O emulsions,  heat  increases the  vapor pressure of the water which
tends to rupture the films around the globules.   The formation of steam within  the
globule  generally will  break a W-O emulsion,  but  the  rate  of steam generation must
be controlled to avoid  excessive foaming.

         A heat-and-pressure system  may be used to break many  types  of emulsions.
The  emulsion  is  heated  in a suitable pressure vessel wherein  water and sediment are
precipitated.   Adjustment of the pH  by addition of  acustic or acid may help facilitate
the breaking  of  an  emulsion with  heat.  There  appears  to be no  general rule to  serve
as a  guide in determining optimum pH.  This  is  a situation  where trial  by  experiment
is the best solution.

         Distillation
         Distillation methods for breaking emulsions  are  effective  in that they involve
the use  of  heat  as  described previously.   In addition, water and  light  ends of  the oil
are removed as  overhead  products,  and the  emulsifying  agent remains with  the  residue.

         Centrifuging
         Comparatively  stable  mixtures sometimes may be separated by  centrifuging.
If there  is a  difference in specific gravity between  the  oil and  water,  centrifugal
force may  be used  to accelerate separation.  Loosely occluded water  is readily
separated from oil or emulsion  by centrifuging.

         Precoat Filtration
         Stable  W-O emulsions, particularly those stabilized  by  finally divided solids,
can  be  resolved  by filtration.  Rotary  vacuum  precoat filters are  commonly  used  for
this  purpose.   In this operation,  the emulsion is filtered through  a layer of  diatomaceous
earth deposited  on  a continuously rotating drum filter.   The  suspended  solid material
and as a result  of removing the stabilizing  of the precoat layer  during that portion of
the cycle  in  which the  drum is submerged  in the material being  filtered.   During  the

-------
                                                                                   239
remainder of the cycle  the  filter  cake and  precoat  are dried.   A knife edge advances
automatically into  the precoat layer to a predetermined depth  for each revolution  of
the filter drum, thereby removing the  sediment  deposited  on the precoat layer along
with a very  thin layer of the precoat  material.

         The breaking of eumlsions occurs as a  result  of rupturing the globules in the
dispersed  phase when  passing  through  the interstices of the  filter cake and precoat
material  and as a  result of  removing the stabilizing solids.   The oil and water phases
of the filtrate  readily separate by gravity difference.  These phases may still contain
dissolved  emulsifiers.   If so,  care should be taken to  prevent the re-formation of the
emulsions by excessive turbulence or mixing before  separation of the oil and water
layers.

         Electrical Methods
         Some  emulsions can be broken under the influence of a strong electrical field
by application of the principle  of the Cottrell  precipitator.  The emulsion is passed
between  two electrodes and subjected  to a  high-potential pulsating  current.   The water
globules  are attracted electrically and coalesce until  the  masses are of sufficient size
to settle  by  gravity.  This method is  used  in the  dehydration and desalting of crude
oil.

         Where electrical desalting is  used  for  treating crude oil feeds, some emulsions
may be blended along with the regular crude oil  charge.   Emulsions can  sometimes be
blended  at  rates ranging from 1 to  10 percent of  the  crude oil  charge if  the water and
sediment  contents of  the combined mixture  are not too high.  Before attempting to
dispose  of emulsions  through a crude oil desalter, consideration should be  given  to the
possibility of cracked material in  the  slop  oil throwing virgin products from the  crude
oil unit  off  specifications.   Equally important for consideration  are  emulsifiers in the
slop oil  that might throw the oil  or water from  the  desalter off specification.

         A  combination of electrical and chemical methods  is frequently successful
where neither  method alone  is effective.

         Chemical  Methods
         Emulsions  can  be broken  by chemicals  which  will balance  or  reverse the  inter-
facial surface  tension on each side of the  interfacial  film,  neutralize  stabilizing
electrical charges, or precipitate  emulsifying agents.

         Anionic and  cationic surf ace-active agents are not compatible and will tend
to neutralize each other.   Rules for breaking emulsions stablizied with electrolytes are:
                                                     -2
         a.  Reactive onions, such as OH   and PO,   ,  will break W-O emulsions.

         b.  Reactive cations, such as H ,  Al    , and Fe  , will   break W-O
             emulsions.

         The use of a heavy metal salt to  form  a flocculent precipitate of a  heavy
metal hydroxide to break dilute O-W  emulsions is typified  by  the  flocculation
process.   The  resulting  clarification or emulstion  breaking is believed  to  result from

-------
 240


destruction  of  the  emulsifying agents by  neutralizing or reversing interfacial-film
electrical charges  and, in turn,  by agglomeration  of the  precipitated floe and ad-
sorption  of  suspended oil  matter  on the floe  surface.

         Any wetting agent will  break a W-O emulsion provided the agent is
effectively  surface-active on the waterside of  the  interfacial  film and a certain
critical amount is  not  exceeded.   An  example of the use of a  surface-active chemical
to break an emulsion is the use  of  a  detergent to  break an asphaltic, solids-stabilized
W-O emulsion by  causing water  to wet the  solid preferentially - thereby displacing
the  loosely  adsorbed asphaltic  emulsifying agent.

         Effective  organic chemicals for  breaking W-O  emulsions generally  have either
an  inorganic group (phosphate) or a reactive group (such  as OH  or  -NhL) and  have
at least  three  focuses of  polarity occurring in  different  terminal  groups.

         Organic chemicals used in breaking W-O emulsions should:
         a.  Be  soluble in both  phases,  i.e.,  the treating chemical contains  both
             hydrophilic  and hydrophobic groups.
         b.  Form  an adsorption  complex with  the predominantly  hydrophobic
             emulsifying  agent present thereby creating a hydrophilic complex that
             can be wetted by water.
         c.  Contain one or more radicals able  to react with calcium (Ca)  or
             magnesium (Mg)  salts  in  the aqueous  phase and result  in insoluble salts
             that  are wetted by  water.

         In  general, the  dewatering of recovered oil or breaking of recovered oil
emulsions in refinery practice is  a series of  batch  operations involving several stages
of water and solids removal.   The simplest system  consists of  holding tanks  operated
on  a batch  basis,  usually with the  addition  of heat and use  of recirculation.   How-
ever, both  cold  and hot  settling are used, with  cold  settling  time varying from one
day to four weeks, and hot settling time usually varying  from  one to three  days at
temperatures in the range of 150 F to 200 F.

         At the  end of each stage of  treatment,  substantially  oil-free water is drawn
off and returned to the refinery  separators and low-water-content oil is decanted  and
returned  to  the refinery for reprocessing*   In most cases,  a layer of emulsion remains
after these  separations.   This  emulsion is generally given  further treatment.  The
methods  used  vary  widely and  are  usually adaptations of the mthods described  earlier
in this chapter.

         The dewatered oil may  be charged  to crude oil units,  with or without
desalters; to thermal or catalytic cracking units; to coking units;  to visebreakers;  or
to separate  distillation units.   Water  content of  recovered oils  fed  to crude oil units
or separate  distillation units may run  as  high as  2 percent, but the  water content of
recovered oil fed to the other types of units is generally limited to 0.3 to 005 per-
cent.

-------
                                                                                  241
         The following figures  3 and 4  illustrate a rather simple continuous process  for
breaking emulsions emanating from storage tanks and a rather sophisticated process for
de-emulsifying all  refinery^wide sources.

         Gravity Sedimentation and Phase  Separation
         The processes of gravity  sedimentation  and phase separation combine  the
effects of specific  gravity,  viscosity,  and  time  to  provide for the separation of sub-
stances  of  different densities.   Conical  bottom vessels  are utilized for most applica-
tions, especially where a continuous drawoff or a  concentrating  effect is  desired.

         Centrifugation
         The use of Centrifugation permits  the mechanical dewatering  of sludges through
centrifugal  force.   Dewatering  by this method  may be defined as sedimentation under
the influence of forces  greater  than gravity.  Within  a centrifuge,  centrifugal force
acts on a suspended particle in the sludge, causing it to settle through the liquid.
It would seem  logical to  design machines for maximum centrifugal force,  but  this  is
not the only consideration.   While additional centrifugal force  helps  settle solids, it
also increases the  difficulty of  discharging the settled  solids.   Thus, for maximum
efficiency,  a compromise must be  made.

         Several variables which affect  gravity  sedimentation  also effect sedimentation
within centrifuges; these variables are:

         1 .  Particle  size and shape,
         2.  Density differential between solid  particles  and the liquid,
         3.  Concentration  of the particles,  and,
         4.  Liquid viscosity.

         Sludge solids in  suspension are a  combination of granular,  fibrous and
flocculent  particles.  Granular particles in dilute  solution  are generally considered  to
settle at terminal velocity,  independent of one  another,  and  without  change in size
or shape.   When  the  particles become concentrated however, volumetric considerations
predominate and the observed settling velocity appraoches zero.   Flocculent and
fibrous  particles do not settle  at  a constant velocity since they  tend  to cluster during
sedimentation, bringing about a continuous change in size,  shape, and relative
density.   Extremely fine particles  which may  not  settle under normal  gravity separa-
tion can be made  to settle  by increased gravitational  forces.   In general  those sludges
which spearate  most readily and concentrate to  the greatest degree  by sedimentation
are also those  which are  dewatered most  efficiently by centrifuges.

         Two of the major types of centrifuges in industrial  use today  are the  solid
bowl and the basket type.   These two types may  be used singly or in  combination to
remove  solids from solution.  When used  together, flow  is first  directed to  the solid
bowl centrifuge where the  large solids are removed and  then  to  the basket  type
centrifuge  where finer solids are removed and emulsions  broken  due to the higher
centrifugal  forces in this  unit.  The solid  bowl  sludge dewatering centrifuge assembly
consists of an outer horizontal cone (bowl) and  an inner screw  conveyer,  each
rotating  at  slightly different speeds.  Sludge is  fed to the centrifuge  through  the
central  screw and  enters the bowl  about halfway along the  side  of the cone.

-------
 242

Centrifugal action  forces both  liquid and solid phases to the  walls  of  the  cone and
down the cone  to  the  large end,  where water in the top  layer of the liquid pool  is
drawn off.   The solids,  which  accumulate closer to  the wall,  are conveyed by the
screw  to the small end of the  cone where they are discharged.   Turbulence due to
the screw,  limits  the  solids  concentration  obtained,  and maximum solids  concentrations
from  this type  centrifuge are normally  in the  range of  5-10 percent.

         As opposed to the  continuous  operation of solid bowl centrifuges,  basket type
centrifuges  operate on a  timed cycle.   The  units resemble  the  rotating tub of a spin
dryer used in home laundering  with the exception that  the  tub  has  a hopper at the
bottom.   Slurry is  fed to  the  tub  (basket) during the  first portion  of the  cycle,  and
the spin of the basket creates  centrifugal  action up  to  13,000  times the force of
gravity which forces liquid  and solids against  the  outer walls.  Part of the  liquid
travels up the  wall and solids  layer and out over the top lip of the basket and  into
a  discharge trough.  According to a timed cycle, remaining  liquid is  then  drawn off
by a pipe which slowly moves toward the wall from  the center of the basket.  AS
the pipe approaches the wall,  it  begins picking up a concentrated solids slurry, and
a  valving  change routes these  to  the bottom hopper.   A new cycle then begins.
After several such  cycles, a hard solids layer accumulates  near the wall and this
layer is  then removed with  a  knife edge at  a  slower rpm and these solids  fall to the
bottom  of  the hopper  where they  combine with the other concentrated slurry and are
discharged.  Using this type of centrifuge, overall sludge concentrations of 8-15
percent are possible.

         The introduction  of flocculation aids, such  as  polymers,  has increased  the
range  of materials  that can  be dewatered  satisfactorily  by centrifuges.   The degree of
solids  removed  can be regulated over a wide  range depending on the  amount of
chemical  coagulant used.   Wetter sludges result from chemical  aids because of the
increased capture of fines.

         Rotary Drum  Vacuum  Filtration
         Vacuum filters are  the most widely used type  of mechanical sludge dewatering
device.   The filtration process is  used  to  separate the  solids  from  the  liquid by  means
of a porous media  which retains the  solids but allows the liquid to pass through.
Filter  media employed include Nylon,  Dacron,  and  Polyethylene  cloth;  steel mesh; or
tightly wound steel coils.   The types  of rotary drum  vacuum  filters currently in  use
differ  primarily in the way  cake  is removed from  the drum.  The filtration  process  is
accomplished by means of a horizontal drum covered  with the filter media.   The drum
is  rotated  in a  tank with  approximately one quarter  or  more of the drum submerged in
wet  sludge.  Valves  and piping are arranged  so that as the drum rotates a vacuum
applied on the  inner side  of the filter media  draws water from  the sludge  and holds
the layer of sludge to the drum.   The  application of a vacuum is continued as the
drum  rotates out of the slurry.  This pulls additional  water from the sludge,  leaving a
moist cake of sludge on the outer surface which is removed prior to re-entering the
tank.  Cake is  removed by  a  knife that moves on the  filter drum  as it  rotates,

         Precoats and  filter  aids are  often used  to speed filtration rates  or collect more
of the fine  particles  in the  slurry.  The precoat,  usually diatomaceous earth,  is a

-------
                                                                                   243

finely divided, hard-structured  solid that forms an open,  noncompressible cake.
When  applied  on  the filter cloth,  the precoat acts as the primary filter medium and
permits high removal  of  fine particles  from the slurry.  A filter aid may also  be mixed
with  the slurry where it becomes distributed throughout the cake,  thereby keeping the
cake  relatively open  for flow and continuously supplying a large surface for adhesion
of finely divided  solids.   This method is particularly good when filtering colloidal
solids that are  not  to be recovered from  the  cake.

         Filter  media may consist of cloth, paper,  or woven  or porous  metal  as  men-
tioned before.  Criteria on which a filter medium is selected must include the ability
to remove the solid phase, high liquid throughput for a given pressure  drop,
mechanical strength,  and  chemical  inertness to the slurry or any wash fluids.   In
operation, some filter cake solids  usually penetrate  the filter  medium and fill  some of
the pores.   In  normal cases,  five  to  25 percent of the pore volume of  the filter
medium  if filled with solids.   As a result, resistance  to flow  through the medium
increases greatly  - in some cases to such  an  extent that  the  filtration rate is
seriously reduced.

         Properties  which  affect the vacuum filtration dewatering process include:

         1 .   Solids feed  concentration,
         2.   Sludge and  filtrate viscosity,
         3.   Sludge compressibility,
         4.   Chemical  and physical characteristics of the sludge,
         5.   Operating vacuum,
         6.   Drum  rotational  speed,
         7.   Drum  submergence,
         8.   Type  and  porosity of the filter media,  and
         9-   Effect of chemical additions.

         Pressure  Filtration
         The filter  press is the  simplest  of all pressure  filters with the most widely
used  type being the plate-and-frame  press.   As  the name implies,  the  plate-and-
frame is an  assembly of alternate solid plates, the faces  of which are grooved to
permit drainage.   Between these  plates  are hollow spaces in  which the  cake  collects
during filtration.   A  filter medium,  usually some sort of  a fabric,  covers both faces
of each  plate.   As filtration  proceeds,  cake  builds  up on the filter cloths  until the
cakes being formed on  each face of the frame meet  in the center.  When  this
happens, the flow of filtrate  which has  been decreasing  continuously as the cake
thickness increases, drops  off rapidly.   Usually,  filtration is  stopped  well before this
occurs.   After  the  filtration sequence, the press  is disassembled and the solids are
collected either manually or mechanically and discarded.

         Filter  presses can be built for high pressures  and can handle the  filtration of
either heavy or light slurries.   The filter press has many advantages, the  greatest of
which are its simplicity,  flexibility,  and ability  to  operate at high pressures.  A
filter press,  with  proper operation, will produce  a denser,  drier cake (up to  60 per-
cent  solids)  than  any other type of filter.  The  sludge and filtrate properties  which

-------
244

 affect the filter press dewatering process  are  the  same as those for vacuum  dewatering.
 The operation of separating waste  oil from oily emulsion by  means of a filter press
 opearated at elevated  temperatures is being practiced.   A process in which oily
 sludges are  subjected to high pressure filtration enabled  the  oil to be  removed  in the
 form  of a separable  phase.  This process  produced a cake  with a high concentration of
 solids and low  concentration of  oil which would be  suitable for land farming.   The
 cake produced  by high pressure  filtration  contained very little volatile material which
 could be removed by incineration.  Significant oil  recovery in the process  reduces
 the overall  cost of the disposal.   The economic incentive  of recovering oil from oily
 sludges will  increase as crude  oil  supplies become less available.

         Incineration
         A properly designed and  controlled incinerator  is satisfactory for burning com-
 bustible waste provided air pollution standards can be  met.  A high control tempera-
 ture  is needed  for efficiency,  prevention  of  excess  air pollution,  and odor  control.
 Incinerators  are readily adaptable  to the  disposal  of virtually all  oxidizable wastes,
 including streams which contain  inorganic or  other non-combustible materials.   Com-
 bustible wastes which contain  10,000 BTU per pound or  more can  usually be incin-
 erated without  the  requirement of  an auxiliary  combustion  chamber.   In some
 instances, where the heat value is less  than  10,000 BTU per pound, it is possible  to
 enrich the liquid waste with another fuel.  Incineration  is being  used  at  many  large
 plants where final disposal  of  the  solid  waste is a problem.   It has  the advantage  of
 freedom  from most odors, independence  of weather conditions, and  high reduction of
 the volume  and weight of the  end product to be  disposed of.   There is a minimum
 size  of treatment plant below which incineration  is not considered economical  since
 there must be enough waste solids  to warrant  reasonable  use  of the expensive  equip-
 ment involved.

 CASE HISTORY

         A major Gulf Coast Refinery is  presently in the midst of a major process ex-
 pansion program to essentially  double their throughput  and  increase their ability and
 flexibility in processing a wider slate of  crude  oil,  including  high  sulfur  mid-eastern
 crudes.   New facilities to  be  installed  include Crude  and  Vacuum  unit with Electro-
 static Desalting,  FCC,  Hydrotreater,  Gas Plant,  Product Treating Plants  and Sulfur
 Plant.   Inherent  in the design of  new,  additional processing facilities  was an
 examination  and  evaluation of existing  offsite facilities such as raw  water and waste
 treating,  steam generation and tankage.   In   conjunction with  a consultant, a major
 program  for  modification of and  addition  to the waste  treatment facilities  was under-
 taken in  order  to assure compliance  with  both State  and Federal discharge require-
 ments.   Concurrent with and as  a  part  of this operation,  an evaluation of the  "oily
 sludge" sector of the plant operation was also performed.  The following  describes
 the  history of this evaluation.

         Old  Facilities (Figure 5)
        All oily wastes  plus cooling tower blowdown  water,  boiler blowdown water,
 lime  softener blowdown, and zeolite regenerant water  were  discharged to the oily
 sewer.   The oily waste sewer discharged  to an  API Separator where  gravity separation

-------
                                                                                 245
of free  oil and  water was accomplished,  followed by an air flotation unit,  employing
a cationic polymer as the sole chemical  feed.   The API effluent was sprayed to the
atmosphere to accomplish  cooling  and was collected in a basin and discharged to two
aeration basins  connected  in series.  Aeration was  supplied by a 15 hp aerator in each
basin.  The waste  was discharged to a settling basis and  thence to a lagoon.   The
lagoon  discharged through a  measurement weir to the ship channel.

        All storm waters were collected and  discharged  to the  second  aeration  basin
for treatment  prior to discharge to the lagoon.

        Ballast water was pumped  to a ballast  water tank  where oil was separated by
gravity  and discharged to  ballast  water holding  basins  and thence  to the second
aeration basin for further  treatment  prior to  discharge  to the lagoon.

        Recovered oil  from the API separator was sent  to an  emulsion  breaking process
and  the resultant sludge from recovery was discharged  to  a  sludge pond.  The float
from  the  air flotation unit consisting of sludge and some oil was also discharged to
the sludge pond.  Efforts  have been  made to recover this oil  in the emulsion breaking
process  but the inclusion of  the air float sludge rendered the  process  inoperable,  pro-
bably due to  the presence of calcium and magnesium salts in  the air  float sludge.

        The sludge ponds did not dewater readily and removal  and subsequent land
disposal of the  sludge  had become a problem.

        Due to  the inclusion  of highly alkaline boiler  blowdown,  filter  backwash
water,  lime softener blowdown and  softener  regenerant waste,  the pH of the API
influent averaged 9.0.  The average total steam load  was 304,000 Ibs/hr and at
10%  blowdown, the  discharge,  at an average pH of 11.5,  to  the oily  water sewer
was  60  gpm,  or 17% of the  average volume input  to the API  Separator.  In addition,
the discharges from the lime softener, filter backwash, zeolite regenerant wastes  as
well  as some  spent  caustics from the refinery process contributed to the high pH.

        Since  it  is well known that  high  pH conditions tend  to cause  oil/water
emulsions, the  efficiency  of gravity separation of free oil  at the API Separator was
impaired,  reducing  the quantity of recoverable oil  at  that point.

        The inclusion of boiler water blowdown in the  oily  waste stream produced a
softening  effect within the oily waste system,  resulting in a  rather heavy precipitate.
The  average  blowdown flow  of 60 gpm or 715,000 Ibs.  per day was the equivalent
to 114  Ibs. of  soda ash,  175 Ibs. of caustic,  and  44  Ibs. of  tri-sodium phosphate
per day.   The  precipitate  formed  was removed as float in the  air flotation  unit con-
tributing  to the  excessive  sludge problem at this point, and  prohibiting further
recovery  of oil.  In addition, the oil adsorbed  on  the floe particles carrying over
from  the  air flotation unit contributed to the oil content of the final effluent  as  did
the emulsified oil due to  the pH.

        The emulsion breaking process operated very satisfactorily upon  recovered  oil
from  the  API  separator but could  not handle the float  from  the air flotation unit  due
to precipitated  calcium and  magnesium salts.

-------
246

         Sludge from the emulsion breaking process  and the float was discharged  to the
  sludge  ponds  where after a  period of dewatering it was mechanically removed and
  sent to land disposal.   The  sludge  was very difficult to dewater due  primarily to the
  air float  sludge and the amount of rainfall and presented a difficult problem.

  INITIAL  RECOMMENDATIONS  AND SYSTEM UPGRADING

         The most economical  and feasible point to  accomplish more  efficient  oil
  removal and recovery was at the API  separator.  This could be accomplished by
  diverting high alkaline  wastes from the oily waste stream to lower the pH of the
 stream entering  the API separator to prevent emulsification of oil  and achieve more
  efficient  gravity separation  of oil.

         The diversion of boiler blowdown alone, which  represented 17% of the flow
  volume to the API  separator,  would increase the  retention time considerably to  ac-
  complish  better gravity separation of oil.

         A second beneficial  effect of diverting alkaline waste streams was the elimina-
  tion of precipitated calcium  and magnesium  salts which constitute the  greatest part
  of  the  sludge volume discharged to the sludge ponds  by way of float from the air
  flotation  unit.  Thirdly, the float, with  the  absence of precipitated matter,  could
  be  discharged to the emulsion breaking process for additional oil  recovery.  Of
  course, a prime benefit of diversion of the alkaline wastes would be the  reduction
  of  emulsified oil content of the discharge to final waste  treatment facilities  and
  final outfall to the ship channel.

         The reduction in sludge  volume discharged  to the  sludge  ponds would
  facilitate dewatering and ultimate final  disposal.

         The water treatment  plant blowdown  was diverted  from  the main process
  sewer and pumped   into  a holding pond.  It  was apparent immediately  that oil
 recovery  was  more   efficient  and sludge and emulsion  formation less  severe.   However,
  due to the  project  timing and financial  considerations,  it was  decided not to try to
  recover oil from either the  API  sludge or DAF  float.  This would be attempted  when
  the  expansion was   complete and more  permanent facilities installed.

  NEW  REFINERY SYSTEM

         Waste Oil
         The refinery has been  treating  waste slop oils and emulsions  in the following
  manner - that is,  place the  slop oils  into 500  BBL tanks, heat,  add  a chemical,
 and  cool.  Emulsions were easily broken  and oil and water were  separated.   The oil
 was  returned to the  crude units  and the water was recycled  to the  waste  treating
 unit.

         This method was very successful excep that during the  hot summer months the
 two  existing 500 BBL tanks  were sometimes overloaded due  to  insufficient cooling.
 The  operating cycle was as  follows:

-------
                                         o                                      247
       1.   Fill tank and  heat  to  130-160 F  - this required 4 to 7 days.
       2.   Cool tank - 4 to 5 days (Atmospheric Cooling.)

       With only two tanks and periods of high oil  loadings,  a tank that  has  not
sufficiently  cooled may  need to be  drained so  that it can be  refilled.   The warmed
drainage was usually not completely demulsified.

       For  the expanded refinery,  the  following  treatment  of  the waste  slop oil  was
proposed.

       1 .   Results of the  tests  by  local waste  water  treatment  chemical  vendors
            indicated that  cold settling of oily water emulsions did not result in
            demulsification.
       2.   Add  four 500  BBL tanks with mixers,  heating/cooling coils and sloped
            bottom for treating  slop oil.   Separated  oil  from these tanks could be
            pumped to existing  slop oil tanks for surging prior to pumping  the  oil
            back to the crude  unit  feed tanks.
       3.   During periods in which  excessive amounts of slop oils were  collected,
            the existing slop oil tanks could  be converted back  into demulsification
            treating service.

       Oil  Sludges
       The  refinery was treating their sludges  in  the following manner.  Both  API
bottom sludges and  air flotation skimmed material  were sent to sludge  pits  where the
oily solids settled.  The oil and water overflowed to second and third pits.  The
water was recycled to the air  flotation unit and  oil routed to slop  oil treating.  The
thickened sludge was pumped to another  pit for drying.   This  method worked satis-
factorily.   However, the  new  waste water treating facilities would be constructed  in
the same area  that  the  existing sludge  pits are now  located,  therefore an  intemnin
and permanent solution  to  the  disposal  of waste solids was  required.

       Two disposal alternatives for  oily solids were  compared  in order to  determine
the most cost effective  method  of waste solids  disposal.   The  projected waste oily
sludge inventory is  shown  in Table  6 and a mass  balance diagram  for oily  solids is
given in Figure 6.

       Alternate No. 1 - Contract Disposal

       The  bottoms from both gravity oil separators would be  combined in  a cone
bottom 500  barrel tank  with the air  flotation skimmings  being  stored in another
identical tank.   The separator  bottoms  will concentrate  to  approximately 15%  solids
and the skimmings  will  concentrate  to approximately  3%.   After emulsion  breaking
decant oil  from the tanks  would be routed to the  slop oil system, oily water to  the
old area API,  and sludge  to contract disposal,,

       Contract  Disposal Site:  A  60 barrel vacuum  truck  would be  needed at
approximately two  week intervals,  to haul  sludge  to  a disposal site.   Since tank
bottoms and oily sediment  will  be removed intermittently, it was assumed  that a HO
barrel vacuum  truck would be  used  for hauling direct to a disposal site.   Current

-------
248

 vacuum truck rates are $17.50 to  $23.00  per  hour,  and 4  hours  per  load  was used to
 estimate hauling costs.  The  estimated  capital cost and  yearly  operating  cost  is in-
 dicated in  Figure  7 based on  the project waste solids  volumes  given  earlier.

         Alternate  No.  2 - Incineration  and Landfill

         The separator bottoms, air  flotation skimmings, oily sediment,  and tank
 bottoms would be  stored in a  1000 barrel  tank before  being charged  to the  incin-
 erator.  Although  variation in the oil content is  expected, an  oil  content of  10%
 was used  for  estimating purposes.  An average solids concentration of 4% was used
 for the estimate.   The cost considerations  of  incineration can significantly influence
 the  processing steps required.   The major  factors  are fuel  costs,  loss  of  oil  recovery
 potential, and the capital cost of the incinerator.   The principal parameters which
 determine  the above costs are solids production rate, water concentration,  and oil
 concentration.  Due to the complexity  of  incinerator design and  the  absence of data,
 the  cost for incineration was  based on  data from  similar systems at other refineries.
 The capital and  operating costs  are shown in  Figure 7.

         On the basis of the preceding discussion and preliminary  cost estimates the
 most cost  effective temporary  solution to solid waste disposal  at the refinery is to
 continue  the  practice  of  contract disposal  of  oily solids  following thickening.   It
 must be recognized that the proposed system is not the ultimate solution to  the
 refinery's  waste solids problems.   However, the  proposed system is  a  prerequisite  for
 most future ultimate solids dispoal  systems.

         The future  of contract disposal sites is  presently  in doubt  as is future regula-
 tions governing landfill and  land farming operations.   In order  to determine the
 magnitude of the waste solids disposal problem  at the  refinery,  the suggestion  is
 made that the refinery initiate an  investigation to determine the  feasibility  of com-
 bining  the  oily and biological solids  and either incinerating the  wastes or dewatering
 them prior  to  land farming or contract disposal.   Also to be considered and, in
 fact, in progress  is a  program to reclaim oil  from the sludge  prior  to dewatering
 and ultimate  disposal.  The study  should commence after the  expanded refinery is in
 operation and should  encompass  all solid wastes  generated within the refinery.   This
 recommendation is made on the  basis  of limited  available  land  area.  The fact that
 contract disposal  costs will continue to  increase must be recognized.  It is  suggested
 that it  would be  in  the refinery's  best  interest to minimize dependence on off-site
 disposal facilities.

 BIBLIOGRAPHY

 (1)  Manual on Disposal of Refinery Wastes,  Volume On Liquid Wastes, American
      Petroleum Institute,  1969
 (2)  Manual on Disposal of Refinery Wastes,  Volume on Solid Wastes, American
      Petroleum Institute,  1963
 (3)  Economics of  Refinery Wastewater  Treatment, American Petroleum Institute, 1973
 (4)  Aqueous  Wastes,  Milton  R. Beychok,  John Wiley & Sons,  1967

-------
                                                                                249
 (5)  "Sludge Disposal:  A Growing Problem",  Petroleum  Refiner, April, 1965
 (6)  "Waste Disposal  Problems Of  The Petroleum Industry",  Industrial Wastes,
     American  Chemical  Society Monograph 118, Reinhold Publishing Corp., 1953
 (7)  Residuals Management in  Industry;  A Case Study of  Petroleum  Refining,
     Clifford S. Russell,  Johns Hopkins University  Press,  1973
 (8)  Oily Waste Disposal  By  Soil  Cultivation Process,  C. Buford Kincannon, EPA
     Report  *R2-72-110,  1972
 (9)  "Land Disposal Of Oily And  Biological Sludge  -  Some Site Considerations",
     Ro  C.  Chatham  and  J. A. Scher,  AlChE 68th Annual  Meeting, 1975
(10)  "Industrial Wastewater Pumps",  J.  A.  Scher,  Chemical  Engineering  Deskbook
     Issue, October 6, 1975
(11)  Private Communications to J. A. Scher From  Client and  Consultant,  1974-5.

 DISCUSSION

 Morris Wiley:  You mentioned  the problem  of sludges  from ballast  reception.   The
 1973  IMCO convention is ultimately  going  to require  shore reception facilities for
 all  kinds  of wastes from  ships and from tankers in  particular.   There are really about
 four major categories  of  wastes.   One  of these is  clean  ballast from tanks that have
 been  cleaned  at sea  with the Load-on-Top  procedure,  one is  dirty ballast  from tanks
 which haven't  been cleaned,  and  one is  the slop oil residue which is  carried  on
 board from  Load-on-Top,  and another is that some refineries accept  the total tank
 washings without any  overboard discharge from  the vessel  while at  sea»  But then  the
 most troublesome kind of waste arises when the ship  has  to  go into ship repair  port.
 They  have to get in and muck out  the bottom of the  tank  and for a very large  tanker
 this can run at a  huge volume of  unpumpable tank cleanings.   Do you have a
 recommendation as to what to do  with  that type of sludge?

 J. A. Scher:   That is a  difficult  question for me to answer.   I  recognize  the
 problem,  in fact,   even a lot  of refineries don't have  the proper facilities  for  handl-
 ing the  quantitites of sludge  that  are involved.  I think you will find in  the long
 run,  in  conjunction with the whole toxic material  system  sludges are going to  be
 handled more  or  less on  a regional basis.   I  think,  and  I am really speculating here,
 that the government  is interested  in  handling these in  a  more uniform  system.  Of
 course,  you are familiar  with the  toxic and hazardous materials  legislation that is
 being generated and  I understand  there is a very strong possibility  of regional  waste
 management for sludges,  etc.  In  the long  run I think you  will  see  a  regional
 approach  to taking these sludges  to a central  area for processing rather than trying
 to do them  on individual  basis, because it  is  very expensive and it  is very difficult
 to do.  As  far as an  individual facility like you describe  it would be a very
 expensive proposition  for them  to  put in a  complete  dewatering, oil handling,  oil
 removal,  sludge disposal  system at a  shipyard.

 Morris Wiley:  I think I  would agree with  that point but then there is a  subsidiary
 question.   P you  are going to consider Baltimore  harbor, for example, or New York
 or Philadelphia, there probably should  be a regional sludge disposal facility,  but
 the question would be should  that facility be  operated in conjunction  with  some  other
 petroleum industry facility such as a  refinery which  would have  a  way of  getting  rid
 of any recovered  oils from processing of this sludge  or should it be a  separate

-------
250
  authority which has no  connection with the petroleum industry?  The  marine people
  would like  to  give  it to the refineries and the  refineries don't really  want  it.  It
  is really an oil industry problem because they  are  carrying our materials.

  J.  A. Scher:  This gets into the  political  sphere.   I  am not so sure the oil  industry
  is going to operate these kinds  of facilities,  I  don't think  they want to operate
  these kind  of facilities.  I  really don't know what the answer  to that is.  I don't
  know whether  I can answer it and give you  a  very good answer but it is something I
  think is  going to  be approached on that type of a  scale  and who  the  operators and
  who the contractors are I don't know, I could  speculate  but  I  am just  guessing.

  Morris Wiley:   I  guess  it is an  open question,  I am asking the question, but not
  really expecting an answer  at this time.

                                                           BIOGRAPHY

                                           Jacoby A. Scher  is a Principal  Environ-
                                           mental  Engineer with  Fluor Engineers and
                                           Constructors, Inc. in  Houston,  Texas.  He
                                           holds  a B.A. in  Chemistry from Rice
                                           University and  an M.S. in Environmental
                                           Engineering from the University  of  Houston.
                                           He  is a Registered Professional  Engineer in
                                           the State  of  Texas.   Mr.  Scher has been
                                           involved in the Industrial  Waste Treatment
                                           Field  for  ten years and has been with Fluor
                                           since  1973.

-------
                                                                                         251
       TABLE 1   ESTIMATED OILY SLUDGE PRODUCTION FROM AIR FLOTATION TREATMENT

Refinery Class
Class
A
B
C
D
E
Air Flotation
Influent TSS
(mg/1)
29.5
35.2
52.2
99.3
29.9
Air Flotation
Effluent TSS
(mg/1)
8.8
10.6
15.6
29.8
8.9
TSS
Removal
(mg/1)
20.7
29.6
36.6
69.5
21.0

Flow
(gal/bbl)
18.0
38.0
42.7
47.3
87.0
TSS
Removal
(Ibs/mbbl)
3.1
9.4
13.0
27.5
15.3
                                                                       MEDIAN
  Economics of Refinery Wastewater Treatment, American Petroleum Institute, 1973.
                                                           15.3
         TABLE 2   COMPOSITION OF SLUDGE WITHDRAWN FROM OIL-WATER SEPARATORS
     API Units with Removal  Equipment

    Average                    Range

(Percent by Weight)        (Percent by Weight)
Water         82.4
Oil           10.9
Volatile Solids   1.4
Ash             5.3
61.2 to 98.3
 I.I to 20.7
 0.5 to  2.3
 0.4 to 15.8
                                    All Survey Samples
    Average

(Percent by Weight)

      66.5
      15.0
      5.5
      13.0
     Range

(Percent by Weight)

   8.9 to  99.5
   O.I  to  59.0
   0.0 to  49.4
   0.3 to  56.2
                   TABLES   COMPOSITION OF OIL-WATER SEPARATOR SLUDGES

                                        (Water-Free Basis)
                Average
           (Percent by Weight)
Oil              46.0
Volatile Solids    14.7
Ash              39.3
                                              Range
                                        (Percent by Weight)

                                           9.4 to 81.5
                                           0.0 to 54.3
                                           2.3 to 85.3

-------
  252
                        TABLE 4  COMPOSITION OF AIR FLOTATION SLUDGE
                                     (Alum Used as a Coagulant)
 Wafer removed by settling I hr, percent by weight
 Froth analysis, percent by weight:
   Oil-free solids
   Oil (by difference)
   Water (after settling I  hr.)
 Ash,  percent by weight of solids
                                       Average

                                         15

                                          3
                                         22
                                         75
                                         27
                                                     Range

                                                  2.5 to 42.0

                                                  1.0 to  5.2

                                                 60.8 to 90.1
                                                  3.1 to 43.8
 Water
 Oil
 Volatile Solids
 Ash
                       TABLES   COMPOSITION OF TANK BOTTOM SLUDGES
                                                                Average               Range

                                                           (Percent by Weight)   (Percent by Weight)
                                        39.4
                                        47.9
                                        4.4
                                        8.3
                                                  5.0 to 98.0
                                                  1.3 to 95.0
                                                  0.0 to 13.0
                                                  0.0 to 55.3
         Solid Waste
TABLE 6  PROJECTED WASTE SOLIDS INVENTORY*


                                            Characteristics
Dry Solids   % Solids   Volume
 Ibs/day      wt/wt    gal/day
1.  Gravity Oil Separator Sludge
    a. Oil Area                     95

    b. New Area                     79

2.  DAF Skimmings
    a. Old Area                     68
3.  Oily Sediment                  1,600
4.  Tank Bottoms                   2,500
                      1.2       1,200   Oily Sludge Containing
                                       up to  10% Oil  &I6%
                      1.2       1,000   Solids After Settling
                      0.5       1,700   Oil, Solids & Emulsion
                                       Breaking Chemicals
Remarks
                                                       Potential Oil
                                                       Recovery
                                                       Potential Oil
                                                       Recovery by
                                                       Chemical &
                                                       Heat Treatment
                    up to 20  Approx.
                               1,000
                               Oily Sediment Contain-
                               ing Clay, Sand, &
                               Carbonate Precipitates
                               and up to 10% Oil
                    up to 20    1,500   Up to 50% Oil
* Excluding Tar and CBM from the HFAIkylation Unit
                                                       Potential Oil
                                                       Recovery if Leaded
                                                       & No n-Leaded
                                                       Sludges are
                                                       Segregated

-------
                                                                                                           253
                      FIGURE  1.   OILY  SLUDGE  PRODUCTION  VERSUS  REFINERY  CAPACITY
          500,000
 . 200,000
^
       u
       a
       a
       a
       u
       
-------
 254
                                                  FIGURE 3
                                   TYPICAL AUTOMATIC WATER DRAW-OFF SYSTEM
 CRUDE OIL
STORAGE TANK

                            ixj-
                                             SLOPE
                                          OIL RETURN
                                  SLOPE
                            OILY-WATER SEPARATOR
                                                                          ?M STEAM TO
                                                                          HEATING COILS
                                                                       t
                                                                                    STEAM TRAP
                                              BALL FLOAT TYPE
                                              EMERGENCY SHUT-OFF VALVE

                                                  FIGURE 4
                                           RECOVERED OIJ, TREATMENT
                               HOT-OIL
                                                                                                   DRY OIL
  OIL
EMULSION
                                                                                                    WATEF

-------
                                               FIGURE  5


                                REFINERY WASTEWATER TREATMENT SYSTEM
  3AI.TAST WATR3
                                                                                        RAINFALL

                                                                                         RUNOFF
BREAKER
                                                                                                         EFFLUENT
                                                            POLYMER  INJECTION
pi
Cn
                                                 OIL TO CRUDE TANK

-------
256
                                        FIGURE  6

                                  MASS  BALANCE  DIAGRAM

                           OILY WASTE SOLIDS DISPOSAL SYSTEMS
       TANK
      BOTTOMS
       OILY
     SEDIMENT
1700 ihfi./Day 13 20%  = 1850 GPP
4115 Ibs./Day
                                    li\ 5 Ihg/Day
                                             GPD
3 15% =
SEPARATOR
BOTTOMS
DAF
SKIMMINGS

615 Ib/D


THICKENING
                                               15%  = 2890
                                                     GPD
CONTRACT
DISPOSAL
ALTERNATE
  NO. I
                                                   IB/I
                                              3 15% =
                                              2890  GPD
                                                          INCINERATION
                                                                         3500  Ib/E
                                                     DRY
                                                    SOLIDS
                                  LANDFILL
                                  ALTERNATE
                                    NO. 2
                                         FIGURE 7
                            OILY HASTE SOLIDS DISPOSAL SYSTEMS
                                      ESTIMATED COSTS
            THICKENING


            CAP. = 349,500

            0 & M = $9,000
                                             HAUL COST AT
                                             S92/T10 BBT..=
                                               521,000
                                                 CONTRACT

                                                 DISPOSAL

                                              590,150 Disposal
                                               Costs ($ 5.09/Gal.
                                                                     INCINERATION

                                                                    CAP. = $200,000

                                                                    C 5, H = 590,000

-------
                  "OILY WASTE DISPOSAL BY SOIL CULTIVATION"

                                   C. B. Kincannon
                       Texas Water Quality Board, Austin,  Texas
INTRODUCTION
    Numerous companies dispose of oily waste materials by adding the sludge to land
areas.  This method of disposal may be referred to as land disposal, land treatment,  land
spreading or land farming.

    Sources of oily sludges include petroleum refinery wastes, oil spills and pipe line
ruptures.  Also, numerous waste materials from industrial chemical and petrochemical
plants contain oil.  The quantity, quality and type or types of oil in oily sludges may
vary widely and often  the oil  exists as mixtures with solids.  Determination of the optimum
disposal method for each batch of sludge based upon chemical and physical composition of
the sludge is not practical.  Generally,  the disposal of oily waste sludges on land areas
has been done on a trial  and error basis.  The control of the disposal site may have been
based solely upon visual  observations, with a minimal of attention toward the quality of
surface water runoff or the effects on the soil or on the subsurface groundwater.

    The  purpose of this paper is to review a study using a soil cultivation process for
disposal of known oil  materials made in 1971 under the partial sponsorship of the Environ-
mental Protection Agency (1). In this study oily materials were spread over the soil, then
mixed with the soil using a roto-tiller type cultivator.  Soil microorganisms were utilized
to decompose the oil at prevailing soil and climatic conditions.  Experimentally determined
oil decomposition rates and other findings during the study suggest that the  process can  be
controlled and  is one  method for disposal of oily sludges at the plant site.

OILY MATERIALS  USED

    Since the oil present in oily sludges from a plant may vary widely in composition and
properties, the selection of representative  oils to use in a study may be somewhat arbitrary,
but should include oils similar to those which may be present in the plant waste  sludges.
For the cultivation  process study, oil feed materials were selected to  represent different
combinations of hydrocarbon types.  The properties and composition of three oils used in the
experiment are given  in Table 1.  The oils include the following:
    crude oil  tank  bottoms containing a natural balance of hydrocarbon types.
    bunker  C (No. 6 fuel  oil),  a residual  fuel oil containing olefinic and aromatic
    components.
    waxy oil, an intermediate refined product containing  highly paraffinic components.

    Differences in  pour points,  viscosity and  hydrocarbon  types are shown in Table  1 for
the three oil materials.
                                     257

-------
 258

TEST PLOTS

    The study included three parallel experiments with each oil material.  A total of nine
test plots were used.  A sketch of the experimental plots  is shown in Figure 1 .  Test plots
were treated identically except for soil nutrient additions.  Plot No. 1  was fertilized
heavily, Plot No. 2 was fertilized moderately and Plot No. 3 was not fertilized.

    Each of the nine plots was 12 x 125 feet; i.e., 1,500 square feet.   Plots were
separated by levees designed so that  rainfall runoff from each plot drained  through a pipe
to a ditch.  A plug in the pipe was used to control the runoff.  The levees  prevented
crossflow of water from one plot to another and kept oil within  the plot area.

    Space between the plots permitted vacuum truck passage for convenience in  spreading
oil evenly over the soil.  The roto-tiller was cleaned  after working each plot to prevent
mixing of oil, soil and fertilizer from one  plot to  another.

    The experimental plots were located in an area which had  been used previously for
oily waste disposal.   Residual oil  in the soil at the start of the experiment was approximately
ten percent by weight.  Rainfall runoff was accumulated  in a holding pond.  The depth of
soil mixed with the roto-tiller was approximately  six inches. The six inch depth  was used
in calculating the volume of the soil  in the test plot.

FERTILIZERS

    In agricultural crop production,  nitrogen and phosphorus are removed from the soil
with the harvested grain or forage.  Replenishing  of the nutrients is required. For a waste
disposal  system such as the soil cultivation process, there should be an equilibrium condition
established which would minimize the need for supplementing the nutrients.  Endogenous
metabolism,  autoxidation of cellular protoplasm,  results in the  release of nitrogen and
phosphorus previously used for synthesis.   The released nutrients are made available for
reuse  in biological  waste treatment systems (2).  No reference  information  was found to
use as a guide in selection of appropriate  quantities of nitrogen and phosphorus which
should be added to the biological system.   Instead, the selection of fertilizer quantities
was based  upon agricultural-oriented experience. Warnings that 1) excess nitrogen
fertilizer elements hinder (poison) bacterial action and 2) excessive total soluble salts may
cause unfavorable osmotic conditions for bacterial growth, were considered in the selection
of the quantities  of nutrient additive.  Excess phosphate was not considered toxic except
for its contribution  to the total soluble salt content.

    Urea and calcium hydrogen phosphate were added initially as fertilizers.  Plots
numbered 1, heavily fertilized, received  1 ,000 pounds of nitrogen (N) and 200 pounds of
Phosphate  (P~0,-) per acre  compared to 500 pounds of  nitrogen and 100 pounds of (P00,-) f°r
the moderately fertilized plots numbered 2. No supplemental nutrient was added to plots
numbered 3.

    Fertilizer materials  added during the  course of the experiment to plots number 1 and
number  2 were based upon the results of periodic  analysis of nitrogen and phosphorus in the

-------
                                                                                   259


soil.  After about 6 months, loss of ammonia to the atmosphere was noted and a change in
the type of nitrogen fertilizer material appeared warranted.  About midway of the project,
ammonia nitrate (NH ,NOo) was used instead of urea and relatively small doses applied
frequently.

EXPERIMENTAL PROCEDURE

     Each  plot was cultivated with a roto-tiller type plow.  Fertilizer materials were added
by manually spreading solid granular fertilizer over the surface of plots No. 1 and 2.
After spreading fertilizer and before adding oily sludge materials, the  plots were again
roto-tilled to mix fertilizer and soil.

     Oily feed materials were transported to the test plot area with a vacuum tank truck,
and distributed over the soil surface by manual  direction of a discharge hose.  After
cultivating with the roto-tiller, the oil-soil mixture appeared uniform  and  presented no
problem at the ambient May temperature of 80 degrees F.  The initial schedule for roto-
tilling was once each two weeks, weather permitting. After six months of cultivation, the
soil was friable, had lost its oily appearance, and based upon test results was suitable  for
a second addition of the oily sludge.   Differences in  the plots were apparent.  The color of
the unfertilized No. 3 plots was darker than for fertilized  Plots No. 1  and 2.

     The date for the second addition of oils was in early February when  the temperature
was in the low 40 degrees F range.  Congealing and solidification of the oils was apparent.
Mixing of the viscous oily  matter into the soil was not successful  until  the ambient tempera-
ture had increased to about 80 degrees F.

     At the end of the  18-month experiment, the soil had again returned to the friable
condition with the appearance of normal agricultural soil.

     Although no attempt was made to grow vegetation on soils previously used for oil
decomposition during this experiment, native grass and plants did sprout and grow on top of
the levee. The soil  in the levees was the same as the starting test plots, which contained
about ten percent by weight (10%w) of oily material.

SAMPLING AND TESTING

     Oil and  nutrient contents were used as controls for the experimental work.  Generally,
oil determinations were made once every two weeks and nutrients were determined once
per month.  Microbial analyses were made monthly, hydrocarbon types of oil added and
extracted from the soil were made for each new addition of sludge to the soil, and oil
content of core soil samples were made at the start, midway and at the end of the 18-month
experiment.

     Obtaining representative samples was a recognized  problem from the start of the project.
The sampling procedure included combining portions of soil from three  sampling points to
form subsamples a, b, and  c,  as shown in Figure 2.  Subsamples a, b,  and c were compo-
sited to represent the whole test plot.

-------
260

 EXPERIMENTAL RESULTS

     The experimental results when using the three oily materials as feed were similar.  The
 relationships of the percent weight (%w) oil in the soil with time are shown in Figure 3 for
 the waxy oil test plots and these results are representative of the other two oily material
 test results.  From the first addition of oil  in May to about November  1970, the oil content
 of the soil decreased markedly.  During the period November 1970 to February  1971, a
 reduction of oil was not apparent.   From February, when a second dosage  of oil was added
 to October 1971,  the oil content again decreased.  The reason  or reasons  for  the period of
 inactivity, November 1970 to  February 1971, may have been due to one,  two or all of the
 following reasons:  During this period, the concentration of oil  in the soil  was near the
 starting content and this residual oil may  be nonreactive or slowly reactive with soil
 microorganisms.  Another reason may have been the lack of available nitrogen and a third
 reason may have been the relatively low temperatures during this period.

     The relationship of the nitrogen concentration and oil removal rate is shown in Figure
 4 and the relationship of temperature and oil removal  rate is shown in Figure  5. Jhe oil
 removal rate was expressed  as pounds of oil removed per cubic foot of soil  (Ibs/ft  ).  During
 the period November 1970 to February 1971, the  ammonia concentration was  20 mg/l or
 less for waxy oil  experiments and the nitrate content was nil indicating a possible lack of
 the nitrogen nutrient.  Also, during this period,  the soil temperature, shown  in Figure 5,
 was less than 40 degrees F and may have been too low for favorable bacterial growth. The
 reason or reasons for the period of  inactivity was not further determined.

     Oil decomposition rates expressed in pounds of oil per month per  cubic foot of soil
 (Ibs/mo/ft ) are given  in Table 2.   For the periods May to November  1970, and February
 to October 1971, the oil decomposition rate averaged Q.9, 0.9 and 0.7 Ibs/ft /mo for
 Plots 1, 2 and 3 for crude oil;  L.8, 1.7 and 0.8 Ibs/ft /mo for  Plots 1, 2  and 3 for Bunker
 C;  and 1 .7,  1 .3 and 0.7 Ibs/ft /mo for Plots 1,  2,  and 3 for waxy oil feed.  For  a three
 month period of November  1970 to February 1971, the oil decomposition rate was minimum.
 The average decomposition  rate per year at these  test conditions was 1.1,  1.0 and 0.6
 Ibs/ft /mo for heavily  fertilized No. 1, medium fertilized  No. 2 and unfertilized No. 3
 plots, respectively.

     Tjne general  conclusion was that the oil decomposition)  rates were approximately 0.5
 Ibs/ft  of soil per month without fertilizer and 1 .0 Ibs/ft /month when fertilized.   Items
 for consideration in developing the cost'of the process using fertilizer are  given in Table 3.
 The fertilizer requirement may be assumed to be  1,000 pounds of nitrogen  and 500 pounds
 of phosphorus per acre  per year. The cost for spreading oil and fertilizer and cultivating
 the soil likely would vary with  location.   An oil decomposition rate of 1 .0 Ib/ft /mo is
 equal to 21,780 Ibs/acre/mo or about 70 barrels (bbls)/acre/mo.  The cost for delivering
 the oil to the area  and distributing the oil over the surface  of the soil was about $l/bbl  or
 $70/acre/mo in 1971 .  The fertilizer material and labor for spreading costs prorated on a
 monthly basis was estimated to be about $23/mo.  Cultivation (plowing) would require
 about 8 hours (hrs)/acre at  $25/hr  and two plowings/mo totals $400/mo.  The total cost/
 acre/mo was estimated to be $493  or about $7 per barrel of oil  at conditions  prevailing in
 1971  .  The cost for disposal of oily sludge containing 33 percent oil was estimated to be

-------
                                                                                  261

about $3 per barrel in 1971.  The cost of disposal  of oily sludge would depend upon the
oil content of  the sludge.  Current  costs for a specific entity or industrial plant would of
course require an independent evaluation.

MICROBIAL ACTION

     Microorganisms in the soil decompose  nitrogenous and carbonaceous materials  into
microbial cellular matter.  When oil or hydrocarbon is the only source of carbon, oil
degrading microorganisms survive and become the  predominant species.  Results of  monthly
determinations of the total aerobic  count indicate that the addition of fertilizer affected
the total aerobic count.  The relationship of total aerobic and oil removal rate and the
effect of fertilizer are shown in Figure 6 for the waxy oil experiments.  The No. 1 plot
(heavily fertilized) generally was higher in total aerobic count than No. 2 (medium
fertilized) and No. 3 (not fertilized) plots.  Also, the microbial count was found to be
sensitive to  change in the oil content of the soil as shown in Figure 7.  The addition of
oil in May 1970 and February 1971  appeared to upset the soil microbial equilibrium and
caused the total aerobic count to be low.

     The predominant species given in Table 4 included Arthrobacter,  Corynebacterium,
Flavobacterium, Nocardia, and Pseudomonas.  Since the area where the experimental
plots were located had been used previously for oily waste disposal, the presence of a
variety of oil  consuming microorganisms might be  expected in the soil.   Flavobacterium
and  Pseudomonas species were present but the microbial  population was different on all
plots. Distribution of the microorganism species were not greatly different  for the  three
oils  used in  the experiment.

OIL  EXTRACTED FROM THE SOIL

     Oils added  to and extracted from soil  plots were quantitatively analyzed for hydro-
carbon types,  qualitatively analyzed by infrared for "fingerprint" differences and
qualitatively analyzed by gas chromatography for boiling point profile.  Detailed
description and results of the analysis have  been reported previously (1).  Some of the more
pertinent procedures and results follow.

     Adsorbents of clay  and silica gel were  used to separate the oils into total  saturates,
resins (asphaltenes, organic acid,  etc.) and aromatics.  The hydrocarbon content of the
fertilized plots related with time are shown on Figure 8.   Generally, the saturate  and
aromatic contents are shown to decrease and the resin content remained unchanged or
increased.  This may indicate the saturate  and aromatic  materials to be decomposed and
the resinous material to be relatively nonreactive with soil microorganisms.  Also shown is
the hydrocarbon content of the oil  originally in the soil.

     Generally from the three parallel experiments, more changes  in composition for the
No.  1 and No.  2 plots  were noted  than for the  No. 3,  indicating an effect of fertilizer.
Further,  differences were noted from infrared analyses.

     The oil originally in  the soil contained saturate groups, small amount of organic acids

-------
262

 and a substantial amount of condensed multi-ring aromatic structures.

      Infrared spectra for crude added to the soil  and after eight months in the soil are shown
 on Figure 9.  Some observations follow:
      1 .  The crude oil feed contained saturates and simple aromatics.
      2.  The oil extracted from the unfertilized plot was similar to the original oil  in the
         soil.  The aromatic were more representative of the polyaromatic groups, and
         saturates were present.
      3.  The oil extracted from the fertilized plot still contained polyaromatic types  and
         some saturates but contained  organic acids, also.
      4.  Organic acid contents are significantly different in the fertilized and unfertilized
         samples.   Saturate peak intensities are related to quantity and are  higher in the
         unfertilized sample.

      In general,  oils from plots that were fertilized showed higher concentrations of organic
 acids and the concentration appeared to be greatest in the heavily fertilized plots.

 INFILTRATION

      Oil and nutrient concentrations at depths of 2, 4,  and 6  feet, determined at the start,
 middle, and end of the project, indicated  that these constituents did not infiltrate the soil
 under the conditions of the project.

 RAINFALL  RUNOFF

      Analysis of accumulated rainwater for oil, ammonia, nitrate, and phosphate contents
 indicated that little, if any,  of these constituents were present when drained immediately
 after a rain.  Long standing appeared to cause higher oil contents.

      The oil content ranged from 30 to TOO mg/1 and infrared examination of these  extracted
 oils  indicated  the  soluble oil  to be essentially organic acids with characteristics quite
 similar with spectra for naphthenic acids.

      Each plot was drained about 36 times during the project and the quantity of oil
 discharged  with the water was equal to about 0.03 Ibs/ft  of oil  in the soil.

      The ammonia  content of  the water varied with the soil nutrient level.  During periods
 when the ammonia was excessive on the No. 1  plots, discharge water contained up to 150
 mg/1 ammonia as  N.  Nitrates and phosphates were generally absent in the discharge.

 APPLICATION OF THE PROCESS

      This study demonstrated that soil  microorganisms decompose petroleum oily waste and
 that  oil and fertilizer chemicals did not penetrate the soil at the location and conditions
 of the test.  However, numerous factors are left unresolved and future experiments should
 gain knowledge concerning the following uncertainties:

-------
                                                                                 263


    1 .  The residual oil in the soil  after the  18-month project contained organic acids,
        and those soluble in rainfall runoff water were primarily naphthenic acids.  Since
        most organic acids are water soluble, this suggests that  naphthenic acids do not
        decompose  in the soil as rapidly as aliphatic acids or that naphthenic acids are
        formed from aliphatic or aromatic materials  at the soil conditions. Speculatively,
        large concentrations of  naphthenic acids may inhibit microbial growth in the soil,
        and water washing of these organic acids from the soil to a separate biotreatment
        system  might improve the oil decomposition  rate in the soil.
    2.  Residual  oil extracted from the soil was characterized by infrared to  be  poly-
        aromatic oils, suggesting this hydrocarbon group to have a slow reaction or to  be
        nonreactive for microbial decomposition at the prevailing conditions. Certain,
        and yet unknown, environmental conditions might improve the decomposition rate
        of the polyaromatics.
    3.  Accumulation and buildup  of metals  or salts contents with  long-range usage may
        affect the microorganism activity. The acidity or alkalinity range, optimum for
        microbial action, may vary with the soil contents.
    4.  The time required for native soil bacteria to become acclimated oil decomposers
        likely depends upon  the soil composition and temperature.
For application of the process, one should follow the suggested action items given in
Table 5.

CONCLUSIONS

    The decomposition rate oLoily sludges (hydrocarbons) by microbial action in cultivated
soil averaged about 0.5 Ibs/ft /mo without fertilizers and about 1 .0 Ib/ft /mo when
fertilizers were added.

    Major bacterial species included Pseudomonas,  Nocardia, Flavobacterium,  Coryne-
bacterium, and Arthrobacter.

    Organic acids are produced and water solubles appear to be naphthenic acids.

    Oil and  fertilizer did not infiltrate.

    Rainfall  runoff water contained organic acids (and possibly  fertilizer).

    The soil  cultivation process is one method which can be used to dispose of oily waste
materials.

REFERENCES

(1)  Oily Waste  Disposal  by Soil Cultivation  Process.  EPA-R2-72-11D, Office of Research
    and Monitoring, Environmental Protection Agency,  December  (1972)
(2)  Manual on  Disposal of Refinery Wastes, Volume on Liquid Waste, Chapter 13,
    Biological  Treatment, First  Edition, 1969, American Petroleum Institute, Division of
    Refining, New York, N. Y.

-------
264

DISCUSSION

Banks:  Has this soil been returned to cultivation and if so, what were the crop yields?

C.  B.  Kincannon:  This experiment did not include agricultural  crop production.  In the
study, the soil  was utilized as a bacterial bed for the decomposition of oils.  I  cannot answer
your question concerning the current use of the experimental area because I am no longer
associated with Shell Oil Company where the work was done.  I refer you to Shell Oil
Company, Deerpark, Texas for further information along these lines.

Morris Wiley:  We have used an alternate method which does not involve cultivation  whereby
you put 0.5 cm of oil sludge over,  say 1 square mile and find that not much happens during
the cold season in say New York.  During the summer and certainly by September, the oil is
weathed completely into little "pebbles" which the earth worms cover up.  You can put
5,000 cubic meters per square kilometer which is a much lower application than you are
discussing.  The difference is to eliminate the cost of having to cultivate, till,  fertilize,
etc.  Would you care to comment on  the potential savings of such a method?

C. B. Kincannon:  I believe we should  know more about the  material put on  the soil.

Morris Wiley:  I should have described that.  It was a refinery disposal pit sludge stabilized
for some years  leaving principally the heavy ends  - 20% oil,  20% water, the rest primarily
solids.

C. B. Kincannon:  With my limited knowledge,  I  would say  if that works - power to you -
    ~    ~
 Morris Wiley:  Do you think that would be cheaper if the land area were available?

 C. B. Kincannon:  If you can place the waste on the land area, not fertilize, till, etc.  and
 the waste will  disappear without damage to the receiving water  or create an air pollution
 problem, the disposal cost should be a minimal cost.

 Morris Wiley:  Would you think that you would have to collect the land run-off in an
 impoundment?

 C. B. Kincannon:  I would  recommend that.

 R.  A. Farnham:  Could you explain the meaning of that  No. 9 you had up there - the 25
 milligrams per  liter, I  thought we were about in the 30% oil range?

 C. B. Kincannon:  As  I pointed out, we found from exploratory  experiments in the
 laboratory that concentrations in excess of 30 percent weight oil caused the oil decomposition
 reaction rate to become slow or nil.  When the soil particle becomes completely enclosed
 with  oil  globules, water is prevented from getting into the soil and water is a necessary
 element  for the bacteria.

-------
                                                                              265
BIOGRAPHY

C. B. Kincannon is an engineer in the Central
Operations Division of the Texas Water Quality
Board (TWQB).  He obtained a  B.S. degree
from Panhandle A and M College (now Panhandle
State University) and was an employee of Shell
Oil Company for 30 years before joining the
TWQB staff.  He is a certified  Grade  A sewage^
works operator and a professional engineer
registered  in the  State  of Texas.
                             TABLE 1  "OIL PROPERTIES"

                                      Crude Oil        Bunker C         Waxy Oil

Lb/Gal                                   7.12             8.52            7.08

Pour Point (°F)                             -5               40              95

Viscosity
    SU    60°F                            60             19000
    SF   122°F                                            120              59

HC Type (Wt%)
    Saturates                              36               18              90
    Resins                                  8               26                0
    Aromatics                             56               56              10

-------
266
                   TABLE 2 "OIL DECOMPOSITION RATE LB/MONTH/CU FT OF SOIL1
                                                                        Nov 70 to
                                                                         Feb 71

                                                                           Min
                                                                           Min
                                                                           Min

                                                                           Min
                                                                           Min
                                                                           Min

                                                                           Min
                                                                           Min
                                                                           Min

Feed
Crude
Oil

Bunker
C

Waxy
Oil

Average



Plot
1
2
3
1
2
3
1
2
3
1
2
3
May to
Nov 70
0.7
0.5
0.2
1.8
1.2
1.8
1.8
1.5
0.8



Feb to
Oct 71
1.1
1.4
1.3
1.8
1.5
0.5
1.6
1.0
0.5




Avg
0.9
0.9
0.7
1.8
1.7
0.8
1.7
1.3
0.7



                                TABLE 3 "ITEMS FOR ESTIMATING COST"
                            Oil Spreading/Month
                               Lb/Cu Ft
                            Fertil izer
                               Nitrogen Lb/Year
                               Phosphorus
                            Cultivation
Quantity
    1
   70

 1000
  500
                    Yearly
                      Avg

                      0.7
                      1.3
                      1.3
                      0.6

                      1.3
                      0.9
                      0.5

                      1.1
                      1.0
                      0.6
                                         TABLE 4 "MICROBIAL ANALYSES'

                                                Predominant Species

                                                 Arthrobacter

                                                 Corynebacterium

                                                 Flavobacterium

                                                 Nocardia

                                                 Pseudomonas
                                  TABLE 5 "FOR APPLICATION OF THE PROCESS1

                    1 .  Prepare Area
                    2.  Provide Treatment Facilities For Rainfall Runoff
                    3.  Obtain Samples To Determine Background Level
                    4.  Add Fertilizer Material
                    5.  Cultivate
                    6.  Distribute  Oil Over Surface
                    7.  Cultivate, Analyze For Oil
                    8.  At Regular Intervals,  Cultivate and Analyze
                    9.  Oil Content Of Soil Should  Not  Exceed 25% wt.
                   10.  Determine Oil  Penetration Rate Into Soil

-------
                                                 267
                 FIGURE I
            EXPERIMENTAL PLOTS
EACH PLOT
12 FT x 125 FT
1500 FT2
750 FT3
                                  TO HOLDING POND
1

^-^


1
J 1





2





3



i





— -





t
1





2





3



t


-—





t
1





2





3



f N


                                            L
CRUDE OIL
               BUNKER C      WAXY OIL
              ( NO 6 FUEL OIL)

                        FIGURE 2
                TYPICAL SAMPLING  POINTS
                          SUB  SAMPLE c
                          SUB SAMPLE  b
                          SUB SAMPLE  a

-------
    268
                                  FIGURE 3
                                 OIL  IN  SOIL
V)

        LJ Od
        O Q
        en
        LJ
        a.
             30

             20

             10



             30

             20

             10
   30

   20

   10
                                      HEAVILY
                                      FERTILIZED
                                      MODERATELY
                                      FERTILIZED
                                               UNFERTILIZED
                    APR
                 JULY   OCT
JAN
APR   JULY
OCT
                FIGURE 4
         EFFECT OF FERTILIZER
   800
E
a.
CL

~

LU
O
O
  400
  200
                                    Q

                                    >d
                                    oo
                      APR
                            3 —    =
                   OCT
   APR
    OCT
                                                           APR

-------
                                                     269
 OIL REMOVED

(Ibs/Ft3 SOIL)
SOIL TEMPERATURE (°F)
                             ->l   CO
                             o   o
                                                m
                                                -n
                                                -n
                                                m
                                                o


                                                0
                                                TJOl
                                                m
                                                -I


                                                m
   OIL REMOVED

  (Ibs/Ft3 SOIL)
     —  ro
     o  o
o
o
  TOTAL AEROBES (xlO7)
IN)

O
00
o
o
o
                                              CD <*)
                                                 m
                                              o
                                              o

-------
270
                  FIGURE 7
           EFFECT OF OIL ADDITION
             ON MICROBIAL COUNT
      o 120
      X
      ~ 100

      m 80
      o
      LJ 60

      ^ 40

      £ 20
      /  i
    /  !
        I
   JL    \
r^-^

         30

         20
         APR
 OCT
APR
OCT

-------
                                271
    FIGURE 8
HYDROCARBON  TYPES
30
20
10
0
h- 30
X
o
g 20
I-
5 10
o
cc
w ^
CL 0
30
20
10
n
—

—
—
•»
\
S
R
A
I
R
A
c
f
A
S
R
A
S
R
•—•™
A
S
B
A
I-SATUR,
*-RESIN<
k-AROMAl
X.
V.
x»
•^
•x.
^
S,
\
\
^:\
•m. '
**-.
X
X
X
•x
X
X




Al
%
rn
s
R
tMita-Bi
A
§
R
A
S
R
A
TES
CS
\
X
\
••»
x»
"* *•»•.
X
X
X
\
X
^s
1''
MJ
\
\

X
N
t^t^^
••K-^
-«-

\
^
•s
*"***»
"* ^
X
X
N
\
\
\
X
\
X.



—




                      CRUDE OIL
                       BUNKER C
                       WAXY OIL
APR   OCT   APR   OCT

-------
                            FIGURE 9
                       INFRA-RED SPECTRA
CN
r\
CN
      UJ
      o
      (O
      <
      tr
100
80
60
40
20
 0

100
80
60
40
20
 0

100
80
60
40
20
 0
                 OH
00
                     C-0
=0  CH2 u"u       OH
C=C  CH3 AROMATIC    AROMATIC
                                             CRUDE OIL
                     UNFERTILIZED
                                            FERTILIZED
  6      8     10    12

WAVE LENGTH (MICRONS)
                                                  14

-------
                THE DISPOSAL  OF OILY WASTES BY LAND FARMING

                                  R. L. Huddleston
                                         and
                                  L. W. Cresswell
                  Continental Oil Company,  Ponca City, Oklahoma

   At CONOCO we have been studying the disposal of oily wastes by land farming for
about three years.  By "land farming" we mean the use of the soil environment for the
planned, orderly treatment of waste.  The work has  been strictly experimental  in nature,
but has been conducted in field plots out-of-doors using actual waste materials.  In this
presentation we  will describe the need for land farming as an oily waste disposal method,
the results of our studies to date, and our view of factors that must yet be addressed before
land farming can be used routinely on an environmentally safe basis for oily wastes disposal.

SOURCES AND  CHARACTERISTICS OF OILY WASTES

   Oily wastes are  generated by numerous routine refinery operations. Examples of such
wastes and their compositions are shown in Table 1.   Amounts of these wastes generated are
not well defined  and vary widely from refinery  to refinery and from time to time within  a
given refinery.  However,  it is probable that a 100,000 BPD refinery generates from 5,000
to 100,000 bbl/year of such wastes.  As  increasingly better jobs  are done in waste water
treatment and  air pollution control, the amount of oily wastes that are generated will
inevitably increase.

   Some  oil can be recovered from these wastes by chemical  and physical means, and it is
very important that we learn how to more effectively and economically deoil such wastes.
It is unlikely, however, that we will ever be able to economically deoil all wastes of these
types.

   The characteristics of waste  oil prevent its  decomposition in  conventional waste treat-
ment systems.  Some disposal methods used in the past are shown  in Table  2.  For various
reasons these methods are rarely used today, and we obviously need suitable alternatives.

   At present, it appears that incineration and land  farming may be the only safe and
possibly economic ways to dispose of oily waste.  Although incineration is expensive, a
fire hazard, poses air pollution problems, and  is difficult to adapt to wastes containing
non-combustible solids and high  water  content, it may have to be used at locations where
land is economically unavailable. An alternative and less expensive disposal  method may
be land farming.

THE  CONCEPT OF LAND FARMING

   It is well known that many microorganisms can  decompose numerous petroleum components.
We must also recognize, however, that petroleum is an extremely complex blend of tens
of thousands of chemical structures.  Our knowledge of how,  and to what extent,  this
complex blend is biodegradable,  is quite small indeed.  Nevertheless, many workers have

                                       273

-------
274

shown that the bulk of at least some crude oils is biodegradable (1,2,3).

   Soil  contains an enormous microbial population as shown in Table 3.  Many of these
organisms can decompose hydrocarbons (4).  Although the water insolubility of petroleum
constituents (see Table 4) adversely affects biodegradation rate, this property helps to "fix"
oily  wastes in place when they are added to the soil.  The extremely large  surface area
offered  by the soil environment (see Table 5) helps spread added oil into thin  films making
the oil more vulnerable to microbial attack.

   A simplistic scheme of oily wastes disposal  by land farming is shown in Table 6. We must
quickly point out that the mechanism shown represents what we "hope" will occur and has
not been proven as yet.  Exceptions may be waste oil constituents that are not completely
mineralized and others that may prove to be recalcitrant to degradation.

   At present we need the answers to a number of questions to define the suitability of land
farming for oily wastes disposal.  Some of these  questions are posed in Table 7.  At present,
we have partial  answers to several, and there  is almost  certainly justification  for using land
farming prior to having complete answers to all of these questions.  If we are  diligent in
providing sufficient data to show that oily wastes land farming is reasonably safe from an
environmental standpoint, we can logically expect little valid opposition to its use in the
near future.  More complete  answers and answers to  some of the less urgent questions  can be
ascertained as land farming is practiced.

CONOCO'S EXPERIMENTAL LAND  FARMING STUDIES

   Billings, Montana.  At a site  near Billings, Montana,  we have studied the land farming
disposal of oily pond  bottoms.  A view of the  area prior to preparation is shown in Figure 1.
The  test site and surrounding  area was examined to assess the plant and animal life, soil
microbial populations and soil composition. Numerous 4^ foot cores were taken to document
subsurface soil  characteristics.  The surface soil was found to be a sandy clay  containing  a
low  humus level and very small amounts  of nitrogen and phosphate.  The Billings climate  is
arid, warm in the  summer and very cold  in  the winter.

   The test area was diked with subsoil,  plowed and disked. The  oily waste,  described in
Table 8, was spread as evenly as possible on the soil  surface using a vacuum truck.  We  later
found that by using a  large diameter hose instead of a truck to spread  the waste, we could
prevent the heavy trucks from compacting and generally disturbing the test  plot soil and  get
more even waste distribution.

   After application, the waste was allowed to dry and weather for three weeks.  This step
changed the waste to  a crumbly consistency, making it  much easier to work into the soil.
Even during this period no significant odor problem was evident.  The waste was then
blended into the upper 6-inches of soil with a tractor-mounted rototiller.  The appearance
of the area at this point is shown in Figures 2  and 3.  A description of the cultivation,
fertilization and sampling pattern used is shown  in Table 9.

   The test plots were fertilized with nitrogen and phosphate.  This was done  because the
soil and waste contained very small amounts of these nutrients, and we did  not want to limit

-------
                                                                                  275


rate of oil degradation by these controllable factors.  An increased soil microbial population
must be preceded not only by available carbon, but also by available nitrogen and phosphate.
These nutrients were added as a mixture of urea and 15-15-15, the only fertilizers available
at the time.

   Thus, the soil at the start of this land farm study contained the waste constituents as
shown in Table 10.

   During the course  of the study,  oil contents were measured using a benzene-methanol
soxhlet  gravimetric technique.  A Leco combustion furnace method was used to determine
total organic carbon content.  Paraffin,  aromatic and resin-asphaltene contents of oil
extracts were estimated by taking the extracts up  in tetrahydrofuran and subjecting them to
thin layer chromatography.

   Over an 18-month exposure period, we observed a 70% oil loss from the soil as shown
in Figure 4.  As also  illustrated, the loss was apparently created by an 88% loss of the
paraffins initially present in  the oil, an 85% loss  of the aromatics and only a 2.5% loss
of the resin-asphaltenes fraction.   The original oil contained 50% paraffins, 30% aromatics
and 20% resin-asphaltenes.  After 18-months of exposure in the soil, the residue contained
only 20% paraffins and 15%  aromatics, while the resin-asphaltenes fraction had increased
to 65%.

   Unfortunately, measurements of the organic carbon content of the soil samples produced
very scattered results as shown in Figure 5.   While the scatter is more pronounced than
usual in this  case, similar scatter is not uncommon when examining field soil analytical
data.  Sample differences are magnified by the difficulty in producing  thorough soil-
waste blending using  only practical field cultivation techniques.  Scatter can be minimized
by compositing large  numbers of samples, very  careful composite blending and sub-sample
taking.   We  have found it necessary to accept  considerable data scatter to economically
conduct our studies.  In spite of the scatter, the TOC data does seem to confirm that the
oil was  not only degraded but that  much of  it was mineralized completely to CO,,.

   Accompanying data over the 18-month period showed that the pH of the soil did  not
change, there was little nitrogen and phosphate loss, and the soil bacterial population
increased about 10-fold.

   As expected,  climatic conditions over the l:8-month  period of soil exposure were not
entirely conducive to good microbial activity.  As can be seen from Figure 4, the air
temperature during 10 of the 18-months of exposure was about 10  C or  less.  In addition,
the soil  moisture  content ranged between only 9 and 18% over the  18-month interval.
For optimum soil microbial activity,  the soil moisture content should be at least 30% (5).
It should also be emphasized that the oily waste studied came from a waste water treatment
pond (bottoms).  As a result, the waste had  been exposed to microbial attack for an
extended period prior to land farming.  Undoubtedly,  the most readily  biodegradable
substances had been removed from the waste prior to land farm exposure.

   It should also be mentioned that crested wheat  was grown on part of the test area.

-------
276


 Visual observations judged the plant growth to be normal and identical  to the control (no
 waste oil) plot crop.

    The  average rate of waste oil loss in the  Billings study is shown in Table 1 1 .  Losses are
 shown for the 18-month period and on an annualized basis.

    Ponca City, Oklahoma.  At Ponca City, we  have been studying the land farming of an
 oily  (crude petroleum) sand waste and, to a limited extent,  the land farming of a refinery
 oily  waste.  Unfortunately,  the refinery waste was inadvertently added to a soil that was
 later found to  have been previously contaminated with oil and  salts.  This problem  has
 complicated evaluation of certain portions of our studies.

    Methods used for test site preparation, waste application and analytical  work were similar
 or identical to those used in the Billings study.  The oily sand was extremely viscous and
 could not be flowed onto the plots.  Due to the  small  size of the oily sand test plots, we
 were able to manually distribute the waste over  them.

    Views of fhe oily sand and refinery waste plots are shown in Figures  6 through 9. Test
 plots were fertilized after waste weathering and soil waste blending.

    The composition of the refinery oily waste is  shown in Table 12.  Addition of this waste
 to the soil produced the waste-in-soil  levels shown  in Table 13.  The oil in the refinery
 waste was found to contain 21% paraffin, 49% aromatics and 30% resins-asphaltenes
 (Table  14).  Amounts  of nitrogen (as NHL) in the fertilized  test plots ranged from 50 mg/kg
 soil to  320 mg/kg soil  and phosphorus (as PO.) ranged from  10  to 50 mg/kg soil. To date
 we have  noted no detectable effect of fertilizer level, within these concentration limits,
 on the  rate of oil degradation.   Even in the most heavy fertilized plots, we have not found
 over 20 ppm NO,-,.

    Over  a 24-month exposure period, we have measured at  35% loss of oil  from the test plots.

    Composition of the oily sand  that we have been investigating is  shown  in Table 15.  In
 oily  sand test plots ammonium nitrogen and phosphate phosphorus levels have been kept above
 50 and 30 mg/kg,  respectively.  Soil water contents have stayed around 15% (about 30%
 of moisture holding capacity) during exposure  and we have observed no difference in pH
 between  test plots and control plots.

    Soil core samples,  30 inches  deep, were taken directly beneath test plots and in central
 areas one year after oily waste was applied  to the test areas.   Results of extractable oil and
 organic carbon measurements on these cores are shown in Table 16.  As can be seen, these
 data indicate that the  waste  has been strictly held within the shallow soil zone  in which  it
 was originally placed.  For  these tests we selected plots to which we had  added 8 and 10
 weight % oil  to the soil  in order to examine most severe test conditions.

    As may be seen in  Figure  10, we have observed about 40% degradation of oil in these
 oily  sand test plots over a 22-month period.  Actually this expression is an  average of what
 we observed over several plots.  A  plot that initially  had 8% oil in the soil was found to

-------
                                                                                  277
   e experienced a 41% oil loss and several 5% oil plots showed an average of 45% oil
   .  This small difference, of course, has a sizable effect on mass oil lost.  This is illu-
have
loss
strated in Table 17.
   Oil composition measurements (see Figure 8) have shown that about 82% of the paraffins
were lost along with 60% of the aromatics.  Only 1% of the resin-asphaltenes fraction was
lost.  This oil  initially contained 22% paraffins,  28% aromatics and 50% resin-asphaltenes.
The 22-month  residue consists of over 80% resin-asphaltenes.

OVERALL PERSPECTIVE OF LAND FARMING DISPOSAL OF OILY WASTES

   Several  companies have been practicing various forms of land farming as a means of
oily waste disposal  for many years.  While most have relied on  visual oil loss as evidence
of success, several  investigators have examined the technical aspects of land farming  in
recent years.  Among those that have carried out such  investigative studies are Shell Oil
Company, Union Carbide and Sun Oil Company.   A summary of oil and oily wastes studied
and reported rates of oil degradation are shown in Table  18.  We have taken the liberty of
converting all reported rates into annualized figures and identical units of measure so they
can be more easily  compared.

   All of these data support the contention reached long ago with  little or no analytical
data;  crude  petroleum  and many of its components are biodegraded in the soil.  Of course,
we still don't  know to  what extent such  materials are degraded  in the soil.   This we must
learn. We have also some data now that oil  in the soil can be  degraded at reasonable
rates in climates as cold and as arid as Montana,  as well as in warmer and more humid
climates such  as Houston and Corpus Christi.  We  have also learned that if we can
mechanically  handle the presence of an initial oil addition of 10% in the soil,  the rate
of oil degradation is greater.  Here at CONOCO, we found that 5% additions are much
more manageable, especially during the initial stages of waste  oil and soil  blending.  We
now have data that indicates that as little as 50 ppm nitrogen and  20 ppm phosphorus
(109 Ib. and 44 Ib. per acre, respectively) are sufficient to support the degradation of
some oils at their maximum rates in some climates. We also have data that indicates oil
in soil can be  decomposed at reasonable rates even when the water content of the  soil
stays below  20%.  We have evidence that waste oil tends to stay tightly bound in  soil to
which it is added while degradation is occurring.  There is some data that indicates normal
crops  can be grown in  soil where oily waste is being decomposed.  All of these data are
encouraging for those of us who hope we can prove the acceptability of land farming as an
oily waste disposal  technique.

   There remains, however, a number of very important factors that must be more clearly
established before we can safely recommend the full scale use of land farming for the
disposal of oily wastes. These factors are listed in Table 19.  We  believe that until these
factors are  reasonably  well established,  the disposal of oily wastes by land farming must
be very carefully controlled and safeguarded against any unforeseen environmental hazards.

-------
  278


 ECONOMICS OF OILY WASTE DISPOSAL BY LAND FARMING

    At present it appears that land farming represents a relatively inexpensive disposal
 method.  An example of what costs might be are shown in Table 20.  We are aware that
 costs will vary widely with land cost,  labor costs, amount of analytical work required, type
 of oily waste applied and other factors.  In this example we have used a land cost of
 $l,000/acre, costed borrowed capital at 10% for  10-years and assumed very small fertilizer
 requirements.  It is obvious that we need more  clearly defined bases for establishing oily
 wastes land farming costs.

    In  summary, we are encouraged by data that indicates land farming to be an effective,
 safe and economic disposal method for at least  some oily wastes.  We believe,  however,
 that considerably more investigative effort must be expended before the concept can be
 safely used on a wide scale basis.  Finally, it is imperative that sound data be  utilized to
 construct reasonable but precise ground rules for the effective, safe and economic use of
 land farming  for oily wastes disposal.

 REFERENCES

 (1) Atlas,  R. M. and R. Bartha,  "Simulated Biodegradation of Oil Slicks Using Oleophilic
     Fertilizers," Environmental Science  & Technology, Vol.  7,  No. 6, pp. 538-541 June
     (1973).
 (2) Bartha, R.,  "Biodegradation of Oil  on Water  Surfaces," NTIS, AD-DOOO 722, March 13
     (1975).
 (3) Kator, H., et al,  "Microbial Degradation of  a Louisiana  Crude Oil In  Closed Flasks and
     Under Simulsated Field Conditions," Proceedings of the 1971  Conference on Prevention
     and Control of Oil  Spills, Washington,  D. C.
 (4) Crow,  S. A., et al, "Microbiological Aspects of Petroleum Degradation in the  Aquatic
     Environment,"  La Mer, Bulletin  de  la Soc. Franco-Japonaise d'Oceanographic, Vol.
     12, No. 2, pp. 95-112, (NTIS COM-75-10298, May 1974) (1974).
 (5) Alexander M., Introduction  to Soil  Microbiology, John Wiley & Sons, Inc.,  New York
     (1961).
 (6) Lai, M.G., etal,  "Determination of the Molecular  Solubility of Navy Oils in Water,
     NTIS, AD-784414, June 6 (1974).
 (7) Ministry  of Defense Working Party,  "The Fate of Oil Spills  at  Sea," NTIS, AD-763042
     (1973).
 (8) Kincannon,  C. B., "Oily Waste Disposal by  Soil Cultivation Process," Environmental
     Protection Technology Series, EPA-R2-72-110, December (1972).
 (9) Raymond, R.  L., et al, "Assimilation of Oil  by Soil  Bacteria," Preprint No.  24-75,
     API Division of Refining Meeting, Hyatt Regency O'Hare, Chicago,  III.,  May (1975).
(10) Francks,  H. C., et al, "Disposal of Oil Wastes by Microbial Assimilation," NTIS, Y-
     1934,  May (1974).

-------
                                                                                  279

DISCUSSION

John Smith:  This is not a particular question, but just a general comment concerning the
sol ids wastes, the problem facing the petroleum industry.  As most of you know the EPA
has  contracted a study for the potentially hazardous waste and that document is being
prepared to reference the guidelines.  They have now advertised for  contractual bids
another study in the same line reference  to the investigation of various methods of elimina-
tion of these hazardous wastes.  The first study was predicated on the incineration, land
farming, etc. and now they are coming out asking contractors to look into other methods
of waste disposal from petroleum operations.  The point being here that  EPA is very aware
of and is going  to be spending money reference to solids waste disposal so it is a very timely
topic.

Robert Huddleston:  I will  speak for all of us  - I  hope they are successful in finding more
economical and acceptable disposal methods.

R. A. Farnham:  We have been doing some land  farming experiments and you mention soil
water contents.  Now you are talking weight percent water rather than  percent field
capacity, are you?

Robert Huddleston:  Yes, for the soils we have examined, the water holding capacities tend
to average about 60 weight %.  Thus, when these soils contain 30%  moisture, the moisture
content is condusive to good microbial  activity.

R. A. Farnham:  That's for clay?

Robert Huddleston:  Yes, clay type soils.

Anonymous:  For those of us who have trouble with numbers, your number describing Shell's
rate of oil degradation experience was  165 g oil  lost per kg of soil per year.  Was that
fertilized or unfertilized?

Robert Huddleston:  I used the fertilized  rate in the table.

-------
280
BIOGRAPHIES

   Lewis W. Cresswell  is a Research Scientist in
the Environmental Group of Continental Oil
Company's Research and Development Department.
He has a B.S.  degree in Chemical Engineering
from Mississippi State University.  Prior employ-
ment includes  Continental's Process Engineering
Environmental  Group and the U.S. Environmental
Protection Agency,  Denver Region.
   Robert L. Huddleston is supervisor of the
Environmental  Research Group in the Research and
Development Department of Continental Oil
Company, Ponca City, Oklahoma.  He holds B.S.
and M.S. degrees in Microbiology from the
University of Oklahoma and has taken additional
studies from Oklahoma University and Oklahoma
State University in Microbiology and in Environ-
mental Science. He is currently Vice President
of the Society  For Industrial Microbiology.  Bob
has been a  member of Conoco R&D for almost 20
years and has authored more than 25 papers
dealing with various aspects of industrial
microbiology and waste treatment.

-------
                                                                                         281
                           TABLE 1 "SOURCES OF OILY REFINERY WASTES'
Crude Oil Tank Bottoms
Slop Tank Bottoms
API, CPI Separator Bottoms
Biotreatment Solids
Coker Slowdown Solids
Water Impoundment Pond Bottoms
                                                               Typical Weight % Composition
Water
15
64
36
96
20
40
Solids
45
4
47
4
40
40
Oil
40
32
17
0.4
40
20
                         TABLE 2 "HISTORICAL DISPOSAL OF OILY SOLIDS'
                       Burning
                       Landfill
                       Surface Pitting
Road Surfacing
Ocean Dumping
Burying
                             TABLE 3 "MICROBIAL DISTRIBUTION IN SOIL1

                                 (TAKEN FROM ALEXANDER, 1961) (5)

                                     Organisms Per Gram of Soil
Depth
(cm)
3-8
20-25
35-40
67-75
135-145
Aerobic
Bacteria
7,800,000
1,800,000
472,000
10,000
1,000
Anaerobic
Bacteria
1,950,000
379,000
98,000
1,000
400

Actinomycetes
2,080,000
245,000
49,000
5,000
-

Fungi
119,000
50,000
14,000
6,000
3,000

Algae
25,000
5,000
500
100
-
                           TABLE 4 "HYDROCARBON SOLUBILITY IN WATER"

              (TAKEN FROM MING, 1974, AND "FATE OF OIL SPILT AT SEA," 1973) (6,7)

                                                             PPM in Distilled Water
                 Benzene
                 Xylene
                 Octane
                 Anthracene
                 C   Paraffin
                 Cnrysene

                 Gas Oil
                 Kerosene
                 Bitumen
           1800
            175
              1
              0.075
              0.002
              0.002

      0.0003 - 0.00000008
      0.0001 ,-0.2

-------
282
                           TABLE 5 "SOIL PARTICLE SIZE AND SURFACE AREA1

                                (TAKEN FROM ALEXANDER, 1961) (5)
                                    Diameter                    No.  Particles         Surface Area
                                       mm                       Per g Soil            sq cm/g
 Fine Gravel                       2.00-1.00                            90               11.3
 Sand                              1.00-0.05                 722-722,000          22.7-227
 Silt                               0.05 - 0.002                    5,780,000                454
 Clay                                    0.002               90,300,000,000             11,300

 *Based on spherical shapes and maximum diameters
                               TABLE 6 "LAND FARMING OILY WASTE"

                 Waste Analysis
                 Land  Selection and Preparation
                 Waste Application
                 Fertilizer Application
                 Cultivation and Other Care

                 Oil + 00 + N + PO.     Microbial       Microbial Cells + COO + H0O
                        2         4     Activity                           2    2
                 TABLE 7 "QUESTIONS ABOUT LAND FARMING OILY WASTE SOLIDS'

 1 .  What type of waste can be applied?
 2.  How long does it take for the oil to be decomposed?

 3.  Is there any oil residual? How much? What is it?
 4.  Are there any environmentally harmful materials formed?
 5.  Will the oil  or other waste components migrate through the soil?
 6.  Mechanically, how is the waste best applied and blended  with the soil?
 7.  How much fertilizer must be  added?
 8.  Is subsequent cultivation required?
 9.   Should plant growth be established on the site?
10.   How do different climatic conditions affect the process?

-------
                                                                                        283
Analysis
TABLE 8 "BILLINGS REFINERY WASTE POND BOTTOMS"




WT %         Metal Content      mg/kg     Metal Content
m
g/kg
	 •- 	 	
Extractable Oil 15
Water 72
Solids 13
Total Organic Carbon 10



Silicon
Calcium
Chromium
Sodium
Barium
Lead
Vanadium
TABLE 9 "TREATMENT OF BILLINGS

0 1 2
Cultivation R(l) R R
Fertilization 1
Sampling (3) 115
(1) R= Rototiller
(2) D = Disc
(3) Numbers refer to samplings per month
TABLE 10 "INITIAL BILLINGS LAND
Analysis WT %
Extractable Oil 0.93
TOC 1.2



Months
3 4


4 2



12,900
10,900
1,900
810
370
50
10
Iron 12,000
Aluminum 4,300
Magnesium 890
Zinc 690
Copper 360
Nickel 10

LAND FARM TEST PLOTS"
of Exposure
7 12


1 1



FARM SOIL-WASTE COMPOSITION
Analysis
Nitrogen
Iron
Chromium
Zinc
Lead
mg/kg
130
720
115
41
3

13 14 15 17 18
D (2) D D

4 413



DUE TO WASTE ADDITION"
Analysis mg/kg
Phosphate 20
Calcium 650
Sodium 50
Copper 22


-------
284
                             TABLE 11 "OIL LOSS - BILLINGS LAND FARM"

                             18 Months                          Annual ized
                           15,840 Ib/Acre Soil
                           7.2gAg Soil
                                  10,560 Ib/Acre Soil
                                  4.8gAgSoil
                          TABLE 12 "PONCA CITY REFINERY TEST OILY WASTE1
 Analysis

 Extractable Oil
 Water
 Solids
 Total Organic Carbon
 Analysis

 Extractable Oil
 Water
 Solids

 Total Organic Carbon
WT%
Metal Content
fig/kg
Metal Content   mg/kg
39
24
37
53





TABLE
WT% of
4.8
3.0
4.6
6.5





Silicon
Aluminum
Lead
Sodium
Zinc
Barium
Copper
Manganese
Vanadium
13 "PONCA CITY REFINERY
Soil Metal Content
Silicon
Aluminum
Lead
Sodium
Zinc
Barium
Copper
Manganese
Vanadium
21,000
3,200
1,400
870
400
240
150
130
21
WASTE IN
mg/kg of
2,600
390
170
107
50
30
18
16
3
Calcium 13
Iron 3
Magnesium
Potassium
Chromium
Strontium
Titanium
Nickel
Cadmium
SOIL"
Soil Metal Content
Calcium
Iron
Magnesium
Potassium
Chromium
Strontium
Titanium
Nickel
Cadmium
,000
,000
880
480
370
180
150
21
3

mg









                                                                  mg/kg of Soil

                                                                     1,600
                                                                       370
                                                                       108
                                                                       60
                                                                       46
                                                                       22
                                                                       18
                                                                        3
                                                                        0.4

-------
                                                                                           285
    TABLE 14 "OIL COMPOSITION IN PONCA CITY REFINERY WASTE SUBJECTED TO LAND FARM STUDY'

                         Analysis                           WT %

                         Paraffins                           21
                         Aromatics                         49
                         Resins-asphaltenes                  30
                             TABLE 15 "PONCA CITY OILY SAND WASTE"

                         Analysis                          VVT %

                         Extractable Oil                    27
                         Water                             4
                         Solids (sand)                       69
                         Total Organic Carbon               10

                         Metal Content   mg/kg      Metal Content         mg/kg

                         Sodium         2,000      Calcium               1,000
                         Magnesium        260      Chromium                90
                         Copper            90      Lead                    90
                         Vanadium          90      Nickel                 130
       TABLE 16 "CORE ANALYSES - OILY SAND LAND FARM TEST PLOT AND CONTROL AREAS"
                                           Sampling A	              Sampling B	
                       Core Depth       Control        Test Area           Control        Test Area

Extractable Oil           0-6"          0.15           4.79                0.11           6.35
   (Wt. %)               6-9"          0.20           0.19                0.05          0.10
                        9-12"         0.09           0.17                0.02          0.05
                       12 - 15"         0.09           0.08                0.03          0.05
                       15 - 18"         0.11           0.18                0.03          0.05
                       18 - 24"                        0.14                0.05          0.05
                       24 - 30"                        0.06

Organic Carbon           0-6"          1.90           5.74                1.48          6.15
   (Wt. %)               6-9"          0.79           0.79                0.87          1.53
                        9- 12"         0.26           0.31                0.93          1.04
                       12 - 15"         0.29           0.37                0.81           0.91
                       15 - 18"         0.25           0.60                0.62          0.79
                       18 - 24"                        0.78                0.52          0.60
                       24 - 30"                        0.61                              0.56

-------
286
                 TABLE 17 "OIL LOSS % AND RATE - PONCA CITY OILY SAND LAND FARM STUDY"

                              Initial %             % Oil Loss             Mass of Oil Lost
                              Oil In Soil              Per Year                Per Year
                                                     25                 27,500 Ib/acre soil (12 g/kg soil)

                                                     22                 38,720 Ib/acre soil (18 gAg soil)
                      TABLE 18 "REPORTED RATES OF OIL LOSS DURING LAND FARM STUDIES"

                   (TAKEN FROM SUN, SHELL, UNION CARBIDE, NEW ZEALAND DATA) (8,9,10)

                                                            Gm Oil Lost     Approximate Initial Oil in Soil
                                                            Per kg/soil/year      Concentration (wt. %)

     Refinery Oily Waste
            CONOCO -  Billings                                      7                     1
                        Ponca City                                   9                     5
            Shell -       Houston                                    165                    10

     Petroleum Crude Oil
            CONOCO - Ponca City                                  17                     5
            Sun -        Marcus Hook, Tulsa, Corpus Christi             14+2                 1
     Lube Oil Waste
            Sun -        Marcus Hook, Tulsa, Corpus Christi            18+5                5
            New Zealand                                           27~                  2
                                                                   38                     5
                                                                   69                     7
                                                                  454                     9

     Waste Vacuum Pump Oil
            Union Carbide - Oak Ridge                              171                     7

     #6 Fuel Oil
            Shell -          Houston                                311                    10
            Sun -            Marcus Hook, Tulsa, Corpus Christi        20+7                 2

-------
                                                                                           287
             TABLE 19 "FACTORS TO ESTABLISH PROPER USE OF LAND FARMING"
1.   Firm establishment that oily waste is not mobile with respect to waste composition,  soil composition
    and climate.

2.   Final extent of waste oil  decomposition in the soil as related to oil composition.

3.   Demonstration that no dangerous substances are present or produced when oily wastes are land
    farmed.

4.   What types of vegetation can be grown on oily waste land farms and what effects does such growth have
    on rate and extent of oil degradation.

5.   Need precise  nitrogen and phosphate requirements for oily wastes land farming.

6.   Establishment  of ground rules for safe and economic application of land farming for oily wastes
    disposal.
               TABLE 20 "ESTIMATED COST OF LAND FARMING OiLY WASTES"

                          100 tons of waste
                          25% oil
                          Application limited by oil degradation
                          Oil degradation rate - 15 g/kg soil/yr.
                          Farm life - 10 years

                                                  Annual Cost ($)                  % of Total

Land                                                 355                            16
Site Preparation                                      540                            24
  Diking, Fencing, Cultivation
Waste Hauling (5 miles)                               275                            13
Waste Spreading                                      160                             7
Fertilizer, Spreading                                  115                             5
Cultivation of Soil-waste (4/yr)                        160                             7
Samples - Analytical Support                          600                            27

TOTAL COST                                      $2,205

Cost per ton of waste:  $22.05

-------
288
                      FIGURE 1


Photograph of  Billings'  Land Farm Area Prior to Waste

                     Spreading
                                 FIGURE  2


            Photograph of Billings'  Land Farm Area After Waste  Tilled In



                                                   .-vf
                                 "  - .
                                            ..
                                                       • :.    ~-;r-"-«»-
                                           • "    .-»       ^* Tit^ •

                                                           ^£*&&,
                                  . .     ,
                                         ,
                                             "
                                                ^  .-;,
                                                    *

-------
o
                                       FIGURE 3

                   Photograph of Billings' Land Farm Area After Waste
                                  Tilled  In -  Close up
                                                                           289
H
Z
UJ
(A
UJ
or
Q.
o
      100
      80
60
      40
      20
                                      FIGURE  4
                          BILLINGS WASTE OIL LAND-FARM STUDY
                         ASPHALTICS
                                                               4
                                                                                100
                                                                          80   =1
                                                                               O
                                                                          60   Z
                                                                               UJ
                                                                          40   0=
                                                                               u.
                                                                               o

                                                                          20   ^
                                      10
                                       12    14
                                                         16    18
                               MONTHS  OF EXPOSURE
 u
 o
 ^**
 a.

 UJ


 ir
 <
      30
20
       10
     -IQL

-------
290
 o
 UJ
 a:

 o
 o
100




80




60




40




20
                                FIGURE  5


                       TOO - BILLINGS  LAND FARM
                                  I     I
         0
                           8    10     12
14
16    18
                           MONTHS OF EXPOSURE
                                 FIGURE 6

            Photograph of Ponca City Refinery Waste Land Farm Area

                             Prior to Preparation


                         -  •


                   %* -,** ...  .
                                                                          3
                                                                          •

-------
                    FIGURE 7

Photograph of  Ponca City Refinery Waste Land Farm Area
                After Waste Addition
                                                             291
                                       -  -*w
                                       -*-w~   .^.
'iS
                                   •
                                                     -.
                                                       '
                    FIGURE 8
Photograph of Ponca City Refinery Waste Land Farm Test
                 Area - Wheat Growth

-------
    292
 us
 CO
 UJ
 cr
 Q.
 o
 00
                    FIGURE 9
Photograph of Ponca City Oily Sand Land Farm Test
                    Sys tern
      W:x?^jr-jag    ^   _   ^
                                     FIGURE  10
                        PONCA  CITY CRUDE  OIL LAND-FARM STUDY
       100
                                                                 -ilOO
                                                                                   80
                                                                                      J
                                                                                      0
                                                                   60
                                                                                   40
                                                                     U.
                                                                     0
        20 -
                                                                  20
                                                             18   20   22
O
o
       30
       20
                                 MONTHS  OF EXPOSURE
x

o

(D
      - 10

-------
             SESSION  V


"ACTIVATED CARBON TREATMENT"


   Chairman:


   Nicholas D. Sylvester

   Professor of Chemical Engineering

   University of Tulsa,  Tulsa,  Oklahoma


   Speakers
   Davis L. Ford

   "Current State of the Art of Activated Carbon Treatment"


   C.T.  Lawson

   "Cautions and Limitations on the Applications of Activated
   Carbon Adsorption to Organic Chemical Wastewaters"


   Joyce A. Rizzo

   "Case History: Use of Powdered Activated Carbon  in an
   Activated Sludge System"


   Leon H. Myers

   "Pilot Plant Activated  Carbon Treatment of Petroleum
   Refinery Wastewater"


   M.A. Prosche

   "Activated Carbon Treatment of Combined Storm and  Process Waters'
                   293

-------
 294
BIOGRAPHY
Nicholas D. Sylvester
    Nicholas D. Sylvester is Professor of Chemical
Engineering and Director of the Environmental Protection
Projects program at the University of Tulsa.  Dr.
Sylvester received his B.S. degree from Ohio University,
and his Ph.D. from Carnegie-Mellon University both in
Chemical  Engineering.  Before coming to the University
of Tulsa he taught at the University of Notre Dame.
Nick  is a  member of the American Institute of Chemical
Engineers, Society of Petroleum  Engineers, American
Chemical  Society, Society of Plastics Engineers, Society
of Rheology and the American Society of Engineering
Education.

    Professor Sylvester is currently  conducting research
in the following areas:  two-phase flow,  drag reduction,
environmental protection, chemical reaction engineering
and improved oil recovery.  Dr.  Sylvester has more than
50 technical publications and has been principal investi-
gator  of funded research totalling nearly $600,000.

    In addition to his professional interests, Nick is
active in youth athletic programs, having coached junior
high basketball and elementary school  baseball  and
soccer. He is also an avid, though, inept, golfer.

-------
        "CURRENT STATE OF THE ART OF ACTIVATED CARBON TREATMENT"

                             Davis L. Ford, Ph.D., P.E.
                                 Senior Vice President
                       Engineering-Science, Inc., Austin,  Texas

     The treatment of wastewaters using activated carbon has received wide attention for
several  years - more recently catalyzed by the development of effluent quality guidelines
pursuant to Public Law 92-500.   These effluent quality criteria and the accompanying
development documents prominently mention carbon as an applicable and attractive
treatment concept, particularly  in the Best Available Treatment Economically Achievable
(BATEA) process mode currently  stated as  necessary to produce the 1983 quality level
objective.  Until recently,  most of the literature has dealt with exploring theoretical
concepts and documenting experimental results.  As more  information on full-scale opera-
tions is  becoming available, however, it  is  considered appropriate to review the current
state of the  art of activated carbon treatment - both in  municipal and industrial sectors.
It is the purpose of this  treatise,  therefore,  to present pertinent and current information
relative to the  activated carbon treatment of municipal and industrial wastewaters.  A
brief discussion of adsorption concepts and carbon characteristics also is  included.

ADSORPTION  CONCEPTS AND THEORY

     Molecules are held together by cohesive forces ranging from strong valence bonds to
the weaker van Der Waals forces of attraction.  These attractive forces are satisfied  in the
solid phase interior molecules, having the ability to capture  certain fluid molecules  as
they contact the surface,  van Der Waals forces are the bases for the adsorption of waste-
water constituents onto  carbon which has  been activated to maximize this interphase
accumulation of liquid constituents at the surface or interphase of the solid phase.

     The rate at which substances are removed from the  liquid phase (adsorbate) to the
solid phase (adsorbent) is of paramount importance when evaluating the efficacy of acti-
vated carbon as a wastewater treatment process.  Unfortunately, the task of quantifying
the many forces acting at the solid-liquid interface is a formidable one.   Developing a
mathematical expression which describes the dynamic phenomenon occurring in a  continuous-
flow/fixed-bed reactor  has been difficult because of multi-variable influences.  The overall
adsorption rate  represents the combined effects of diffusion through a  laminar layer of fluid
surrounding  the constituent, surface  diffusion, and adsorption on the internal pore surfaces.
Most mathematical solutions for  equations which describe concentration/time profiles are
limited  to the special case in which  only  one of these phenomena controls the overall rate
of adsorption (1).

     One expression for  the  continuous-flow regime assumes the diffusion of the constituent
through  the  liquid phase and through the pores of the carbon  (which are rate-limiting),
then combining these  resistances in  an overall mass coefficient term.  Using this rationale:
                     =  k2r(Cs'C)
                                      295

-------
296

 where:
              q         =    flow rate
              C         =    concentration of the adsorbate
               s
              D         =    adsorbent bed depth
              C         =    equilibrium adsorbate concentration
              k r       =    overall mass transfer coefficient

     A more convenient expression of Equation 1  is in terms of the adsorbate rate with
 respect to the weight of the carbon in  the columns:

              qdC/dM =  ly/X (Cs-C)                                     (Eq. 2)
 where:
              M        =    weight of the  carbon in the column
              X         =    packed density of the carbon in the column

     Another proposed model  predicts four successively decreasing adsorption rates would be
 observed as the adsorption proceeds to equilibrium.  The  initial rate would be limited by
 the  rate of adsorbate  transfer across the film layer, film diffusion, or, if sufficient turbulence
 existed,  control  would be exerted by the combined rate of external surface adsorption and
 macropore  filling.  After the external  surface adsorption  capacity was exhausted, there would
 exist three secondary adsorption rates controlled, respectively, by the filling of the macro-
 pore (an  effective radius of 5,000 to 20,000 A°), the transitional pore (20 to 100 A°), and
 the  micropore (10 to 20 A  effective radius).  This model is illustrated in Figure 1  (2).  It
 is inherent in this model that the intraparticle transport occurs as a series of adsorption/
 desorption  steps,  each linear with respect to time, and  their summation resulting in a time/
 linear  function.

     The  development of adsorbate removal  kinetics on a  batch basis can  be used to appro-
 ximate carbon effectiveness and predict organic residual  levels.  The  adsorption isotherm
 is used for  this purpose and is defined as a  functional expression for the variation of adsorption
 with concentration of adsorbate in bulk solution at a constant temperature.  The isotherm is
 expressed in terms of  removal of an impurity - such as BOD, COD, and color - per unit
 weight of carbon  as a function  of the equilibrium impurity remaining in solution.  Linear
 plots as shown in  Figure 2 can be expressed in terms of the empirical  Freundlich equation.
 This expression relates the amount of impurity in  the adsorbed phase to that in solution:


              -fi        =  KC1/n                                            (Eq. 3)

 where:
              X         =    amount of impurity  adsorbed
              M         -    weight of carbon
              C         =    equilibrium concentration of impurity in solution
              K,n      =    constants

     The  Freundlich isotherm  is valid within the context of a batch test for pure substances
 and  some dilute wastewaters.  As shown in Figure 2, its application is limited in certain

-------
                                                                                  297


cases when a significant portion of the organic impurities are not amenable to sorption,
resulting in a constant residual, regardless of the carbon dosage.

     The constants "n" and "K" can be used to define both the nature of the carbon and
the adsorbate. A high  "K" and "n" value, for example, indicate good adsorption through-
out the concentration range studied.  A low "K" and  "n" value would infer  low adsorption
at dilute concentrations with high adsorption  at the more concentrated levels.  Variations
of the constants for selected wastewaters are shown in Table 1  (3).

FACTORS WHICH INFLUENCE ADSORPTION

     There are many factors which influence both the rate and  magnitude of  adsorption -
underscoring the  difficulty in developing predictive models which would apply to all
complex wastewaters.   A brief discussion of the more important factors is presented  herein.

     Molecular Structure.  The molecular structure, or nature of the  adsorbate, is partic-
ularly important  in dictating the degree of adsorption that can actually occur.  As a rule,
branched-chain compounds are more sorbable than straight-chain compounds, the type and
location of the substituent (functional) group  affects adsorbability, and molecules which
are low in polarity and solubility tend to be preferentially adsorbed.  Unless the screening
action of the  carbon pores actually  impedes,  large molecules are more sorbable than small
molecules of similar chemical nature.  This is attributable to more solute chemical bonds
being formed, making desorption more difficult.

     Inorganic compounds demonstrate a wide range of adsorbability.  Disassociated salts -
such as potassium chloride and sodium sulfate - are essentially nonsorbable.   Mercuric
chloride and ferric chloride are relatively sorbable, and iodine is one of the most adsorbable
substances known.  Generally, however, a significant reduction  in  inorganic materials is
not expected  in carbon systems.

     Organic compound sorbability can be classified to some extent.   Primary alcohols and
sugars,  for example, are resistant to adsorption,  while ethers and certain organic acids
are highly sorbable.  Recently published experimental data presented in Table 2 are
indicative of the sorbability of many organic  compounds (4).  Additional  sorbability data
conducted independently are presented in Table 3.

     Solubility.   An increase in solubility acts to oppose the attraction of the adsorbate to
carbon.  Thus, polar groups which have a high affinity for water usually diminish adsorption
from aqueous solutions.  Conversely, the greater adsorption of the higher aliphatic  acids
and alcohols is attributed in part of their relatively lower solubility in an aqueous solution.
There are exceptions to this, as in the case of the highly soluble chloracetic acid (5).

     lonization.  lonization is generally adverse  to adsorption  by carbon as strongly-
ionized materials are poorly adsorbed.   Hydrogen ions, which  are significantly adsorbed
under some conditions,  would be an exception to this.  Some negative ions, therefore,  are
more sorbable when associated with  hydrogen  ions.  For this reason,  mineral acids - such
as sulfuric acid - are sorbable at higher concentrations.

-------
298


     A change in ionization can drastically affect adsorption.  A low pH, for example,
promotes the adsorption of organic acids whereas a high pH would favor the adsorption of
organic bases.  Phenol adsorbs strongly at neutral or low pH while the adsorption of the
phenolate salt at a high pH is poor.  The optimum pH is therefore solute-specific and must
be determined for each wastewater.

     Temperature.  As adsorption reactions are generally exothermic and high temperatures
usually slow or retard the adsorption process,  lower temperatures have been reported to favor
adsorption  (1,5).  Very little  information has  been presented, however, which documents
significant shifts  in adsorbability within the temperature range of 65 F to 90 F (typical of
most wastewaters). Lower temperatures should increase adsorption, but the effect in aqueous
solutions is  very small.

     Adsorption of Mixed  Solutes. Most wastewaters contain a myriad of compounds which
may mutually enhance, interfere, or act independently in  the adsorption process.  Factors
which affect overall adsorption of multiple adsorbates include the relative molecular size
and configuration, the relative adsorptive affinities, and the relative concentrations of the
solutes (1).   Predictive models obviously require  validation for complex  wastewaters, as
extrapolation from investigations using synthesized wastes containing controlled concen-
trations of selected adsorbates may not reflect all of the interactions occurring in the waste.

     A summary of the factors  which potentially influence adsorbability is presented in
Table 4.

PROPERTIES OF ACTIVATED CARBON

     Activated carbons are made  from a variety of materials including weed, peat,  lignin,
bituminous coal,  lignite,  and petroleum residues.  Granular carbons produced  from medium
volatile bituminous coal or lignite have been  most widely applied in the treatment  of
wastewater  as they are relatively inexpensive  and readily available.  The activation of
carbon is essentially a two-phase process which includes burning off the amorphous decom-
position products and enlarging the pores in the carbonized material (6).  The burn-off, or
carbonization, phase involves drying  the carbon  at approximately 170 C,  heating the
material to  270  C to  280  C with the evolution of carbon monoxide, carbon dioxide, and
acetic acid, and, finally, completing the carbonization process at a temperature of 400 C
to 600 C.   The yield following carbonization  is approximately 80 percent. The intermediate
product is then activated  by using carbon dioxide or steam at a  temperature of  750  C to
950  C, burning off decomposition products, exposing and widening the pores in the development
of macroporous structure.   In the activation process, the kind  of adsorptive powers  developed
are determined by:
     1 . the chemical nature and concentration of the oxidizing gas;
     2. the extent to which the activation is  conducted;
     3. the temperature of the reaction; and
    4. the amount and kind of mineral  ingredients in the  char  (5).

    The proper activation conditions  provide  an  oxidation which selectively erodes the
surface so as to increase the surface area, develop greater porosity, and leave the  remaining

-------
                                                                                  299
atoms arranged in configurations that have specific affinities.
     Activated carbons from coke,  coal, or lignite have specific properties, depending on
the material  source and the mode of activation.  Property standards are therefore helpful
in specifying carbons to perform  a specific task.

     As a rule, granular carbons  made from  calcined petroleum coke have the smallest
pore size, the  largest surface area, and the highest bulk density.  Lignite carbon has the
largest pore size, least surface area, and the lowest bulk density.   Bituminous coal has a
bulk density  equal to that of petroleum coke and an average pore size and surface area
somewhere between  those of petroleum coke and  lignite-based carbons (7).  A brief
description of carbon properties follows:

     Total Surface Area;  This is  the surface area of carbon expressed in square meters per
     gram, normally measured by the adsorption of nitrogen gas by the BET method (8).

     Carbon  Density: Apparent density is the weight  in grams of one ml of carbon in air.
     Bulk density, backwashed and drained,  is often used and is usually expressed in
     pounds per cubic foot.

     Particle Size Distribution:  The particle size distribution is critical in terms of
     hydraulic  loading and backwash rates.  Commonly manufactured particle size ranges
     for granular  activated carbons expressed  in limiting U.S. Standard Sieve Sizes
     include  8  x  16, 8 x 30, 10 x 30, 12 x 40, and 20 x 40.  Effective sizes (sieve
     opening at which 10  percent of the material passes) range from 0.55  mm to 1 .30 mm.
     In general,  the uniformity coefficient (the millimeter opening at which  60  percent
     of the material  passes divided  by the millimeter opening at which 10 percent of
     the  material passes) for granular activated carbon should not exceed  2.1.

     Adsorptive Capacity:  The best measure of adsorptive  capacity  is the effectiveness of
     the  carbon in removing the critical constituents (BOD, COD, color, etc.) from the
     wastewater in question.  Various tests, however,  have been developed to give
     relative removal capacities  of activated  carbon under specific  conditions.   Phenol
     number is  used as an  index of a carbon's  ability to remove taste and odor compounds;
     tannin is representative of organic compounds added  to water by decayed vegetation;
     and iodine and  molasses numbers are used to show if a carbon is activated.   The
     iodine number,  defined as the milligrams of  iodine adsorbed by one gram of carbon
     (with an iodine  concentration  in the residual  filtrate is 0.03) is probably the most
     widely used  method of expressing carbon  capacities.   It generally  can be correlated
     to the ability of an activated carbon to adsorb low molecular weight substances
     while the molasses number correlates the  carbon's ability to adsorb higher molecular
     weight substances. The iodine number measures the micropores having an effective
     radius of less than 20 angstroms and the molasses number measures the transitional
     pores ranging from 20 to 500 angstroms.

     These are the principal general parameters used in specifying carbons, the  objective
being the selected carbon

-------
300

     1 .  has the adsorption  capability to meet the effluent requirements;
     2.  incurs minimal losses occurring during carbon transport and regeneration;
     3.  has good hydraulic characteristics with respect to head  loss or pressure drop; and
     4.  represents the most cost-effective media to accomplish the prescribed task.
 These, of course, represent general parameters of specification and should be augmented by
 test data and process requirements developed from bench- or pilot-scale evaluation using
 representative wastewater samples and selected carbons.

     Comparative properties of the most widely used carbons in wastewater treatment -
 those  from lignite and bituminous coals - are shown in  Table 5 (6); a broader presentation
 of properties from commercial carbon  manufacturers also is tabulated in this Table (9).
 Typical properties of a petroleum-base powdered carbon developed but not yet marketed by
 AMOCO are shown in Table 6.

 REGENERATION

     Because of economic and solid waste  disposal considerations,  it is generally more
 feasible to  regenerate spent carbon for subsequent reuse than to  dispose of it. In the
 regeneration process, the objective is to remove from the carbon porous structure the
 previously-adsorbed materials,  thus reinstituting its ability to adsorb impurities.  There are
 several modes of regeneration which can be applied including thermal, steam treatment,
 solvent extraction, acid  or base treatment, and chemical oxidation.  Of these, only
 thermal regeneration using  a multiple-hearth or rotary-tube  furnace is widely applied in
 wastewater treatment.  The discussion on regeneration  therefore centers around thermal
 treatment.

     Thermal regeneration refers to the process of drying, thermal desorption, and high
 temperature (1,200 F to  1,800  F) heat treatment in the presence of limited quantities of
 oxidizing gases such  as water vapor, flue  gas, and oxygen (10).  Multiple-hearth furnaces
 are  the most commonly used, although rotary kilns or fluidized-bed furnaces are occasionally
 applied.

     The sequential stages in thermal regeneration are shown in Table 7.  Of these steps,
 the  gasification stage is the most critical.  The system  should  be controlled  in order  to
 selectively gasify the sorbed organic material while minimizing  the gasification of the
 carbon structure.  Basically, there are three major variables involved  in thermal regenera-
 tion.  These include  furnace temperature, residence time, and the carbon loading.  Of
 these  three, furnace  temperature is controllable, although it may take several hours to
 adjust, the residence time can be changed by varying the rabble-arm speed in a multiple-
 hearth furnace, or the rotation  rate and slope of the tube in a rotary kiln.  Very little  can
 be done to  change the carbon loading, which affects the severity of regeneration  required.

     The  multiple-hearth furnace is the most commonly used system for granular carbon
 regeneration.  A schematic diagram of a typically-designed multiple-hearth furnace is
 shown in Figure 3.  The wet spent carbon  is added at the top of the furnace and drops from
 hearth to hearth, being  raked along by the rabble arms. The  temperatures shown  for the
 various hearths are typical  gas temperatures for granular carbon  regeneration.  It  was

-------
                                                                                  301

found that the addition of steam on hearths four and six gave a more uniform distribution
of temperatures throughout the furnace.  The effect of steam  is to reduce the apparent
density and  increase the iodine number of  the regenerated carbon (6).  Normally, about
one pound of steam per pound of carbon  is used.  Natural  gas or fuel oil  is usually added
to supply the auxiliary heat.  Although the fuel requirement  varies, generally about
3,000 BTU/lb of carbon and 1,300 BTU/lb of steam generated is required.

     Rotary kiln furnaces are not being installed in new facilities but many are still in use.
The kilns are usually direct-fired,  counter-current units with steam injection with up to
10 percent excess  air.  Again, natural gas or fuel oil is commonly used as auxiliary fuel.
The heat efficiency of rotary kilns  is less than that of multiple-hearths.

     Fluidized-bed systems for  powdered carbon regeneration have been evaluated on a
pilot scale and appear to have promise (11, 12, 13).  In this process a bed of inert
granular material such as sand  is fluidized by the  upward flow of hot gases and the wet
spent carbon is injected directly into the bed.  The inert bed particles provide a reservoir
of heat which is rapidly transferred to the  spent carbon particles.  The heat economy of
the fluidized-bed  furnace is less favorable than the multiple-hearth furnace or the rotary
kiln - providing afterburners are not required.

     There is considerable debate as to the effect  of regeneration on carbon capacity,
carbon losses during the regeneration cycle, and hydraulic characteristics of the regen-
erated carbon. Much of this is focused on the regeneration effects on lignite as compared
to bituminous carbon.   The  manufacturers  of lignite claim  that,  after a number of regen-
eration cycles, both carbons tend to become more like  each other  and perform the same (9).
Specifically,  lignite - while softer and lighter than bituminous  - is less susceptible to
change during regeneration and can be regenerated under less severe conditions  (lower
temperature and shorter residence time).   The bituminous manufacturers note that lignite
carbon losses per cycle are  higher than bituminous, indicating a higher operating cost,
and bituminous adsorption capacities based on molecular weights ranging from phenol at
94 to dextran at 10,000 are significantly higher than lignite  (14).  It is not the intent in
this writing  to favor one over the other, but simply to present information from manufacturers
of both types.  Based on discussions with independent evaluators of bituminous and lignite,
several trends in thinking evolve.  First,  few full-scale plants using lignite are currently
in operation,  necessitating  an  evaluation  based on pilot-plant studies.  There appears to
be little firm evidence that bituminous carbon capacities are higher than those of lignite,
although screening tests should be conducted in any event to determine the optimum carbon
to perform a specific task.  Somewhat higher losses have been noted during regeneration
of lignite as compared to bituminous (six to nine percent for lignite, and four to seven
percent for bituminous over three cycles).   Lignite requires lower temperature for its
regeneration and more skilled  regeneration operation and  control.  There has been little
evidence to show significant hydraulic differences as demonstrated by head loss charac-
teristics in the two carbons  following regeneration. Carbon selection is  therefore predi-
cated on process test results, capital  cost  requirements, and an annualized cost evaluation.

-------
302

EXPERIENCE IN CARBON TREATMENT OF MUNICIPAL WASTEWATERS

     Treatment  of municipal wastewaters using activated carbon has been practiced on a
full scale for over 10 years.  For this reason, ample performance records are presently
available for a realistic evaluation of this application, both in the physical/chemical and
biological effluent polishing modes of treatment.  A summary of these systems  has been
tabulated elsewhere, but  it is the intent in this section to document selected municipal
facilities and discuss salient features of these systems which are pertinent to this overall
discussion on carbon.  The two effluent polishing systems selected  for discussion are the
South Lake Tahoe  and Colorado Springs facilities.  The physical/chemical discussion
centers around  the new plant being constructed for the Cleveland Regional Sewer District,
with the Garland, Texas;  Rosemount, Minnesota; and Pomona, California plants also
included.  A syllabus of these selected case histories is presented as follows.

     South Tahoe Tertiary Treatment System.  The first major and most widely publicized
application of activated  carbon in treating domestic wastewater was the system constructed
for the  South Tahoe Public Utility District in 1965.  The plant was conceived  to polish the
effluent from an existing activated sludge facility to a quality level which would have
little or no impact on the receiving  waters in an  ecologically-sensitive area.

     The facility consists of a chemical mix-coagulation, precipitation, and clarification
unit, an ammonia  stripping  tower (which  has been used infrequently), a recarbonation and
settling basin,  mixed-media filters, carbon adsorption, and final chlorination.  A simplified
flow diagram is shown in Figure 4 (15).  The water quality at various points  in the process
following 18 months of operation is presented in Table 8 (16).  The more recently reported
water quality at the various stages of treatment is presented in Table 9 (9).  The carbon
(bituminous) efficiency per  regeneration period at the South Tahoe Plant is cited in
Table 10 (6).

     It should be recognized that this facility was a demonstration project, involving expensive
capital  and annual expenditures.  With the possible exception of nitrogen, the effluent
quality should represent the best level obtainable in treating a domestic wastewater effec-
tively using maximum biological treatment polished by chemical treatment,  filtration, and
carbon adsorption.

     Colorado Springs Tertiary Treatment  System.  As at Tahoe, the Colorado Springs
treatment system involves chemical treatment, filtration, and carbon adsorption for polishing
a slipstream of  biologically treated effluent.  There are basically two regimes of historical
data from this facility. The first was when the slipstream was taken from an overloaded
trickling filter  effluent (the flow diagram schematically illustrated in Figure 5).  The second
regime follows  the addition of a  contact stabilization facility to the biological treating
component of the system.  This indicates  that the  influent to the polishing portion of the
system, the design data for  which are presented in Table  11, has a lower concentration of
biodegradable organics in the second regime than in the first (17).  This  is detected in a
general comparison of the effluent quality observed during the two periods.  Effluent  quality
data from a three-month  period  in 1972 indicate  an approximate level obtained during the
first  regime.  The  average quality values for the  reactor-clarifier influent and effluent,

-------
                                                                                 303

filter effluent, lead carbon column effluent, and polishing carbon effluent are tabulated
in Table 12.  The data averaged during the second regime of operation are shown in
Table 13 (17). Considerable improvement is noted in the comparison, underscoring the
need for adequate biological pretreatment if the tertiary system is to realize maximum
performance.  It  is noted that the Tahoe and Colorado Springs  (second regime) effluents
are relatively similar - both plants have adequate biological pretreatment.  The  deteriora-
tion of effluent quality when this  is not the case is evidenced in Table  12.

     The Colorado Springs facility is using a bituminous coal activated  carbon, using a
multiple-hearth furnace regeneration system.  As in the other carbon systems, there has
been sulfide generation problems in the carbon vessel.  This  problem has been ameliorated
by adding a copper sulfate solution (100 mg/l as CuSOJ  to the top  of the carbon columns
and  altering the backwashing frequence to remove biological solids.

     Cleveland Regional Sewer District Physical/Chemical System.  A physical/chemical
wastewater  treatment plant of which activated carbon is an  integral  part is presently  being
constructed for the Cleveland Regional Sewer District (CRSD) at the Cleveland Westerly
plant site.  This system is designed to receive an average flow of 50 MGD with peak flows
of 100 MGD.  A flow diagram of  this system is shown in Figure 6.   Extensive pilot-plant
treatability studies were conducted prior to finalizing the design as  there is a significant
industrial contribution in the raw  waste load (18).  Moreover,  assurance of meeting the
NPDES  requirements of 20 mg/l  BOD (30-day average) and  30 mg/l  (seven-day average)
was  required.  These studies, combined with those more recently conducted by CRSD,
provided some interesting results with respect to carbon adsorption.

     The raw waste load of the Westerly collection system documented during the two
pilot studies (1970-71  and 1974-75) is shown in Table 14.  The strength of this wastewater,
combined with the organic nature of the industrial component, created some difficulty in
meeting  the required permit levels during the pilot-plant  evaluation.  Although these
results could be overly-pessimistic for various reasons, CRSD evaluated several process
alterations with the objective of improving process efficiency and effluent quality.  Most
of this investigation centered around the use of ozone injection - both  as a post-carbon
and  as a precarbon mode of operation.  The postozonation step was originally conceived
as strictly a disinfection step, although some further reduction of BOD  through ozone
oxidation was anticipated.  However, studies indicated the ozone requirement for disin-
fection to the required  level of  200 fecal coliforms per 100 ml  ranged from 2.6 to 10.9
mg/l (19).  The higher doses were attributed to the ozone demand from the sulfides
generated in the  carbon columns and ferrous bicarbonate resulting from iron slugs in the
influent.  This ferrous species was solubilized in the recarbonation step and was not
altered while  passing through the  carbon columns.  As the higher demands of ozone are
economically  prohibitive, reducing  the ozone-demanding constituents or looking closer
at chlorine was merited.  Moreover, postozonation did not improve  organic effluent
quality - as indicated  in Table 15.  Ozonation prior to the  carbon columns was then
applied by CRSD  with encouraging results.  This quality improvement in terms of a
frequency analysis of carbon column effluent BOD is shown  in  Figure 7.  The same trend
in terms of COD  is shown in Figure 8. These results infer that some  nonsorbable  compounds
are converted to more sorbable intermediates through ozone transformations.  This possibility

-------
304


is substantiated in the literature, documenting ozone oxidation of aldehydes, ketones,  and
alcohols to carboxylic acids, the product being  more sorbable than the reactants (20, 21).
Based on these results, the possibility of injecting ozone prior to the carbon  columns is
being seriously considered because of the potential advantages of:
     1.  possibly transforming nonsorbable compounds to sorbable compounds, enhancing
        carbon adsorption efficiency;
     2.  reducing the bacterial load  to the carbon columns through particle sterilization; and
     3.  increasing the dissolved oxygen level in the  carbon column influent, reducing  the
        possibility of sulfide production and anaerobic bacterial activity.
There may be problems concerning this approach, such as the technical problems involved
with injecting ozone in the  pressurized  carbon influent line, the high  decomposition rate of
ozone,  and  enhancing corrosion potential in the carbon reactors.  Moreover, disinfection
following carbon adsorption still needs to be resolved.  However, the  advantages presently
appear to outweigh the disadvantages and this approach is being seriously considered as a
process  modification.

     Garland Physical/Chemical System.  A physical/chemical treatment plant is presently
being constructed  at Garland,  Texas.  This 30 MGD  plant is being designed to produce an
effluent having less than 10 mg/l BOD and  TSS.  The raw wastewater quality used for design
is presented in Table 16 (22).   The flow diagram for this  facility  is shown in  Figure 9.   The
carbon basins are common-wall concrete facilities, each  having a surface area of 950 square
feet and a depth of 10 feet.  Nine basins can treat the 30 MGD, while the remaining basin
is off-line for the  backwashing or regeneration.   The wastewater is in  contact with the  carbon
for a minimum of 30 minutes and the upflow rate is  2.5 gpm/ft   at design flow.   A 14.5 foot
multiple-hearth  furnace is designed to regenerate 80,000 pounds of carbon daily.  Based on
the treatability studies, an effluent quality of 15 mg/l COD and 10 mg/l BOD is predicted
(22).

     Rosemount Physical/Chemical System.  The Rosemount, Minnesota physical/chemical
facility has  been in operation since  1974.   The  system receives between 0.3 and 0.6 MGD
and  consists of screening, chemical clarification using either lime or ferric chloride,
prefiltration, upflow activated carbon adsorption, postfiltration, clinoptilolite ammonia
exchangers, and disinfection.  During the first year of operation, the  average BOD removal
was  90 percent,  the average effluent BOD concentration  being 23 mg/l.  More  recently,
however, operating problems have been reduced and  the reported effluent BOD  has  been in
the range of five to 10 mg/l (23).

     Pomona Physical/Chemical Pilot Plant. A  pilot-scale physical/chemical system has
been operating for 27 months in Pomona,  California, under the auspices of the Los Angeles
County  Sanitation District and  the Environmental Protection Agency.  The objectives of
this  study included an evaluation of the long-term effectiveness of granular activated
carbon in the removal of soluble organic matter  from  chemically-clarified municipal waste-
water, controlling hydrogen sulfide generation in the carbon columns,  and determining the
effects of repeated thermal generations  on carbon characteristics and performance.

     A flow  diagram of the pilot system  is shown in  Figure  10, the design data for which are
presented in Table 17 (24).  The primary organic control parameters used in these studies

-------
                                                                                305

were total chemical oxygen demand (TCOD) and the dissolved chemical oxygen demand
(DCOD).  A summary of the overall performance of the physical/chemical system is
presented in Table 18. These quality levels represent average values observed during the
27 months of study.  For comparative purposes,  the effluent quality from an 8.0 MOD
activated sludge system treating the same wastewater is also included.  As noted in Table
18, the major  portion of pollutants, with the exception of DCOD, was removed in the
clarification phase.

    The problems of excessive biological growths in the columns, particularly the
biochemical reduction of su I fates, has persisted in many of the case histories cited.  A
review of this  problem was therefore an  integral part of the Pomona study. As micro-
organisms have the hydrogen acceptor preference  of molecular oxygen, nitrate, sulfate,
and oxidized organics, it  follows that an environment having  the presence of sulfates and
organics and the absence of molecular oxygen and nitrates would favor the anaerobic
sulfate reducers.  The sulfide levels in the effluent approached as high as six mg/l.  This
was a major concern as sulfide production was correlated directly to net head  loss  in the
column (head  loss before the daily backwash minus the head loss after  backwash).  The
fact that the column head loss was caused primarily by abundant biological growths in the
column is underscored by the graphical relationship depicted  in Figure 11 (24).  Several
methods of ameliorating the biological proliferation and sulfide production were under-
taken.  This concluded oxygen addition, intensive air/water  backwash, chlorination, and
sodium nitrate addition to the  carbon column at an average dosage of 5.4 mg/l (as N) was
determined to  be the most effective in terms of retarding sulfide generation in the  carbon
column.

    The spent carbon used in the regeneration studies was backwashed, dewatered to
approximately 50 percent  moisture, conveyed to a six-hearth  furnace, and regenerated at
temperatures ranging from 1,650  F to 1,790 F.  Steam in the amount of 0.6 Ib/lb of
carbon was added to the lower two hearths to enhance the regeneration.  The  regeneration
cycle normally took from 53 to 66 hours to complete.  The effects of regeneration  on the
physical properties of the  carbon are shown in Table  19 (24).   A reduction in the iodine
number (a measure of the extent to which the micropores have been cleared) is noted,
although the molasses number (a measure of the extent of macropore clearing)  is essentially
unchanged. The ash content of the carbon, which measures the amount of calcium and
other  inorganic residues picked up by the carbon during service increased over 60  percent
from the virgin level during the first regeneration.  The ash increase during the second
regeneration was less,  however, with a  decrease during the third regeneration.  The over-
all carbon regeneration loss (bituminous coal) ranged from 2.5 to 6 percent with an average
loss of 4.3  percent over three regenerations.

    Summary.  A summary of the municipal wastewater treatment using the tertiary appli-
cations of activated carbon has been published recently and is shown  in Table 20 (6).  A
similar summary for a straight physical/chemical application of carbon is shown in Table
21 (6) and in Table 22 (25).

EXPERIENCE IN CARBON TREATMENT OF  INDUSTRIAL WASTEWATERS

    Although  ample  carbon treatment data  from full-scale facilities are becoming available

-------
306

 from the municipal sector, much of the performance data in industrial applications evolves
 from pilot-plant studies.  There are several full-scale carbon systems treating industrial
 wastewaters,  however, and this section includes results from pilot studies and selected
 operational systems.  Specifically, results from pilot-plant studies - primarily in the refinery
 and petrochemical sector - will be cited, as well as the full-scale systems at the ARCO
 Refinery near Wilmington, California, the Reichhold Chemicals system in Tuscaloosa,
 Alabama,  and the British Petroleum Refinery in Marcus  Hook, Pennsylvania.

     Pilot  Carbon  Studies.  There have been pilot studies conducted  in the industrial sector
 evaluating activated carbon as both a total (physical/chemical) process and  as an effluent
 polishing (tertiary) unit.  Nine of these studies have been conducted by the  author for
 various petrochemical and petroleum refining facilities  (26).  The efficiency ranged from
 50 to 86 percent COD removal,  as noted in Table 23.   It should be recognized that these
 results were obtained using virgin  carbon with controlled hydraulic and feed rate regimes.
 The attendant problems of full-scale operation and  the effect of using regenerated carbon
 on process efficiency should be considered when translating these results to what might
 occur in an operating carbon treatment system.

     A comprehensive pilot-plant study treating petroleum refinery effluents  by activated
 carbon was conducted recently by the Environmental Protection Agency (27).  Both  API
 separator effluent and biologically-treated effluent were charged to the columns in  order
 to obtain a comparative evaluation.  The results of this study in terms of BOD and COD
 removal are plotted in Figures 12 and 13, respectively.  It is apparent that,  when operated
 in parallel, the biological system  was more effective in removing  BOD and COD -
 particularly the former.   This is consistent with the results observed in pilot studies
 conducted by the  author.  It is also noted that carbon adsorption following biological
 treatment  was particularly effective in reducing both the BOD  and COD to low levels.  The
 residual COD is in the same range as that cited in Table 23 when the carbon application
 mode and  influent COD levels were similar.  The complete  results of this study are shown
 in Table 24.   It is noted that there is no  removal  of cyanides or ammonia although there
 was a reduction of the cited organic constituents, particularly phenols.  There was surpris-
 ingly good removal of chromium, copper, iron, and zinc although the exact mechanisms
 of removal were not determined.   The carbon capacity observed in the columns was  0.31
 Ib TOC removed/lb of carbon, while the isotherm determined capacity was 0.12 Ib/lb.
 This difference was attributed to biological activity observed in the column.  The carbon
 regeneration activity analysis is reported in Table 25,  the change in  iodine and molasses
 numbers showing trends similar to those observed in previous studies.   The investigators
 emphasized that these data were generated using  virgin carbon in  the columns, and  cautioned
 that the presence of iron or aluminum salts present in the effluent could have a deleterious
 effect on the  carbon  through the regeneration  cycle. They  hypothesized  that aluminum
 salts can remain on the surface of the carbon during regeneration, reducing the effective
 surface area of the carbon and reducing its adsorption capacity.  Moreover, iron salts can
 catalyze oxidation reactions of the carbon and the gases during regeneration, thus perma-
 nently damaging the  carbon structure (27).

     Extensive pilot-plant studies have been conducted  recently by Union Carbide evaluating
 activated  carbon  as a tertiary process treating effluent from an activated sludge plant

-------
                                                                                   307

 receiving petrochemical wastewaters (28).  The objectives of this study were both to
 establish a basis of justification for applying activated carbon and to optimize its design.
 Controlled organic and hydraulic application rates were applied to pilot-scale carbon
 columns.  The results of the various throughput rates are listed in Table 26.  The first
 verity which can be deducted from this Table is the fact that,  as the activated sludge
 improves in quality, the efficiency of the adsorber in  terms of COD or soluble organic
 carbon (SOC) improves.  The improved efficiency with decreasing hydraulic throughput
 rates is also apparent.  There was still a 30 to 50 percent  COD residual  in the adsorber
 influent attributable to organic compounds which are neither biologically degradable nor
 sorbable on activated carbon.  There was a high BOD residual at the same throughput
 rate, although this residual was reduced to 35 to 50 percent of the adsorber  influent at the
 uneconomical throughput rate of 0.15 bed volumes per hour.  This BOD residual  includes
 low molecular weight-oxygenated organics such as aldehydes, ketones, alcohols, glycols,
 and other polar organics which are adsorbed  to a very limited extent.  Other observations
 from this study are cited as follows:
     1. breakthrough curves for multi-component wastewaters  tend to be poorly defined
        and sporadic rather than sharp sigmoidal wavefronts;
     2. in multi-bed series adsorbers, the lead bed removes the more readily-adsorbable
        organics and has the highest adsorptive capacity - typical capacities through both
        beds ranged from 0.2 to 0.4 Ib COD/lb of carbon and 0.1 to 0.3 Ib BOD/lb of
        carbon (the observed capacities for the first bed generally were twice  as high as
        those observed in the second); and
     3. the maximum hydraulic application rate  that appeared to be technically justifiable
        was 0.5 bed volumes per hour (through both columns of the series).

     Full-scale Carbon Studies.  There are currently several  full-scale operational  carbon
 treatment systems  in the  United States. Three systems  from which data are available  are
 cited here.  These include the Atlantic Richfield (ARCO)  system in California, the
 Reichhold system in Alabama, and  the British Petroleum (BP) system in Pennsylvania.

     ARCO Carbon Treatment System (Watson Refinery)
     The first full-scale carbon treatment  system in  the petroleum refining industry was
 installed by ARCO at their Watson Refinery near Wilmington, California. This facility
 was necessitated by a resolution from the  Los Angeles  Regional Water Quality Control
 Board which limited the amount of  COD that could be discharged by industry into the
 Dominguez Channel in Los Angeles County.  At the time,  process wastewaters  could  be
 treated in the County's primary treatment unit, but they could not accommodate storm
 runoff  from the Refinery.  This presented a problem as  the  storm water  collection system
and the process sewers were interconnected.   Therefore, during periods of rainfall, the
 combined storm runoff and process flow could not be sent to the County nor could it be
discharged  to  Dominguez  Channel because of its  COD concentration.  A  holding basin
and a carbon treatment plant were therefore deemed to be the most efficacious way of
treating this combined  flow on an intermittent basis (29).
     The treatment system, completed in 1971, is illustrated schematically in Figure 14.
The process wastewater flows through an API  separator and then through pH adjustment  and
chemical flocculation tanks.  Following the addition of both coagulants and polyelectro-
lytes, the flow is routed to one of two circular dissolved air flotation units.  At this

-------
 308

point, the oil and grease (O  & G) levels are normally below 15 to 20 mg/l.  Under dry
weather conditions, the flow goes to the County for further treatment.  During periods of
rainfall, it is diverted to a 50 million gallon holding  basin where it is surged and conveyed
"as required" to the carbon treatment plant.  The plant  consists of 12 adsorber cells, the
flow to which is controlled by a handwheel-operated  slide gate.  Each bed can be back-
washed with  treated water from a backwash sump, the backwash water being returned to the
holding basin.  The carbon handling and regeneration system includes storage hoppers for
spent and regenerated carbon, a pump,  eductor, piping, and controls to convey the  carbon
from any cell to the regeneration furnace and back, and the multiple-hearth furnace with
gas scrubber.  It is gas-fired  and  supplemental steam and air are added on two hearths.  An
afterburner sect-ion is  separately gas-fired to  raise the off-gas temperatures to approximately
1,450 F (necessary to combust the organic vapors in the exit gas).

     The actual design criteria for the system are presented in Table 27.  Based on a
probability distribution of two typical runs presented in  Figure  15, the effluent COD was in
the range of  the predicted level if the influent concentration did not exceed the design
basis.  Chlorination was attempted once the effluent COD exceeded the design level  on
the premise that the halogenated  compounds would be more effectively adsorbed and the
overall performance would  be enhanced.  There was no  noticeable improvement in effluent
COD when this step was concluded, however (29).  Although oil and grease was not cited
as a design parameter, the  observed effluent  concentration and its dependence on the influent
level is shown in Figure 16.  The performance of this  full-scale system in terms of COD
removal approximates the design basis and the observed  reduction of COD and O & G is
consistent with pilot studies using similar wastewaters.  It was noted,  however,  that algal
proliferation  in the holding basin adversely affected the carbon plant performance.  When
excessive algal growths developed, a deterioration of effluent quality and more frequent
backwashing  was characteristic of the system. At one time, the algae became so concen-
trated  that the carbon plant had to be shut down.  Copper sulfate was added to the holding
basin to minimize  algal growth and  ease this  problem.

     The most significant variance of observed performance from the basis of design was the
carbon capacity.  Although a precise determination was not possible, the loading based on
several runs ranged from 0.30 to 0.35 Ib COD removed/lb of carbon, rather than the  1.75
Ib COD/lb carbon prediction cited  in Table 27.

     With the exception of  carbon capacity, the carbon  plant generally performed as expected.
It  is no longer in operation, however, primarily because the treatment requirements imposed
by Los Angeles County have been changed.

     Reichhold Carbon Treatment  System (Tuscaloosa Plant)
     The  Reichhold Chemical  Plant produces sulfuric acid, formaldehyde, pentaerythritol,
sodium sulfate, sodium sulfite,  orthophenylphenol, and a number of synthetic resins.  A
carbon adsorption  system was designed to treat an effluent flow of 500,000 gpd having an
average BOD concentration of 390 mg/l, a COD level of 650 mg/l, and a pH ranging from
5.4 to 12.3 (30).  An agreement  between the Company and the Alabama Water Improvement
Commission established an organic removal objective of 90 percent.

-------
                                                                                 309
    A physical/chemical treatment plant using activated carbon was designed and
constructed and Figure 17 presents a simplified flow diagram.  The process streams are
routed to a 2.5-day retention equalization basin where the water is  pumped to an acid
mixing chamber. Concentrated sulfuric acid is added to maintain a  pH range of 6.5 to
8.5.  A nonionic polymer is added prior to the flocculation chamber.  Following gravity
sedimentation, the  water is pumped to one of two moving-bed adsorbers.  Each adsorber
contains  124,000 Ib of granular activated carbon.  At the design flow rate of 175 gpm/
adsorber,  the  empty bed  contact time is approximately three hours.  Treated  water is
collected in a trough at  the top of each adsorber where it flows  to a final retention tank
and then to the  river.  A conventional regeneration furnace with an afterburner  and wet
scrubber system  is used for reactivation of the carbon.

    In the first  few months of operation, the results have been reported as meeting  the
effluent quality requirements.  The effluent BOD has been in the 35 to 40 mg/l range,
while a 90 percent  removal of COD  infers an effluent at the 65  to 70 mg/l level  (30).

    British Petroleum Carbon Treatment System (Marcus Hook, Pennsylvania)
    The  British  Petroleum (BP)  Refinery is a 105,000 barrel/day Class "B" refinery located
in Marcus Hook, Pennsylvania. In order to comply with discharge standards  prescribed by
the Delaware  River Basin Commission (DRBC), a preliminary engineering program and
treatability study was undertaken by the Company.  Based on these results, the decision to
treat the effluent from the existing API separator to the required  level using a filtration/
carbon adsorption system was made.

    The  treatment system, placed into operation in March 1973, is shown in  Figure 18
(31).   The API separator  effluent flows to an intermediate surge  basin where it is  pumped to
three  downflow  filters, operated in parallel.  The flow rate ranges from seven to  12 gpm/
ft , depending on the operation, and the media consists of 2.5 ft of anthracite and 4.5 ft
of sand.  The  fJlters are  backwashed using filtered water, with provisions for  air scouring
at 7.1 scfm/ft  . The design filter pressure is 47.5 psi  and the maximum  allowable pressure
drop through the media is 6.5 psi.  The backwash interval is 12  hours.

    Three carbon adsorbers (10 ft diameter and 65 ft high) are operated in parallel.  Each
adsorber  contains 92,000 pounds of granular  activated carbon  in  a bed depth  of 45 feet.
An additional  8,000 pounds occupy the lower and upper cone  areas.  Flow to the three
adsorbers is controlled by the level in the filtered water holding  tank.  The upward flow
rate is 8.5 gpm/ft  , which gives an  empty bed contact time of 40 minutes.  Spent carbon
at the  bottom  of the adsorber is pulsed out at the rate of approximately 1,000 Ib/day.
Fresh  carbon is then added to the top of the column from a feed  hopper.  The design
criteria and activated carbon properties for the system are shown in Table 28  (32).

    The regeneration facility is a five-foot diameter multiple-hearth furnace.  The design
feed rate to the six-hearth furnace is 125 Ib/hr.  In an atmosphere controlled by the addition
of steam,  the adsorbed organics are volatized and oxidized.   In order to insure complete
oxidation, all flue  gases pass through an afterburner fired by refinery fuel gas and maintained
at a temperature of 1,350°F.   A wet scrubber is  included for gas cooling and the removal
of particulate  matter.  The design data for the thermal regeneration  system is shown in

-------
310

Table 29.

     As the full-scale system has been in operation for over two years,  performance data are
available (31,  32).  During the first months of operation while the adsorber still contained
virgin carbon,  the COD removals  were determined to be independent of influent concentration
but dependent on contact time. This relationship is shown in Figure  19.  The oil and grease
removal during this period,  shown in  Figure 20, followed two different regimes. Good removal
was observed initially prior to  the start of carbon pulsing.   However, once the pulsing began,
taking the pulsed bed out of operation, the adverse effect on the remaining two columns is
reflected  in terms of deteriorating effluent quality.  The reduction of oil  removal with increasing
increasing influent  concentrations is attributed to both the oil removal  mechanism and pulse
bed "off-line"  mode of operation.

     There have been four distinct phases of operation since the carbon system came on line.
The first period of data reported herein occurred when virgin carbon was  in the  adsorbers and
the foul water  condensate stream from the Fluidized Catalytic Cracking Unit (FCCU) was not
yet included in the raw wastewater stream.  The second period included the  FCCU stream,
but virgin carbon was still present in  the adsorbers as the wave front had  not yet reached the
top of the carbon bed.  The third period encompassed a time following  a  complete  turnover of
the carbon bed and still  included the FCCU stream. The fourth period  excluded this stream
and incorporated a  modification to the column septum design, still following a complete carbon
turnover. The  average and maximum adsorber effluent concentrations for several parameters
observed during each period is tabulated in Table 30.   A significant deterioration  in quality
following the inclusion of the FCCU foul water condensate and complete  carbon turnover in
the reactor is noted in Figure  21 .  This has been attributed to inadequate pretreatment of the
API separator effluent in terms of O &  G and  soluble organic removal as  well as significant
buildups of anaerobic biological growths and oily materials in the carbon media.  A 40
percent decrease in adsorptive capacity of the regenerated carbon also has been observed.
The iodine number of the regenerated carbon is 560 to 680, while the virgin carbon was in
the range of 950 to 1,000 and  the molasses number of the regenerated carbon was 280,  or
approximately 50 higher than  the virgin carbon. This indicates u decrease in micropores and
an increase  in macropores during the  regeneration cycle.

     The Company is currently  reviewing methods of improving process performance over that
observed in the  last three periods of operation, or possibly changing the system  concept
altogether.  Particular emphasis is being placed on more adequate pretreatment, including
biological oxidation, as steps necessary to effectively use the existing  carbon system.

SUMMARY

    Activated  carbon treatment of industrial wastes, while promising,  must be carefully
evaluated before process decisions are made and capital funds are committed.  As  noted in
the pilot- and full-scale case histories  presented herein, breakthrough  geometry and adsorption
kinetics of multi-component wastewaters are difficult to define,  many organic compounds
are not amenable to carbon adsorption, and the effects of  regeneration on carbon capacities
are variable and unpredictable.  For  these and other reasons, comprehensive testing and
technical  reviews are a necessary  prerequisite  to process commitment.

-------
                                                                                  311
     Unfortunately, most of the literature regarding carbon treatment of industrial waste-
waters centers around pilot-plant results.  There are,  however, sufficient data from pilot
studies and the two full-scale systems in the petroleum refinery industry to draw general
conclusions,  at least within this industrial category.  For example, a distribution of long-
term average COD concentrations in effluents from carbon adsorbers treating refinery
wastewaters is presented in  Figure 22 (33).  These  levels are relatively consistent with
those residuals reported in the  petrochemical  industry (28).  A similar presentation of oil
and grease effluent levels from a full-scale refinery carbon system is shown in Figure 23.
Carbon adsorption capacities in terms of Ibs of COD removed/lb of carbon exhausted have
ranges from 0.2 to 0.4  in the petrochemical industry and from less than 0.1  to 0.55 in the
petroleum refining industry. These are lower than  reported carbon capacities for municipal
wastewaters, as shown in Figure 24, emphasizing the inaccuracies which can occur by
extrapolating results from the treatment of one wastewater and using them as the basis for
predicting another.

SYNOPSIS

     A review of the current state of the art of activated carbon treatment has been pre-
sented.  Basic concepts of activated carbon treatment have been included, as well as
pertinent municipal and industrial case histories with which the author  is familiar.  It is
recognized that new truths pertaining to this subject become known on  a  continuing basis.
However, in evaluating process concepts, developing design  bases, predicting  effluent
quality,  and finalizing management decisions in terms of constructing control systems with
attendant capital  commitments, one must base these judgments on the current state of the
art.   It is toward the objective of defining  the art  of activated  carbon treatment that this
information is presented.

     In the pursuit of this definition, certain apparent verities emerge.  Some of the more
significant include the  following:
     1 .   Adsorption theory  is rigorous for single solutes, but becomes less definitive when
         applied to wastewaters containing multiple components with varying molecular
         weights and chemical characteristics.  A  good example is the poorly-defined
         and erratic breakthrough geometry observed in the carbon treatment of complex
         industrial wastewaters.
     2.   Many classes of organic compounds are not amenable to carbon  adsorption -
         particularly oxygenated organics - and show up as residual BOD, COD, or TOC
         in carbon column effluents.  This  limits the overall process efficiency of pure
         physical/chemical treatment systems.  As many of these residual compounds are
         biodegradable, activated carbon as a  polishing process is generally capable of
         producing a  better quality of effluent than is the strict physical/chemical appli-
         cation.
     3.   Anaerobic conditions which prevail  in carbon reactors have caused difficulty
         through the proliferation of biological growths on the  carbon media.  Sulfide
         generation has been particularly troublesome.  Although biological activity in
         the column has the potential  of increasing carbon capacities and is hypothesized
         to offer regeneration  possibilities, full-scale experience has indicated that
         uncontrolled biological growth in  carbon  reactors has more negative features than

-------
312

         positive.
     4.  Ozonation of wastewaters prior to carbon adsorption has been demonstrated on a
         pilot scale to enhance adsorption in the columns. This is attributed to the ozone
         products being more sorbable than the reactants.  Other possible attractions of
         this application are controlling biological densities on the carbon media through
         partial disinfection and maintaining an aerobic environment in the reactor column.
     5.  Carbon capacities used as design criteria generally have been overstated and have
         not been realized in full-scale experience.  Breakthrough concentrations which
         force premature regeneration cycles in single-operated columns, and desorption
         phenomena which cause low capacities in series-operated polishing columns, are
         partially responsible for this.
     6.  Design criteria for carbon adsorption systems should be sensitive to:
         (a)   wastewater constituents and their classification  in terms of adsorbability,  and
               the effluent residual potential; and
         (b)   effect  of the selected carbon media with respect to capacity, resistance to
               abrasion, regeneration impacts, and hydraulic characteristics; and
         (c)   the biological growth potential  with the associated effects on carbon capacity,
               backwash requirements,  and hydraulic characteristics of the flow through
               the column; and
         (d)   the necessary pretreatment requirements for control of suspended solids
               (organic and inorganic), oil and grease, dissolved oxygen,  biological
               population, and other constituents  which affect carbon adsorber performance.

 REFERENCES

  (1)  Weber,  Walter J.,  Physicochemical  Processes for Water Quality Control, John Wiley
      and Sons, Inc.,  New York, (1972).
  (2)  Bell,  Bruce A. and  Molof, Alan H., "A New Model of Granular Activated Carbon
      Adsorption Kinetics," Water Research,  Vol.  9,  Pergamon  Press, London, (1975).
  (3)  Ford,  D. L., "The Applicability of Carbon Adsorption in  the Treatment of Petro-
      chemical Wastewaters," Proceedings, The Application of New Concepts of Physical-
      Chemical  Wastewater Treatment, Sponsored by the International Association of Water
      Pollution Research and the American Institute of Chemical Engineers,  Vanderbilt
      University,  Nashville, (September,  1972).
  (4)  Giusti,  D. M.,  Conway, R. A., and Lawson, C. T., "Activated Carbon Adsorption
      of Petrochemicals," Journal, Water Pollution Control Federation, (May, 1974).
  (5)  Hassler, John W., Purification With Activated Carbon, Chemical  Publishing Co.,
      Inc.,  New York, (1974).
  (6)  Environmental  Protection Agency,  Process Design Manual  for Carbon  Adsorption,
      Technology Transfer Manual, (October, 1971).
  (7)  American Water  Works Association, "AWWA Standard for Granular Activated Carbon,"
      Journal, AWWA, (November,  1974).
  (8)  Smisek, M., and Cerny, S., Active Carbon, Elsevier Publishing Co.,  New York,  (1970).
  (9)  De John, P. B.,  "Carbon from  Lignite or Coal,  Which is  Better?", Chemical Engineering,
      (April,  1975).
 (10)  Loven, A. W.,  "Perspectives on Carbon Regeneration,"  Chemical Engineering Progress,
      (November, 1973).

-------
                                                                               313


(11)  Federal Water Quality Administration,  "Development of a Fluidized Bed Technique
     of the Regeneration of Powdered Activated Carbon," Water Pollution Control Res.
     Series ORD-17020 FBD, (March, 1970).
(12)  Environmental Protection Agency, "Powdered Activated  Carbon Treatment of
     Combined and Municipal Sewage," Water Pollution  Control Res. Series 11020 DSO,
     (November, 1972).
(13)  Berg, E.  L.,  etal, Chemical Engineering Proceedings 67, No. 107, 154, (1970).
(14)  Fuchs, JohnL., Private Communication to Engineering-Science, Inc., (June, 1975).
(15)  Culp, R. L., and Roderick,  R.  E., "The Lake Tahoe Water Reclamation Plant,"
     Journal,  Water Pollution Control Federation, (February, 1966).
(16)  Slechta,  A. F., and Culp, G.  L., "Water Reclamation  Studies at the South Tahoe
     Public Utility District," Journal, Water Pollution  Control Federation, (May, 1967).
(17)  Ford, D. L., personal communication with the City  of Colorado Springs, Colorado,
     (July, 1975).                                                         --
(18)  Batelle - Northwest and Zurn Environmental  Engineers,  "Westerly Advanced Waste-
     water Treatment Facility,  Process Design Report and Appendices," Cleveland, Ohio,
     (1971).
(19)  Guirguis, W.  A., et  al, "Ozonation Studies at the  Westerly Wastewater Treatment
     Center," Second  International Ozone Symposium, Montreal (May,  1975).
(20)  Evans, F. L., Ozone in Water and Wastewater Treatment, Ann Arbor Science, Inc.,
     (1972).
(21)  Snoeyink, V.  L.,  Weber, W. J. and Mark,  H. B.,  "Sorption ofPhenol and Nitro-
     phenol by Active Carbon," Environmental Science and Technology,  (October, 1969).
(22)  McDuff,  D. P., and Chiang, W. J., "Physical Chemical Design for Garland, Texas,"
     The Applications of New Concepts of Physical - Chemical Wastewater Treatment,
     Sponsored by the  International Association of Water  Pollution Research and the
     American Institute of  Chemical  Engineers, Vanderbilt University, Nashville,
     (September, 1972).
(23)  Engineering-Science, Inc.,  verbal communication with the Metropolitan Sewer
     Board, Rosemount, Minnesota, (1975).
(24)  Directo,  L.  S., and Chen, C. L.,  "Pilot Plant Study of Physical Chemical Treat-
     ment," 47th Annual Water Pollution Control  Federation Conference,  Denver,
     Colorado, (October,  1974).
(25)  Engineering-Science, Inc. and  Cleveland Regional Sewer District, Report on
     Evaluation of Continuing Westerly Pilot-Plant Studies, prepared for CRSD,(June,
     1975).
(26)  Ford, D. L., and Buercklin, M. A., "The Interrelationship of  Biological-Carbon
     Adsorption Systems for the Treatment of Refinery and Petrochemical Wastewaters,"
     6th  International Association of Water Pollution Research Conference, Jerusalem,
     (June, 1972).
(27)  Short, T.  E.,  and  Myers,  L. A., "Pilot Plant Activated Carbon Treatment of
     Petroleum Refinery Wastewaters," Robert S.  Kerr Environmental Research Laboratory,
     Ada,  Oklahoma,  (1975).
(28)  Lawson, C.  T., "Activated Carbon Adsorption for Tertiary Treatment of Activated
     Sludge Effluents from  Organic Chemicals and Plastics Manufacturing Plants - Appli-
     cation Studies and Concepts," Research  and  Development Department, Union Carbide
     Corporation,  South Charleston, West Virginia, (August, 1975).

-------
314

 (29)  Environmental  Protection Agency,  Refinery Effluent Water Treatment Plant Using
      Activated Carbon, Environmental Protection Technology Series, EPA 660/2-75-020,
      (June, 1975).
 (30)  Shumaker, T.  P., "Granular Carbon Process Removes 99 - 99.2% Phenols," Chemical
      Processing, (May, 1973).
 (31)  McCrodden, B. A.,  "Treatment of Refinery Wastewater Using Filtration and Carbon
      Adsorption," Advanced  Petroleum Refinery Short Course, Principles and Practice in
      Refinery Wastewater  Treatment, University of Tulsa,  (June,  1973).
 (32)  McCrodden, B. A.,  "Treatment of Refinery Wastewater Using Filtration and Carbon
      Adsorption," presented at Technology Transfer Conference, Activated Carbon in Water
      Pollution Control, Sponsored by the Pollution Control Association of Ontario, and  the
      Canadian Society for Chemical Engineering, (October,  1974).
 (33)  Engineering-Science, Inc., Report to the National Commission on Water Quality,
      Petroleum Refinery Industry - Technology and Cost of Wastewater Control,  (June, 1975).

 DISCUSSION

 Ed Sebesta:  Were these pilot plant data taken with regenerated carbon and equilibrated
 carbon?

 Davis Ford:  No. It  was virgin carbon.  That's a good point.  Many of the pilot plants
 studies that we have run have been with virgin carbon.  It is not that we don't recognize
 that it would be more applicable with  regenerated carbon, it is just sometimes that it is
 hard  to get that much regenerated carbon from the vendor and that has been our problem.
 One  would think it  would  be  more practical to run all these pilot studies with regenerated
 carbon.

 Bob Huddleston:  Do you have any data that indicates whether or not the effects  of ozone
 were anything  other than simply killing the microbes?

 Davis Ford:  Yes.  They ran some  GC work to identify the  nature of the compounds after
 organization and I don't have the results in this paper.  Just in talking with these people,
 it's what you really  have done is gone from the ketones, aldehydes to carboxylic acids,
 there  has been a shift and  the carboxylic acids being more ameniable to carbon absorption,
 so its a combination of  the biological growth but also the transformation of these oxygenated
 organics to carboxylic acids.

 M. J<_. Mutton:  Discuss if  you will the cost of ozone. Do  you have any figures on this?
 I expect it is quite expensive.

 Davis Ford:  I left cost  completely out of the paper because that  is almost another paper
 itself; but considering the  concentration of ozone  that we think we have to have, it is
 probably about 4 or  5 milligrams  per liter is within the cost effective parameters that we
 have  set to shis point.   So  we think  it  is cost effective.

 Anonymous:  Do you have  some comments on phenol  removal ?

-------
                                                                                 315


Davis Ford:  No.  Phenol  is highly sorbable, most of the data we have seen on phenol
removal has been excellent and the only exceptions to this was when we just had so many
physical problems, the biological growths and oil and grease, that the phenol removal
has gone down.  I think that is shown in one table.  If you don't have this constraint,
phenol removal should be  good,  carbon is a good phenol remover.

Morris Wiley:  The little village where  I live has a new sewer system and biological
treatment plant.  It is an extended aeration without any primary treatment, then a second
stage nitrification and finally a sand filter. They are able to do less than one part per
million BOD and less than  one part per  million ammonia in the effluent. Now I should
caution that it  is only operating at one  quarter of rated  capacity.  Thus, it appears that
if one  has to go to these very low organic and nitrogen contents in the effluent, that one
can do it by a conventional biological system  simply by spending a  lot of money.  I
estimate that this is about $1,000 per family per year, total annualized cost for this
system including the sewer connections  which  were put in.  Have you done any economic
studies to try to figure out  which would really be preferable, an advanced  biological
reactor system versus an activated  carbon where you  need a very low concentration of
pollutants in the effluent?

Davis  Ford: Let me speak in general terms here.  First, you  have a good analytical
chemist at that plant.

Morris Wiley:  Well,  I have some reservations about  the precision of the tests.  The plant
operator is a Texaco employee running the plant in his spare  time with his wife and his
son, but it is a good quality effluent and the county  is accepting their test reports.

Davis Ford:  Let me say, from the cost studies  that we have run, we have seen biological
treatment is more cost effective than carbon, I don't think there is any question about
that. When you get down  to the fact that a biological system can't make a permit and
you have to go to carbon,  you know some people are not sympathetic with  cost effectives,
but if you can get to those levels biologically, it just intuitively has to be more cost
effective than carbon.  Carbon has gone up, it is 50-60<:  now for bituminous.  When we
started running these studies three  or four years ago, it was 32$ or 30<: and just because
of these lower carbon capacities I  have stated, biological treatment has to be more cost
effective.  If we can make it with biological treatment,  that is certainly the theology that
I  would like to see.

Morris Wiley:   Then  it looks as though our research should be concentrated on enhancing
the effectiveness this biological treatment as Jim Grutsch pointed out earlier?

Davis Ford:  I do not disagree with that.  I also think we have got to be realistic, there
are certain limitations to biological treatment.  When you look at BAT, if  and when it
comes, it is going to take some real engineering to make those numbers biologically.
Neale Fugate:  Do you have any data on the possible adsorption of heavy metals,
conversely desorption of  heavy metals on carbon?
or

-------
316


Davis Ford: Yes, sir.  I  have it, but it is not presented in the paper.  I have just seen some
data on that.  There is slight reductions of some of the metallic ions  probably as metallic
hydroxides.  I think in some literature absorption of some metallic ions has been reported.
It has not been significant, something like 10-15% reduction of zinc, mercury,  cadmium,
some of these substances  but percent removal is quite low, but some removal has been noted.
I  am not sure of the mechanism as a physical removal of zinc or sorption of zinc  iron, but
one of those two mechanisms.

Robert L. Wortman: It is my understanding that ozone is a very effective phenol remover,
it removes the phenol quite effectively,  it is probably even better than activated carbon
(through oxidation).  It would seem to me that may be secondary conventional treatment
followed by ozonation may be as effective as secondary treatment followed by carbonation.
My question is,  have you had  any experience on  this in combination  with  carbon filtration
and I  am interested in treating each one individually, that is considering ozonization as a
treatment method?

Davis Ford: Let me answer that two ways.  First of all, I think I mentioned in the talk
that we tried preozonation after carbon with bad  results. However,  there is a treatment
plant, financed in  part by EPA which goes on line in about two months and that  is at Estes
Park, Colorado, which is exactly the system you  described.  Other than that plant,  I am
not personally familiar with a real good biological  treatment system including nitrification
followed by ozone  but that system is designed that way. That  is the  upper sanitation district
in Estes Park, Colorado, the plant is completed and is probably going through shake down
right  now and we will get some good full scale information from that  facility in the next
few months.

Anonymous:  I think there is also some work being done in  Pomona, I don't know exactly
what  it was, but on a different combination of ozonation and various schemes; but to further
explore the Estes Park, if you say you  had a  multimedia filter and then go in ozonation, this
may be quite effective to work with.

Fred Gowdy:  Davis, I wonder if you would give us some of the reasons you believe why the
carbon  capacities that we have realized in some of our full scale  plants,  haven't been any-
where close to those predicted by various pilot studies?

Davis Ford: I am not really sure  that I know the answer to that, I can just hypothesize.
First of all, I think Union Carbide has presented carbon capacities in lead columns as far
as polishing columns were concerned,  so that would indicate the operation has a lot  to do
with carbon capacity, in other words, how well you can operate the  carbon utilization in
a series mode of operation to get the operable  capacity of the  overall system.  You can
pretty well  do that  in a pilot scale. I am not sure you can do  it in a full scale quite so
easily.   It is easier to operate  that pilot scale, I  assure you, than a full-scale plant. So
I  think part of the answer lies in  the operation. Secondly,  it is possible that you had more
physical problems in terms of biological growth, influent constituents in the full  scale that
you don't perceive  in a pilot scale. You might control the pilot scale a little better and
for this reason are able to realize a  better carbon capacity.  Those are the two main reasons
I  think,  I just can't prove it.

-------
                                                                                 317

Ed Soger:  I  wondered if you would like to elaborate on the back washing features of these
columns.  Why is that required if you have adequate pretreatment, are you removing
suspended solids build-up?

Davis Ford:  I  think the best case history to point to there is the BP, they have got some
data on their filters, good filters which work quite well; but the effluent suspended solids
are probably in the 10-15-20 milligram per liter range and these influent suspended  solids
to the carbon  column are going to accumulate after a certain period of time, combined
with the biological proliferation  in the columns.   I think you always have to have the  flexi-
bility operational  flexibility of back washing columns as well as the filters.  I  think that
has been the experience to date.  You may not have to do it, but it is an operational
flexibility that you should have.   It seems like even if you go in with zero suspended
solids, some point in time you are going to have to back wash the filters or the carbon
columns.

Pat DeJohn:  One of the things that one of the gentlemen asked about the loadings,  most
of the loadings that are reported  are reported with virgin carbon and the properties of
activated  carbon change quite substantially when  you regenerate carbon and I  think that is
part of the BP problem, they are  not able to maintain their phenol standards because of
the loss in capacity, adsorptive capacity and consequently, their system has been sub-
stantially under design.  The other thing that you  mentioned about why you should be  back
washing,  that BP plant is a  pulse bed unit and there are fines that are created due to the
regeneration of the carbon and they have no way to remove those fines; consequently they
recycle them back into the  absorber and they pulse that unit a lot of times because of
pressure drop problems and that is another thing about that particular plant.

Davis Ford:  I  agree with that.  In the past. Pat, when we have had to use virgin carbon,
we have just applied a factor on  the capacity, reduced capacity, just an empirical factor
and that is a less favorable  way of doing it than using regenerated carbon; but  we just
haven't had that luxury a lot of times.

J.  J. Chavez:  I was just wondering if you could elaborate on the regeneration phase
from the economic side by which a refinery could  put in one regulation facility and  the
options  open to him if he can't afford one?

Davis Ford: There again I didn't really get into cost.   I believe the reported  regeneration
cost now is something like  15-20<: per pound.   Some manufacturers, I believe,  are offering
a regeneration service.  If you have a small system, it doesn't make sense because of
economy in  scale to put in your own regeneration  facility, to have a contractor regenerate
it, if that's  possible, but I think  if you go into regeneration which in most areas you would
have to for most size of plants, you  have to count on about a 15-20$ per pound regeneration
cost.

-------
318
 BIOGRAPHY

      Davis L. Ford holds a B.S. in Civil Engineering
 from Texas A & M University and M.S. and Ph.D.
 degrees in Environmental Health Engineering from the
 University of Texas at Austin.  Dr. Ford is currently
 Senior Vice  President and Member of the Board of
 Directors of  Engineering Science,  Inc., in Austin,
 Texas.  Dr.  Ford has written 4 books, 20 reports, 60
 publications in the field of environmental  engineering
 and has consulted for over 50 industries, the United
 Nations (WHO and PAHO), the EPA and various state
 and municipal agencies.

         TABLE 1 "COMPARATIVE ANALYSIS OF BATCH ISOTHERM DATA
                   REFINERY & PETROCHEMICAL WASTEWATERS"
                                                          K
      OIL SEPARATOR (PRIMARY) EFFLUENT
          Refinery - Petrochemical Complex No. 1
          Refinery - Petrochemical Complex No. 2
          Refinery No. 3

      SECONDARY (ACTIVATED SLUDGE) EFFLUENT
          Refinery - Petrochemical Complex No. 1
          Refinery No. 3
          Refinery Secondary Effluent
          Refinery Secondary Effluent
          Refinery Secondary Effluent

      SINGLE ADSORBATE
          Phenol
          Dichlorethane (pH 4)
                       (pH6)
                      (pH 10)
0.0290
0.0036
0.0140
0.0062
0.0043
0.0051
0.0038
0.0020
0.1110
0.0045
0.0038
0.0041
0.77
0.80
0.36
0.60
1.00
0.96
1.08
0.69
5.80
1.82
1.67
1.49

-------
                                                                                                           TABLE  2

                                                                       AMENABILITY OF  TYPICAL ORGANIC COMPOUNDS
                                                                                  TO  ACTIVATED  CARBON ADSORPTION  [4]
               Compound

Alcohols
  Methanol
  Ethanol
  Propanol
  Butanol
  n-Amyl alcohol
  n Hexanol
  Isopropanol
  Ally! alcohol
  Isobutanol
  t Butanol
  2 Ethyl butanol
  2Ethylhexanol
Aldehydes
   Formaldehyde
  Acetaldehyde
   Propionaldehyde
   Butyraldehyde
   Acrolcm
   Crolonaldehyde
   Benzaldehyde
   Paralo>riyde
Amines
   Di N Propylamme
   Butylamtne
   Di N Butylamme
   Allylamme
   Elhylenediamme
   Dielhylenelriamme
   Mcnelhanolamine
   Dicthanotamme
   Tntihanclamme
   Monoisopropanolamme
   Dusopiupanolamme
 Pyridines & Morpholines
   Pyndine
   2 Methyl 5-Ethyl pyridine
   N Methyl morpholme
   N Ethyl morpholme

 Aromatics
   Benzene
   Toluene
   Elhyl benzene
   Phenol
   Hydfoqumone
   Aniline
   Styrene
   Nitrobenzene
 Esters
   Methyl acetate
   Ethyl acetate
   Piopyl  acetate
   Butyl acetate
   Primary a my I  acetate
Concentration (mg/l)

Molecular
Weight
320
46.1
60.1
741
88.2
1022
60 1
58.1
74 1
74.1
1022
1302
30.0
44 1
581
72 1
56.1
701
1061
1322
101 2
731
1293
571
601
1032
61 1
1051
1491
751
1332
791
1212
1012
1152
78.1
92 1
1062
94
1101
931
1042
123 1
74 1
88 1
1021
1162
1302
Aqueous
Solubility
(*)
00
00
oo
7.7
17
0.58
OO
oo
85
00
0.43
007
oo
00
22
71
206
155
0.33
10.5
oo
oo
oo
oo
oo
oo
oo
954
OO
oo
87
OO
si. sol
OO
oo
007
0047
002
67
6.0
34
003
019
319
87
2
068
02


Initial (Co)
1,000
1.000
1,000
1.000
1,000
1,000
1.000
1,010
1,000
1.000
1.000
700
1,000
1,000
1,000
1.000
1.000
1,000
1.000
1,000
1.000
1.000
1,000
1.000
1,000
1,000
1,012
996
1.000
1,000
1.000
1,000
1.000
1,000
1,000
416
317
115
1,000
1,000
1.000
180
1,023
1.030
1,000
1.000
1.000
985


Final (C|)
964
901
811
466
282
45
874
789
581
705
145
10
908
881
723
472
694
544
60
261
198
480
130
686
893
706
939
722
670
800
543
527
107
575
467
21
66
18
194
167
251
18
44
760
495
248
151
119
Adsorbability'

g compound/
g carbon
0007
0020
0038
0107
0155
0191
0025
0.024
0084
0.059
0170
0138
0.018
0022
0057
0106
0061
0092
0.188
0148
0,174
0103
0174
0063
0021
0062
0015
0057
0067
0040
0091
0095
0179
0085
0 107
0080
0050
0019
0161
0167
0150
0028
0196
0054
0100
0149
0169
0175

Percent
Reduction
36
100
189
534
71.8
955
12.6
219
419
295
855
985
9.2
11.9
277
52.8
306
456
940
739
802
520
870
31 4
107
294
72
275
33.0
200
457
473
893
425
53.3
950
792
843
806
833
749
88.8
95.6
262
505
752
846
880
               Compound

Esters
  Isopropyl acetate
  Isobutyl acetate
  Vinyl acetate
  Ethylene glycol monoethyl ether acetate
  Ethyl acrylate
  Butyl acrylate
Ethers
  Isopropyl ether
  Butyl ether
  Dichloroisopropyl ether
Glycols& Glycol Ethers
  Ethylene glycol
  Diethylene glycol
  Triethylene glycol
  Tetraethylene glycol
  Propylene glycol
  Dipropylene glycol
  Hexylene glycol
  Ethylene glycol mono methyl ether
  Ethylene glycol monoethyl ether
  Ethylene glycol monobutyl ether
  Ethylene glycol monohexyl ether
  Diethylene glycol monoethyl ether
  Diethylene glycol monobutyl ether
  Ethoxytriglycol
Halogenated
  Ethylene dichlonde
  Prupylene dichlonde
Ketones
  Acetone
  Methylethyl kelone
  Methyl propyl Ketone
  Methyl butyl ketone
  Methyl isobulyl ketone
  Methyl isoamyl ketone
  Diisobutyl ketone
  Cydohexanone
  Acelophenone
  Isophorone
Organic Acids
  Formic acid
  Acetic acid
  Propiomc acid
  Butyric acid
  Valeric acid
  Caproic acid
  Acrylic acid
  Benzoic acid
Oxides
  Ptopylene oxide
  Styrene oxide
  Dosage 5 g Carbon C/l solution
Concentration (mg/l)

Molecular
Weight
102.1
116.2
86.1
132.2
100.1
128.2
1022
130.2
171.1
62.1
1061
150.2
194.2
761
134.2
1182
76.1
901
1182
1462
1342
1622
1782
990
113.0
581
72 1
861
1002
100.2
1142
1422
982
1201
1382
460
601
741
881
1021
1162
721
1221
581
1202
Aqueous
Solubility
(*)
2.9
0.63
2.8
22.9
20
02
1.2
003
0.17
OO
oo
oo
oo
00
00
oo
oo
oo
00
099
oo
00
oo
0.81
0.30
00
268
43
v si sol.
19
0.54
0.05
2.5
0.55
1 2
OO
OO
oo
00
24
1.1
oo
029
405
0.3


Initial (C,)
1.000
1.000
1.000
1.000
1.015
1,000
1,023
197
1.008
1.000
1,000
1.000
1.000
1.000
1.000
1.000
1,024
1.022
1.000
975
1.010
1,000
1,000
1,000
1.000
1.000
1.000
1.000
988
1.000
986
300
1,000
1,000
1,000
1.000
1.000
1.000
,000
,000
,000
,000
,000
1,000
1,000


Finil (C|)
319
180
357
342
226
43
203
nil
nil
932
738
477
419
884
835
386
886
705
441
126
570
173
303
189
71
782
532
305
191
152
146
nil
332
28
34
765
760
674
405
203
30
355
89
739
47
Adsorbability

I compound/
t carbon
0137
0164
0.129
0.132
0 157
0193
0162
0039
0200
00136
0053
0.105
0116
0024
0033
0122
0028
0063
0112
0.170
0087
0166
0 139
0.163
0.183
0043
0.094
0139
0.159
0169
0.169
0.060
0.134
0194
0.193
0047
0.048
0065
0119
0.159
0.194
0129
0183
0.052
0.190

Percent
Reduction
68.1
82.0
643
65.8
77.7
95.9
800
100.0
100.0
68
262
523
581
116
165
614
135
310
559
87 1
436
827
697
81 1
929
218
468
69.5
807
848
85.2
100.0
66.6
972
966
235
240
326
595
797
970
645
91 1
261
953
                                                                                                                                                                                                                                        CO

-------
  320
                   TABLE 3 "RELATIVE AMENABILITY TO CARBON ADSORPTION OF TYPICAL
                               PETROCHEMICAL WASTEWATER CONSTITUENTS"

Compound                            % Adsorption       Compound                                 % Adsorption

Ethanol                                   10             Vinyl acetate                                  64
Isopropanol                                13             Ethyl  acrylate                                  78
Acetaldehyde                              12             Ethylene glycol                                 7
Butyraldehyde                             53             Propylene glycol                               12
Di-N-propylamine                         80             Propylene oxide                                26
Monoethanolamine                          7             Acetone                                       22
Pyridine                                   47             Methyl ethyl ketone                            47
2-Methyl 5-ethyl pyridine                  89             Methyl isobutyl ketone                         85
Benzene                                   95             Acetic acid                                    24
Phenol                                    81             Proprionic acid                                33
Nitrobenzene                              96             Benzoic acid                                   91
Ethyl  acetate                              50

Initial Compound Concentration = 1,000 mg/l       Powdered Carbon Dosage =5,000 mg/l
         TABLE 4 "INFLUENCE OF MOLECULAR STRUCTURE AND OTHER FACTORS OF ADSORBABILITY"

 1 .   An increasing solubility of the solute in the liquid carrier decreases its adsorbability.

 2.   Branched chains are usually more adsorbable than straight chains.  An increasing length of the chain decreases solubility.

 3.   Substituent groups affect adsorbability:

     Substituent Group                                      Nature of Influence

     Hydroxyl                             Generally reduces adsorbability; extent of decrease depends on structure of
                                          host molecule.
     Amino                                Effect similar to that of hydroxyl but somewhat greater.   Many amino acids
                                          are not adsorbed to any appreciable extent.
     Carbonyl                             Effect varies according to host molecule; glyoxylic are more adsorbable
                                          than acetic but similar increase  does not occur when introduced into
                                          higher fatty acids.
     Double Bonds                          Variable  effect as with carbonyl.
     Halogens                             Variable  effect.
     Sulfonic                              Usually decreases adsorbability.
     Nitro                                 Often increases adsorbability.
     Aromatic Rings                        Greatly increases adsorbability.

4.   Generally, strong ionized solutions are  not as  adsorbable as weakly ionized ones; i.e., undissociated molecules
     are in general preferentially adsorbed.

5.   The amount of hydrolytic adsorption depends on the ability of the hydrolysis to form an adsorbable acid or base.

6.   Unless the screening action  of the carbon pores intervene,  large  molecules are more sorbable than small molecules
     of similar chemical nature.  This  is attributed to more solute carbon chemical  bonds being formed, making
     desorption more difficult.

7.   Molecules with low polarity are more sorbable than highly polar  ones.

-------
                                                                                                            321
                                                         TABLE 5

                                PROPERTIES OF SEVERAL COMMERCIALLY AVAILABLE CARBONS*(6]
PHYSICAL PROPERTIES
Surface area,  nWgm (BET)
Apparent density, gm/cc
Density, backwashed and drained, Ib/cu. ft.
Real density,  gm/cc
Particle density, gm/cc
Effective size, mm
Uniformity coefficient
Pore volume,  cc/gm
Mean particle diameter, mm
SPECIFICATIONS
Sieve size (U.S. std. series)
  Larger than  No. 8   (max.
  Larger than  No. 12  (max.
  Smaller than No. 30 (max.
  Smaller than No. 40 (max.
Iodine No.
Abrasion No., minimum
Ash (%)
Moisture as packed (max. %)
                          %)
                          %)
                          %)
                          %)
                                                ICI
                                             AMERICA
                                           HYDRODARCO
                                               3000
                                             (LIGNITE)

                                            600-650
                                            0.43
                                            22
                                            2.0
                                            1.4
                                            0.8
                                            1.7
                                            0.95
                                            1.6
8

5

650
     1.5
     0.9
   CALGON
 FILTRASORB
     300
    (8x30)
(BITUMINOUS)


 950-1050
 0.48
 26
 2.1
 1.3   1.4
 0.8   0.9
 1.9 or less
 0.85
 1.5   1.7
  5

  900
  70
  8
  2
 * Other sizes of carbon are available  on  request from the manufacturers.
'* No available  data from  the manufacturer.
— Not applicable to this size carbon.
  WESTVACO
   NUCHAR
     WV-L
    (8x30)
(BITUMINOUS)


  1000
  0.48
  26
  2.1
  1.4
  0.85   1.05
  1.8 or  less
  0.85
  1.5  1.7
 5

 350
 70
 7.5
 2
    WITCO
     517
   (12x30)
(BITUMINOUS)


    1050
    0.48
    30
    2.1
    0.92
    0.89
    1.44
    O.bO
    1.2
    5
    5

    1000
    85
    0.5
    1
                                       TYPICAL PROPERTIES OF 8 X 30-MESH CARBONS
                                                                                    [9]
          Total  surface area, m2/g
          Iodine number, min
          Bulk  density, Ib/ft3 backwashed and drained
          Particle density wetted  in water, g/cm3
          Pore  volume, cm3/g
          Effective size, mm
          Uniformity coefficient
          Mean-particle dia., mm
          Pittsburgh abrasion number
          Moisture as  packed, max.
          Molasses RE (relative efficiency)
          Ash
          Mean-pore radius
                                                                        LIGNITE
                                                                        CARBON

                                                                        600-650
                                                                          600
                                                                           22
                                                                         1.3-1.4
                                                                           1.0
                                                                        0.75-0.90
                                                                       1.9  or less
                                                                           1.5
                                                                         50-60
                                                                           9%
                                                                        100-120
                                                                         12-18%
                                                                          33 A
                                                         BITUMINOUS
                                                         COAL CARBON

                                                           950-1,050
                                                             950
                                                              26
                                                            1.3-1.4
                                                             0.85
                                                            0.8-0.9
                                                           1.9 or less
                                                              1.6
                                                             70-80
                                                              2%
                                                             40-60
                                                             5-8%
                                                             14 A

-------
 322
            TABLE 6 "TYPICAL  PROPERTIES OF POWDERED ACTIVATED CARBON (PETROLEUM BASE)1
Surface Area (BET m /gm)
Iodine No.
Methylene Blue Adsorption (mg/gm)
Phenol  No.
Total Organic Carbon  Index (TOCI)
Pore Distribution (Radius Angstrom)
Average Pore Size (Radius Angstrom)
Cumulative Pore Volume (cc/gm)
Bulk Density (gm/cc)
Particle Size     Passes:  100 mesh (wt%)
                        200 mesh (wt%)
                        325 mesh (wt%)
Ash (wt%)
Water Solubles (wt%)
pH of Carbon
                                                                                   !,300
                                                                                   !,700
                                                                                    400
                                                                                     10
                                                                                    400
                                                                                     15
                                                                                     20
                                                                                    0.1-
                                                                                   0.27
                                                                                     97-
                                                                                     93-
                                                                                     85-
                                                          •2,600
                                                          •3,300
                                                          •600
                                                          12
                                                          •800
                                                          60
                                                          30
                                                          0.4
                                                          0.32
                                                          100
                                                          98
                                                          95
                                                         .5
                                                         .0
                                                         -9
     Stage

 Drying

 Thermal Desorption

 Pyrolysis and Carbonization


 Gasification
TABLE 7 "STAGES OF THERMAL REGENERATION"

Approximate Temperature ( F)                     Processes

       Ambient to 212

       212 to 500

       400 to 1, 200
     1,200 to 1,900
                                                    Water Evaporation

                                                    Physical desorption of volatile adsorbed organics

                                                    Pyrolysis of nonvolatile organics and
                                                    carbonization of the pyrolysis residue

                                                    Gasification of pyrolytic residue through
                                                    controlled chemical reaction with water vapor,
                                                    vapor, carbon dioxide, or oxygen
                      TABLE 8 "WATER QUALITY AT VARIOUS POINTS  IN  PROCESS" (16)
Quality Parameter
BOD (mg/l)
COD (mg/l)
Total organic carbon (mg/l)
ABS (mg/l)
PO4 (mg/l  as PO4)
Color (units)
Turbidity (units)
Nitrogen organic N (mg/l as N)
Ammonia N (mg/l as N)
NO- and NO- (mg/l as N)
UncRlorinated:  Coliforms (MPN/100 ml)
               Fecal coliforms (MPN/100 ml)
               Virus
Chlorinated:    Coliforms (MPN/100 ml)
               Fecal coliforms (MPN/100 ml)
                    Raw
                Wastewater

                  200-400
                  400-600

                  2.0-4.0
                   10-15
                   25-35
                      0
                                                 Secondary
                                                 Effluent

                                                 20-100
Separation-
Bed Effluent
   Carbon-
Column Effluent
80-160
-
0.4-2.9
25-30
-
30-70
4-6
25-32
0
2,400,000
150,000


20-60
8-18
0.4-2.9
0.1-1
10-30
<0.5-3
2-4
25-32
0
9,300
930
Negative
8.6
1-25
1-6
<0.01-0.5
0.1-1
<5
<0.5-1
1-2
25-32
0
1 1 ,000
930
Negati\
<2.1

-------
                                                                                                323
          TABLE 9 "WATER QUALITY AT VARIOUS STAGES OF TREATMENT AT SOUTH LAKE TAHOE1

                                                        Effluent
Quality
Parameter
BOD (mg/1)
COD (mg/1)
SS (mg/1)
Turbidity (JTU)
MBAS (mg/1)
Phosphorus (mg/1)
Col i form
(MPN/lOOml)
Raw
Wastewater
140
280
230
250
7
12 6
50 x 10


Primary
100
220
100
150
6
9 6
15 x 10


Secondary
30
70
26
15
2.0
6 6
2.5x 10

Chemical
Clarifier


10
10

0.7



Filter
3
25
0
0.3
0.5
0.1
50


Carbon
1
10
0
0.3
0.1
0.1
50

Chlorinated
Final
0.7
10
0
0.3
0.1
0.1
<2

           TABLE 10 "CARBON EFFICIENCY PER REGENERATION PERIOD AT SOUTH LAKE TAHOE1
                               NOVEMBER 1968 THROUGH JANUARY 1971
            Parameter

Carbon Dosage
 (Ib regenerated/million gallons treated)
Iodine Number
Apparent Density (gm/ml)
          2
Percent Ash

Chemical Oxygen Demand
                                 Spent Carbon
                                 Regenerated Carbon
                                 Spent Carbon
                                 Regenerated Carbon
                                 Spent Carbon
                                 Regenerated Carbon
                                 Percent Removal
                                 Ib COD applied
                                 Ib COD applied/MG
                                  Ib COD removed/MG
                                  Ib COD applied/lb ]
                                   carbon regenerated
                                  Ib COD removed/lb
                                   carbon regenerated
Methylene Blue Active Substances (Methylene Blue Active
 Substances (MBAS))                Percent removal
                                  Ib MBAS applied
                                  Ib MBAS applied/MG
                                  I b MBAS removed/MG
                                  Ib MBAS applied/lb
                                   carbon regenerated
                                  Ib MBAS removed/lb
                                   carbon regenerated
                                                          Average

                                                            207
Maximum

  418
Minimum
  111
583
802
0.571
0.487
6.4
6.8
49.9
28,250
162
81
0.78
0.39
77
995
5.7
4.4
0.027
0.021
633
852
0.618
0.491
7.0
7.2
63.3
54,970
254
149
1.56
0.71
93
1,675
10.7
8.2
0.045
0.039
497
743
0.544
0.478
5.8
5.8
30.1
15,680
105
32
0.52
0.16
58
457
2.6
1.6
0.012
0.007
1          3                                3
-Based on ft  of carbon fed to furnace at 30 Ib/ft
 November 1968 through November 1970

-------
324
Solid Contact Clarifier
   TABLE 11 "COLORADO SPRINGS TERTIARY PLANT DESIGN DATA"

                  48 ft diameter, 11 ft 9 in sidewall,  2 ft cone depth
                  1,809 sq ft surface area         168,465 gallon capacity
                  2 hr detention time and 0.76 gpm/sq ft rise rate  @ 2 MGD flow
Spent Lime Tank


New Lime Holding Tank


Recarbonization Tank


Activated Carbon Adsorbers
 Dual Media Sand  Filters
                  14 ft diameter, 10 ft sidewall depth, 8 ft cone
                  1,948 cu ft                    14,610 gallon capacity

                  14 ft diameter, 10 ft sidewall depth, 8 ft cone
                  l,948cuft                    14,610 gallon capacity

                  14 ft diameter, 14 ft sidewall depth, plus 2 ft freeboard - 154 sq ft
                  2, 156 cu ft  16, 160 gallon capacity - 12 minutes detention time @ 2 MGD flow

                  20 ft diameter, 14 ft sidewall depth, 10 ft of carbon media
                  314 sq ft surface area, 4,396 cu ft    32,970  gallon capacity
                  total tower detention  time is 24 minutes - 4.5 gpm/sq ft (@ 2 MGD flow)
                  carbon bed detention  time is 17 minutes  4.5 gpm/sq ft (@ 2 MGD flow)
                  3,140 cu ft of carbon £ 30 Ib/cu ft   94,200  Ibs carbon
                  1 2 ft diameter, 1 1 ft sidewall depth, 1  ft 6 in top cone
                  113 sq ft surface  area
                  1,243 cu ft capacity           9,323 gallon capacity
                  3 ft of e.s. 1.5 mm sand       5 ft of e.s. 2.8 mm anthrafilt
 PARAMETER

   BOD
   COD
    TOC
    TSS
    Turbidity
    O.PO  (total)
    O.PO^ (soluble)
    MB AS
    PH
    Alkalinity (total)
    Hardness
    C  (as C    )
    C C  (as C COJ
    r- i a      a   •*
    Color
    SuI fates
    Sulfides
    Flow (MGD)
   Lime Dose  (ppm Ca 0)
TABLE 12 "TERTIARY TREATMENT PLANT DATA SUMMARY (FIRST REGIME)"

         R-C INF.      R-C EFF.    FILTER EFF.    LEAD CARBON EFF.  POLISH CARBON EFF.
106
305
76
56
53
32
31
5.32
7.35
197
163
52
130
162
88
0.15
1.93
46.9
120
37.9
31.0
10.1
2.2
0.16
3.22
11.44
282
210
79
197
44
80
0.14
1.93
                                        48.0
                                       114
                                        36.8
                                         3.7
                                         4.2
                                         1.7
                                         1 .6

                                         7.93
                                        69
                                       250
                                        95
                                       237
                                        38
                                       439

                                         1.93
 44.0
 83
 27.9
  2.1
  3.6
  2.2
  2.2
  1.36
  6.91
 63
240
 84
209
 25
439
  0.26
  1.70
 33.0
 64
 23.8
  3.7
  3.2
  1.9
  1.8
  0.51
  7.12
 69
253
 92
231
 13
401
  0.41
  1.70
                     349
 NOTE:  During this period,  the reactor was operated with a deep sludge blanket of approximately six feet.

-------
                                                                                                 325
                TABLE  13 "TERTIARY TREATMENT PLANT DATA SUMMARY (SECOND REGIME)'
PARAMETER
   BOD
   COD
   TSS
   Turbidity
   0. Phosphate (total)
   0. Phosphate (soluble)
   MBAS
   Fecal Coliform
   Fecal Strep
   Average Flow (MGD)
   Average Lime Dose
   Lead  Tower
   Polish Tower
               INFLUENT
                 (mg/1)*

                  102
                  258
                   62
                   54
                   30
                   26
                    4.2
            5 x 105/100 ml
            6.5x 10 /100ml
                      EFFLUENT
                        (mg/1)*

                         8.0
                        15.7
                         2.2
                         2.2
                         1.0
                         1.0
                         0.15
                      225/100 ml
                     1150/100 ml
                         1.5
                       345 mg/1
5 ft of twice-regenerated + 5 ft of virgin carbon
8 ft of twice-regenerated + 2 ft of virgin carbon
   *All units in mg/1 unless indicated.
   The "INFLUENT" is the secondary effluent going to the solids contact clarifier.
   The "EFFLUENT" is the final polish carbon tower effluent.
   Influent to tertiary plant is from a trickling filter plant with a flow of
   23 to 25 MGD; trickling filter plant capacity is 13 MGD.
REMOVAL
(percent)
  92,
  93.
  96,
  96,
  96,
  96,
  96.4
  99.96
  99.82
        TABLE  14 "COMPARISON OF WASTEWATER STRENGTH BATTELLE AND CRSD TEST PROGRAMS"

                                                                   Probability (% of Occurrences)

                                                           10                   50                90
Battelle Series (1970-71) BOD
CRSD Series (1974-75) BOD
Battelle Series (1970-71) COD
CRSD Series (1974-75) COD
Battelle Series (1970-71) BOD/COD
CRSD Series (1974-75) BOD/COD
                                         180 mg/1
                                         115 mg/1
                                         320 mg/1
                                         215 mg/1
                                          0.56
                                          0.53
                                 240 ma/I
                                 170 mg/1
                                500 mg/1
                                350 mg/1
                                  0.48
                                  0.48
 320 mg/1
 250 mg/1
720 mg/1
580 mg/1
  0.44
  0.43
Parameter
TABLE 15 "EFFECT OF POST-OZONATION ON EFFLUENT ORGANIC QUALITY"

                      CRSD  PILOT-PLANT TEST RESULTS*

                            Before Ozonation                             After Ozonation
BOD (unfiltered)
BOD (filtered)
COD (unfiltered)
COD (filtered)
TOC
                                32 mg/1
                                25 mg/1
                                72 mg/1
                                59 mg/1
                                31 mg/1
                                               31 mg/1
                                               24 mg/1
                                               66 mg/1
                                               52 mg/1
                                               29 mg/1
*AII data represent average values over an eleven day period.  Ozone dosage ranges from 4  9 mg/1.

-------
326
                                          TABLE 16 "QUALITY OF RAW WASTEWATER"

                                     GARLAND PHYSICAL/CHEMICAL TREATMENT FACILITY

                 Parameter                                                          Results

                 Total BOD (mg/l)                                                     266
                 Filtered BOD (mg/l)                                                   236
                 Total COD (mg/l)                                                     542
                 Filtered COD (mg/l)                                                   240
                 Suspended Solids (mg/l)                                                233
                 Alkalinity (mg/l as CaCOJ                                            200
                 PH                                                                7.2-7.7
                 P04 (mg/l)                                                            15


                                          TABLE 17 "PCT SYSTEM DESIGN DATA" (24)

           CHEMICAL TREATMENT SYSTEM

                 1 .    Flocculation:  Detention Time, minutes                             45
                      Chemical Dosage             Alum,  mg/l Al                      22
                                                  polymer, mg/l                       0.25

                 2.    Sedimentation:   Detention Time, hrs.                                1.5
                      Overflow Rate,  gpd/sq ft                                       900
                      Underflow, % of plant flow                                        1.25
                      Underflow solids, % by weight                                     2

                 3.    Gravity Thickening: Solids Loading, Ib/day-sq ft                   12
                      Underflow solids, % by weight                                     4

                 4.    Vacuum Filtration:  Yield, Ib/hr-sq ft                               2
                      Cake solids                                                     18

                5.    Sludge Incineration:  Solids Loading, Ib/hr-sq ft                     2

           CARBON TREATMENT SYSTEM

                1.    Carbon Contacting (8 x 30 mesh carbon)  Empty-bed contact time      25
                      Hydraulic surface Loading, gpm/sq ft                               4
                      Backwash  volume, % of plant flow                                  5
                      Sodium Nitrate dosage, mg/l N                                    5.5
                      Carbon Dosage,  Ib/MG                                         250
                      Carbon Regeneration loss, %                                      5

-------
                                                                                                327
           TABLE 18 "SUMMARY OF PHYSICAL/CHEMICAL TREATMENT SYSTEM PERFORMANCE" (24)

                                                                Average Percent Removal
 Parameters

 Suspended Solids (mg/l)
 Turbidity (JTU)
 TCOD (mg/l)
 DCOD (mg/l)
 BOD (mg/l)
 Total  Phosphate (mg/l P)
 Nitrate (mg/l N)
 Color
 PH

 Notes:  1.  Average alum dosage = 25 mg/l Al  (275 mg/l alum)
         2.  Average polymer dosage = 0.3 mg/l Calgon WT-3,000 _
         3.  AS Effluent refers to Activated Sludge Plant Effluent (8 MGD, existing facility)
Raw
Sewage
199

321
49.4

11.1


7.7
Clarified
Effluent
28.3
22.9
95.8
48.6
36.2
1.3
0.9
20
6.8
Carbon
Effluent
6.7
6.3
19.3
13.5
7.8
0.9
1.3
7.8
6.8
Chemical
Treatment
85.8

70.2
1.6

88.3



Carbon
Treatment
76.3
72.5
79.9
72.2
78.5
30.8

61.0


Overall
96.6

94.0
72.7

91.9




AS Effluent*
11.6
7.7
39.5
25.7
8.0


33.1

               TABLE 19 "EFFECT OF REGENERATION ON THE PCT CARBON  CHARACTERISTICS"

                                    Spent Carbon       	Regenerated Carbon (Composite  Sample)
(Composite Sample)
Carbon
Characteristics
Virgin
Carbon
1st
Reg.
2nd
Reg.
3rd
Reg.
Before Quenching
1st
Reg.
2nd
Reg.
3rd
Reg.
After Quenching
1st
Reg.
2nd
Reg.
3rd
Reg.
Iodine No.  (mg/l)

Apparent Density
  (g/cm cu  ft)

Molasses No.

Methylene Blue No.
  (mg/g)

Ash (%)

Mean Particle Dia.
  (mm)
1,040    402    572     570    805     722     773

0.484  0.629  0.585   0.594  0.528   0.537   0.526
                         751     727     721

                       0.565   0.548   0.535
222
259
6.4
1.44
120
147
10.3
1.46
168
153
8.22
1.58
                        154

                        153
213

223
                       8.67   10.7

                       1.48   1.57
 233

 243


11.6

1.50
 230

 246


7.81

1.54
 189

 227


12.0

1.55
 221

 239


12.2

1.50
 204

 245


 9.0

1.43

-------
                                                                 TABLE  20
TERTIARY TREATMENT PLANTS

Site
Arlington, Virginia
Colorado Springe, Colo.
Dallas, Texas
Fairtax County, Va.
Los Angeles, Calif.
Montgomery County, Md.
Occoquan, Va.
Orange County, Calif.
Piscataway, Md.
St. Charles, Missouri
South Lake Taho, Calif.
Windhoek, South Africa

Status
1973
Design
Operating Dec. '70
to Present
Design
Design
Design
Design
Design
Construction
Operating Mar. '73
to Present
Construction
Operating Mar. '68
to Present
Operating Oct. '68
to Present

Average
Plant
Design Capacity
Engineer (MDG)
Alexander Potter/ 30
Engineering-Science
Arthur B. Chafet 3
& Assoc.
URS Forest & 100
Cotton
Alexander Potter/ 36
Engineering-Science
City of Los Angeles 53
CH2M/Hill 60
CH2M/Hill 18
Orange County Water 15
District
Roy F . Weston 5
Moran and Cooke 5.5
CH2M/Hill 7.5
National Institute 1.3
for Water Research
Pretoria, So. Africa


CO
KJ
OO
Total
No. of Contact Hydraulic Carbon Effluent
Contactor Contactors Time1 Loading Depth Carbon Requirements2
Type In Series (Min.) (gpm/ftz') (ft) Size (Oxygen Demand)
Down flow 1
Gravity
Down flow 2
Upflow 1
Packed
Down flow 1
Gravity
Downflow 2
Gravity
Upflow 1
Packed
Upflow 1
Packed
Upflow 1
Packed
Downflow 2
Pressure
Downflow 1
Gravity
Upflow 1
Packed
Downflow 2
Pressure
38 2.9 15 8 x 30 BOD
30 5 20 8 x 30 BOD
10 8 10 8 x 30 BOD
BOD
(by
36 3 15 8 x 30 BOD
50 4 26 8 x 30 COD
30 6.5 26 8 x 30 BOD
COD
30 5.8 24 8 x 30 BOD
COD
30 5.8 24 8 x 30 COD
37 6.5 32 8 x 30 BOD
30 3.7 15 8 x 30
17 6.2 14 8 x 30 BOD
COD
30 3.8 15 12 x 40 COD
< 3 mg/1
< 2 mg/1
< 10 mg/1
< 5 mg/1
1980)
< 3 mg/1
< 12 mg/1
< 1 mg/1
< 10 mg/1
< 1 mg/1
< 10 mg/1
< 30 mg/1
< 5 mg/1

< 5 mg/1
< 30 mg/1
< 10 mg/1
1   Empty bed  (superficial) contact time
   for average  plant flow.
2  BOD:   Biochemical oxygen demand
   COD:   Chemical oxygen demand
50 MGD ultimate capacity

-------
                                                            TABLE 21
PHYSICAL/CHEMICAL TREATMENT PLANTS

Site
Cortland, New York
Cleveland Westerly,
Ohio
Fitchburg,
Massachusetts
Garland, Texas
LeRoy, New York
Niagara Falls,
New York
Owosso, Michigan
Rosemount,
Minnesota
Rocky River,
Ohio
Vallejo,
California

Status 1973
Design
Construction
Construction
95% Complete
Construction
Design
Construction
50% Complete
Design
Operational
Operational
Design

Average
Plant Contact
Ca£acity Time1
Design Engineer (MGD) (Min.)
Stearns & Wheeler 10 30
Engineering-Science 50 30
Camp Dresser & 15 35
McKee
URS Forest & 30 30
Cotton
Lozier Engineers 1 27
Camp Dresser & 48 20
McKee
Ayres, Lewis, 6 36
Norris & May
Banister, Short, 0.6 66
Elliot, Hendrickson (max)
and Associates
Willard Schade & 10 26
Assoc.
Kaiser Engineers 13 26

Total
Hydraulic Carbon Effluent
Loading Depth Carbon Requirements2
(gpm/TtZ) (ft) Size (oxygen Demand)
4.3 17 8 x 30 TOD 30 mg/1
3.7 17 8 x 30 BOD 20 mg/1
3.3 15.5 8 x 30 BOD 10 mg/1
2.5 10 8 x 30 BOD 10 mg/1
7.3 26.8 12 x 40 BOD 10 mg/1
3.3 9 8 x 30 COD 112 mg/1
6.2 30 12 x 40 BOD 10 mg/1
4.2 36 12 x 40 BOD 10 mg/1
(max)
4.3 15 8 x 30 BOD 15 mg/1
4.6 16 12 x 40 BOD 45 mg/1
(90% of time)
1   Empty bed (superficial)  contact  time for average plant flow.

2   BOD:   Biochemical  oxygen demand
   COD:   Chemical  oxygen demand
                                                                                                                                        CO

-------
 330
           TABLE 22 "SUMMARY OF PCT PILOT-PLANT AND FULL-SCALE PLANT PERFORMANCES"

                            Effluent                      Effluent                      Effluent
Blue Plains
 Pilot Plant
Owosso,
 Michigan
Pomona,
 California
Rosemount,
 Minnesota (1st year
Rosemount, Minnesota
 (last 3 to 4  months)
Battelle Pilot Plant
 at Westerly
CRSD Pilot Plant
 at Westerly
 Estimated based on BOD similar to COD removals across clarifier.
2
 Just around  carbon columns.
Raw COD
(mg/l)
320
250-350
321

;ota
«)
t 527
437
COD
16
24-30
19


42
56
% Raw TOC
Removal (mg/l)
95 100
-91
94
-

92
87 90
TOC % Raw BOD
(mg/l) Removal (mg/l)
8 92 150
140
1201
230

240
21 77 206
BOD
(mg/l)
6
8
7.8
23

26
32
RemovaJ
96
84
78.
90

89
84

52




     TABLE 23 "CARBON PILOT-PLANT RESULTS FOR PETROCHEMICAL AND REFINING WASTEWATERS"

Type of Wastewater
Refinery
Refinery
Refinery
Petrochemical
Refinery
Refinery
Refinery
Petrochemical
Design Q
(MGD)
28
1.9
22
3
26
28
8
29
Process
Application
Physical/Chemical
Physical/Chemical
Physical/Chemical
Tertiary
Tertiary
Tertiary
Tertiary
Tertiary
Influent
COD (mg/l)
600
800
670
150
100
300
100
150
Effluent
COD (mg/l)
103
201
143
49
41
50
40
48
Percent
Removal
83
75
79
67
59
83
60
68

-------
                                                                                                 331
                         TABLE 24 "REFINERY WASTEWATER TREATMENT RESULTS"
Parameter

BOD (mg/l)
COD (mg/l)
TOC (mg/l)
Oil & Grease (mg/l)
Phenols (mg/l)
Chromium (mg/l)
Copper (mg/l)
Iron (mg/l)
Lead (mg/l)
Zinc (mg/l)
Sulfide (mg/l)
Ammonia (mg/l)
Cyanides (mg/l)
Turbidity, (TTU)
Color (Std. Color Units)
API Separator
Effluent
97
234
56
29
3.4
2.2
0.5
2.2
0.2
0.7
33
28
0.25
26
30
Carbon Treated
Effluent
48
103
14
10
0.004
0.2
0.03
0.3
0.2
0.08
39
28
0.2
11
15
Biologically Treated
Effluent
7
98
30
10
0.01
0.9
0.1
3
0.2
0.4
0.2
'27
0.2
17
15
                                                           Biological-Carbon
                                                           Treated Effluent

                                                                  3
                                                                 26
                                                                  7
                                                                  7
                                                                  0.001
                                                                  0.02
                                                                  0.05
                                                                  0.9
                                                                  0.2
                                                                  0.15
                                                                  0.2
                                                                 27
                                                                  0.2
                                                                  5
                                                                  1
                        TABLE 25 "CARBON REGENERATION ACTIVITY ANALYSIS"

                                                               Iodine No.
                     Virgin Carbon
                     API Separator Effluent to Carbon
                     Biologically Treated Effluent to Carbon
                                      1,010
                                       906
                                       991
Molasses No.

   216
   405
   304
COD (mg/l)
SOC (mg/l)
BOD (mg/l)
                   TABLE 26 "PILOT-PLANT RESULTS-TERTIARY CARBON APPLICATION"
   Influent
Concentration

     600
     500
     400
     300
     200

     300
     200
     100
      50

     250
     200
     150
     100
      50
      20
^Throughput Rate = 0.6 - 1.2 Bed Volumes per hour
^Throughput Rate = 2.4 Bed Volumes per hour
 Throughput Rate = 1.2 Bed Volumes per hour
Effluent
Concentration
280 ]
230
175
120
65
1402 2503
95 155
30 55
10 25
884 21 85
70 186
54 135
36 68
20 30
10 12
2
.Throughput Rate
Throughput Rate
Average
Percentage Removal
531
54
56
60
68
532 173
53 23
70 45
80 50
654 135
65 7
64 10
64 32
60 40
50 40
= 0.5 Bed Volumes per hour
= 0.15 Bed Volumes per hour

-------
332
                  TABLE 27 "DESIGN CRITERIA FOR THE ARCO CARBON PLANT"
         Number of Rainfall Days Per Year (max.)
         Maximum Rainfall Runoff Rate (per day)
         Maximum Rainfall Runoff Rate (per year)
         Average  Influent COD Concentration
         Average  Effluent COD Concentration
         Carbon Capacity
                                                           30       days
                                                           12.6     Million Gallons
                                                          378       Million Gallons
                                                          250       mg/l
                                                           37       mg/l
                                                            1       Ib carbon exhausted per
                                                                    1000 gal. water treated (1.75
                                                                    Ib COD removed/lb carbon)
TABLE 28 "ACTIVATED CARBON ADSORPTION DESIGN DATA AND ACTIVATED CARBON PROPERTIES"
Rated Flow (Each of Three Adsorbers)
Adsorber Diameter
Adsorber Bed Depth
Contact Time (Empty Bed)
Hydraulic  Loading
Design Inlet Pressure
Pressure Drop Through Carbon
Carbon Inventory    Carbon Bed
                   Adsorber Total
Theoretical Carbon Capacity
Carbon Dosage
                                                                   667
                                                                             gpm
10
45
40
8.5
60
35
92,000
100,000
0.3
ft
ft
min
gpm
psi
psi
Ib
Ib
IbT
                                                                             Ib TOC/lb carbon
                                                     0.86 Ib carbon/1,000 gal Throughput
                             Activated Carbon Properties Filtrasorb 300
                                    (8 x 30 Bituminous Coal)
         Total Surface Area (N, BET Method)
         Bulk Density
         Particle Density Wetted in Water
         Mean Particle Diameter
         Iodine Number, minimum
         Ash
         Moisture
                                                     950-1050
                                                           26
                                                      1.3-1.4
                                                      1.5-1.7
                                                          950
                                                      Max  8%
                                                      Max  2%
                 m2/
                 Ib/ft"
                 g/cc
                 mm
                      TABLE 29 "THERMAL REGENERATION DESIGN DATA"
         Furnace
         Regeneration Rate
         Steam Addition Rate
         Fuel
         Fuel Rate
        Combustion Air Rate
        Design Temperatures
                   Hearth 4
                   Hearth 6
                   Afterburner
                   Hearth 4
                   Hearth 6
                   Afterburner
                   Hearth 4
                   Hearth 6
                   Afterburner
   60" x 6 Hearth with Integral Afterburner
   125 Ib/hr
   125lb/hr
   Refinery Fuel  Gas
   188CFH
    68 CFH
   310 CFH
 5,000 CFH
 1,800 CFH
 8,120 CFH
1,725° F
1,750° F
1,250° F

-------
    FIGURE  I
    ADSORPTION OF  DBS  TO  EQUILIBRIUM
    IN  CONTINUOUSLY STIRRED
    FLOW  SYSTEMS [2]
 160


 140


 120


 100


 80


 60


 40


 20
T	1	1	T
                        T	T
                                 k, = 1.25 mg/g-hr
                     = 1.78 mg/g-hr
          k, = 2.80 mg/g-hr
      -k0 = 4.28 mg/g-hr
           \	I
                    1     I     I     I
               15    21    27    33   39

                    TIME (hours)
                                  45   51
ORGANIC CONSTITUENT IS DODECYL-BENZENESULFONATE (DBS)
57
                                                        333
 2
 O
 00
 rr
 <
 o
 o
 £
O
o
o
          Figure  2

          FREUDLICH  ISOTHERM APPLICATION

          ' DATA CONFORM TO
          -FREUNDLICH
          - ISOTHERM
                     DECREASING
                      WASTEWATER
                       COMPLEXITY
            NON-SORBABLE
              RESIDUAL
                             FREUNDLICH ISOTHERM
                             APPLICABILITY
                             RESTRICTED TO
                             DEFINED LIMITS
          EQUILIBRIUM CONCENTRATION, C  (mg/l)

-------
    334
FIGURE 3
CROSS-SECTIONAL VIEW
OF  MULTIPLE-HEARTH FURNACE
                         CARBON IN
                               GAS OUT
                      Rabble Arm
                     Rabble Teeth
                  Steam
                                                      HEARTH
                                                    I  (200°-300°F)
                                                    2 (300°-450°F)
                                 3 (400°- IOOO°F)
                                                    4 (IOOO°-I600°F)
                                                    5  (I600°-I800°F)
                                                    6  (1600°-1800° F)
                                                     CARBON OUT
FIGURE 4
SIMPLIFIED FLOW  DIAGRAM
                      SOUTH TAHOE
              0=]
MIXED
MEDIA
FILTRATION


MIXED
MEDIA
FILTRATION


RECARBONATION
AND
SETTLING



AMMONIA
STRIPPING
TOWER


                                                            CHEMICAL
                                                            CLARIFICATION
                                                 FINAL EFFLUENT
                                                 PUMP STATION

-------
FIGURE 5
SIMPLIFIED FLOW  DIAGRAM
             COLORADO  SPRINGS  PLANT
     Recycle
     Effluent from
     Primary Clarifier
         Clarifier
  Waste
  Solids
  Return
                       High  Rate Filter

                    Dist.  Box

,
J
ary
er


Dist. Box
Chlorination
Chamber
Irrigation
Canal
                                         Reactor
                                         Clarifier
                                       Recarbonation
                                       Reactor C02
                                           H2S04
                            Carbon Columns
                                                                                  LAS Hi
                                                                                   MIX! "\
                              *- FLOCCULATOR


                              	1
                                                                         Raw
                                                                         Waslewater
FIGURE  6

SCHEMATIC  FLOW  DIAGRAM
WESTERLY ADVANCED WASTE TREATMENT  PLANT
PILOT PLANT
HELECTROLYTIC UNIT
FOR HYPOCHLORITE
GENERATION
                                                                                                                                                                CO
                                                                                                                                                                CO
                                                                                                                                                                Oi

-------
      FIGURE 7

      FREQUENCY ANALYSIS FOR EFFLUENT  BODL25j
      CRSD PRECARBON OZONATION STUDY
      WESTERLY PILOT PLANT
                                                                          FIGURE 8                                       r

                                                                          FREQUENCY  ANALYSIS  FOR EFFLUENT COD1
                                                                          CRSD PRECARBON OZONATION STUDY
                                                                          WESTERLY PILOT PLANT
o
o
00
LJ
 _
Ld
100
 90
 80

 70

 60

 50


 40



 30




 20
      2    5   10    20   30  40  50 60  70  80   90  95

         % OF THE VALUES LESS THAN STATED VALUE



  •   NO  PRECARBON OZONATION

  A   OZONE  <6.2 mg/l

  O   OZONE  =6.2 mg/l

  Q   OZONE  =3.2 —6.2 mg/l
                                                       98 99
Q
O
O
                                                                       LJ
                                                                       UJ
        2    5   10   20  30 40 50 60 70  80   90  95   98

            % OF THE VALUES LESS  THAN STATED  VALUE



     9   NO PRECARBON OZONATION

     A   OZONE  <6 2 mg/l

     O   OZONE  =6.2 mg/ I

     a   OZONE  =3.2 —6.2 mg/l
 CO
 CO

-------
                                                                             337
FIGURE 9                                         r 1
FLOW  DIAGRAM  FOR  THE GARLAND  PLANTLJ
INFLUENT
t
t
IEQUALIZATION!
<

IPRETREATMENTI



1
* PRIMARY
9 CLARIFIER
1
4-
SLUDGE
CONDITIONING
"


TRICKLING
"" FILTERS

	 FILTER
PRESS
1


llNCINER/i
1

CHEMICAL
9 CLARIFICAT^

ILANDFI



TION|
LQ
	 1
1
FINAL
CLARIFIER



j -^RECARBONATIONl-^FILTRATIONlJ






L CARBON
ADSORPTION
< '
(DISINFECTION |
EFFLUENT
 FIGURE 10
 PHYSICAL/CHEMICAL PILOT PLANT SCHEMATIC
                                    POMONA, CALIFORNIA
         FLOCCULATION
                                           ATMOSPHERE
                                      ""~r
                                       1   MAKE-UI-
                                       T r—WATER
                                        buENCH
                                      I  JTANK
                                     — ^t-*—v
                                      EDUCTOR
MOTIVE WATER
                        COMPRESSED
                        AIR FOR
                       • PULSED A/ff
                        CLEANING OF BAGS

-------
    338
            g>
            en
            Q.
            CO
            V)
            o
            LU
            X
            LJ
FIGURE II
SULFIDES  AND HEADLOSS vs TIME
                                                                    CT
                                                                    E
                                                                    CO
                                                                    LU
                                                                    O
                                                                    to
                 0
                  0
             345    678
             STUDY  PERIOD   (weeks)
      FIGURE 12
      BOD  REMOVAL  TREATMENT  RESULTS
Q
O
CD
                                                                             API EFFLUENT
                                                        A BIOLOGICALLY
                                                          TREATED
                                                          EFFLUENT

                                                        O CARBON TREATED
                                                          EFFLUENT


                                                        D CARBON TREATED
                                                          BIOLOGICAL
                                                          EFFLUENT
                              567    8    9    10

                                  STUDY PERIOD (days)
                                            12
13
14

-------
FIGURE 13

COD  REMOVAL  TREATMENT  RESULTS
                                                                                   339
                           5   67   8   9   10  II

                           STUDY PERIOD (days)
                                             12  13  14
                                                                    API EFFLUENT
                                                                 &  BIOLOGICALLY
                                                                    TREATED
                                                                    EFFLUENT

                                                                 O  CARBON TREATED
                                                                    EFFLUENT


                                                                 D  CARBON TREATED
                                                                    BIOLOGICAL
                                                                    EFFLUENT
    FIGURE 14

    ARCO FLOW  DIAGRAM
               WATSON  REFINERY
PROCESS
WASTEWATER
         API SEPARATOR
             I
       NEUTRALIZATION
             I
          CHEMICAL
         FLOCCULATION
             I
        DISSOLVED AIR
          FLOTATION
             T
To LACSD
                                                   HOLDING BASIN FLOW TO
                                                      LACSD OR TREATMENT
                                                      SYSTEM DEPENDING
                                                      ON WATER QUALITY
                                                                 CARBON
                                                                 HOPPERS
          FLOW DIVERTED TO HOLDING BASIN
             WHEN RAINFALL EXCEEDS
             O.I INCHES
                                                    REGENERATION
                                                       FURNACE
                                                                              ~l
                                                                        L	1
                                                               *• FINAL EFFLUENT
                           	CARBON	

-------
       FIGURE 15

       PERFORMANCE OF ARCO CARBON PLANT
                                      COD REMOVAL
    400
O
-3-
CO
                 10   20  30 40 50 60 70 80   90

            %  OF THE VALUES LESS THAN STATED VALUE
                                               99
                                                                     FIGURE 16

                                                                     PERFORMANCE OF ARCO CARBON PLANT
                                                                                                    08G REMOVAL
I          10   20 30 40 50 60 70  80  90

    %  OF THE VALUES LESS THAN STATED  VALUE

INFLUENT •                                EFFLUENT
D INFLUENT!
                                        O EFFLUENT

-------
     FIGURE 17

     CARBON  TREATMENT  SYSTEM
                    REICHHOLD CHEMICALS CO.
                                                                        341
     Influent
               1
           EQUALIZATION
  Nonionic
     Poly
     Acid.
ACID  MIXING
TANK
           FLOCCULATION
           TANK
           CARBON
           ADSORBER
           FEED SUMP
MULTIPLE-
  HEARTH
  FURNACE
                                                                        CARBON
                                                                        ADSORBERS
                                Sludge To
                                Disposal
                                                               V
                                                                  I
                                                                  I
                                                                  I
                                                                  I
                                    	CARBON	
   FIGURE 18

   BP  TREATMENT  SYSTEM  FLOW  DIAGRAM
                                MARCUS HOOK  REFINERY
                                          Backwash
                API Bottoms
                                                                                 itrifuge
  API SEPARATOR        SURGE BASIN

            r--|	T	1
            1000
     Treated
     Process
     Effluent
Once- through
Cooling Water

^

s<
\
Si
^

                                                        MULTIPLE-
                                                        HEARTH
                                                        FURNACE


                                                        	 Fuel
                                                        -— Steam
                                                                 QUENCH
                                                                 TANK
            Air
                 CARBON ADSORBERS

-------
         FIGURE 19
         COD  REMOVAL AS A FUNCTION r  i
         OF ADSORBER  CONTACT TIMEl32J
                                PERIOD I
                                                         FIGURE 20
                                                         ADSORBER  OIL REMOVAL  AS  A         r ,
                                                         FUNCTION  OF  INFLUENT  CONCENTRATION L32J
                                                                                                    PERIOD  I
   O
   •^
   LJ
   (T

   Q
   O
   CJ>
100



90



80



70


60


50



40


30


20


 10


  0
                                                               100
                                                               90
                                                               80
                                                               70
                                                     < 60

                                                     O

                                                     LU 50
                                                     DC
                                                    O  40
                                                               30


                                                               20



                                                               10
                                                               0
40  42   44   46  48   5O   52   54   56   58   60

             CONTACT  TIME (minutes)
                                                                0    10  20   30   40  50   60   70   80  90   100

                                                                              INFLUENT OIL (mg/l)
CN
^t
n
                                                            NITIAL REMOVALS


                                                            REMOVALS AFTER START OF BED PULSING

-------
CO
-*
n
FIGURE 21


FULL-SCALE  PROCESS  PERFORMANCE
                               BP MARCUS HOOK
                                                                             FIGURE 22
EFFLUENT  COD  ATTAINABLE  FROM

ACTIVATED CARBON  SYSTEMS
   100




    90




    80




    70




    60




^  50




O  40
              £E
          PERIOD
                  30




                  20




                  10
                                    PERIOD
                                                              PHENOLS

                                                              COD
                                                                             a
                                                                             o
                                                                             o
                                                                   z
                                                                   Ul

                                                                   3
                                                                   u.
                                                                   LL.
                                                                   UJ
                                                                             <
                                                                             UJ
                                                                   a:
                                                                   UJ
                                                                   o
                                                                   z
                                                                   o
       I  VIRGIN CARBON, NO FCCU FOUL CONDENSATE

       2 CARBON BEDS NOT TURNED OVER,  FCCU

             CONDENSATE INCLUDED


       3 CARBON BEDS TURNED OVER, FCCU
             CONDENSATE INCLUDED

       4 EFFLUENT SEPTUMS  BENT,  NO FCCU

             CONDENSATE, ONE YEAR OPERATING
     450




     400




     350




     300




     250




     200




     150




     100




      50
                                                                                           Systems treating

                                                                                           biological treatment

                                                                                           system effluent
                                                                               A Only full-scale refinery
                                                                                system treating total
                                                                                waste
                                                                                                           915
                                                                                            10
                                                                                                20 30
                                                                                                      50
                                                                                                          70 80  90
                                                                                                                         99
                                                                          PERCENT OF PILOT-PLANT  OR FULL-SCALE SYSTEMS

                                                                          TREATING  REFINERY OR  RELATED WASTEWATERS

                                                                          WITH LONG TERM  MEAN  COD LESS THAN

                                                                          STATED VALUE

-------
  FIGURE 23
  INFLUENT AND  EFFLUENT  OIL AND
  GREASE  DISTRIBUTIONS  FOR A
  FULL-SCALE ACTIVATED
  CARBON  SYSTEM
   o>
   E
   UJ
   cr
   (C

   ^
   O
ISO -


I60


I40


I20
   UJ   IOO
   CO
 80


 60


 40


 20


  0
             Average Influent = 30.2 mg/l
             S D. = 36,3 mg/l
             Coef. Var  = 1.20
             No. of Data Points - 55 over 5 months


             Average Effluent - 7.4 mg/l
             S.D. = ll.l mg/l
             Coef. Var  = I. 5
             No. of Data Points = 50 over 5 months
                                                       FIGURE 24
                                                       CARBON ADSORPTION CAPACITY
                                                       FOR  VARIOUS  PLANTS
u.y
— 0.8
c
o
_a
o 0.7
X)
^ 0.6
T3
-
OQ.2
Q_
<
0 O.I
°c

A A


/
rl-
A
— Blue
S ^
Ay-£
\-Tahoe, Ci
C
/-
/0


-Owosso, Mich.
Plains (DC.)
A / '
fc-Clevt
'osemount, Minn.
I/if.
) ^^
^




'-Colorado Sprin
land Westerly
^'
^





gs, Colo.

/-^1
0







^-^
[
E





o
^-^"
3
3

) 100 200 300 400 500
                                                                     INFLUENT  COD TO CARBON COLUMN (mg/l)
          PERCENT  OF THE VALUES LESS THAN
          STATED VALUE
                                                       A Municipal Wastewater
                                                       O Refinery Wastewater
                                                       D Petrochemicals Wastewater
CO

-------
        CAUTIONS AND LIMITATIONS ON THE APPLICATION OF ACTIVATED
          CARBON ADSORPTION TO ORGANIC CHEMICAL WASTEWATERS

                                    C. T. Lawson
                                         and
                                    J. C. Hovious
              Union Carbide Corporation,  South Charleston, West Virginia

OBSERVATIONS  ON  ACTUAL WASTEWATERS

     Normal practice  in assessing the feasibility  of activated carbon adsorption is to first
conduct batch adsorption isotherms with powdered activated  carbon.  These tests provide
a relative indication of the amount of organic removal achievable by adsorption and the
ultimate adsorptive capacity of the carbon.  Unfortunately,  these data are useful only in
a relative sense - for  comparing the relative merits of two different carbons, or for
comparing the relative amenability of different wastewaters to adsorption.   Isotherm data
are not suitable for designing continuous granular caibon adsorption columns, since the
dynamic effects and interactions in a continuous  bed differ too greatly from the batch
equilibrium situation in an isotherm.

     Batch adsorption  isotherms have been performed on many organic chemical bio-
treated effluents and raw wastewaters  of widely varying  composition and overall organic
concentration (COD or TOC basis).  Figure  1 shows curves of percent organic removed vs.
carbon dosage.  It is obvious the amenability of  these wastes to adsorption  is rather wide -
ultimate percentage organic removals vary from  45% to    90%.  Even for those cases where
    90% organic removal can be achieved by massive carbon dosing in batch tests,  the
organic removals  achievable in continuous adsorbers are lower, often significantly so.

     Comparing batch and continuous data for Plant A bio-treated effluent:

     Ultimate % removal of TOC in isotherm test                           90%
     %TOC removal in continuous bed  adsorber prior to breakthrough
       (1 bed vol/hr)                                                      52%
     %TOC removal before breakthrough (0.5 bed vol/hr)                   64%
Not only was the organic removal predicted by isotherms not achieved, even in early
stages of the  column run with fresh carbon, but doubling  the  bed  depth (halving the
throughput rate) only  raised the amount of TOC removal from 52  to 64%.

     When batch adsorption data are plotted as log (wt.  organic adsorbed/wt. carbon) vs.
Jog (organic concentration remaining in solution),  the Freundlich isotherm  results - X/M
vs. C  .  When the logarithmic plot is extrapolated to C  = C  (the initial  solution  concen-
tration), the resulting (X/M)_  is called the "adsorptive capacity" of the carbon.  These
                           N_-
isotherm capacities are usually°greater, often very much  greater, than the  ultimate
adsorptive capacity exhibited by a granular adsorption bed at exhaustion.  Typical  examples
from prior publications (1):
                                      345

-------
 346
                                         AdsorpHve Capacity, gm COD/gm carbon
   Wastewater                             Isotherm              Column at Exhaustion
Plant A bio-effluent                           1.3                        0.26
Plant D bio-effluent                           0.65                       0.24
Plant G raw wastewater                        0.075                      0.05
Note that carbon had a very low adsorptive capacity for the constituents in Plant G
wastewater.

     It  was further noted that increasing the carbon bed depth yielded not only the afore-
mentioned  limited increase in percentage  organic removal,  but also a diminished adsorptive
capacity.  Operating two beds in series until each  bed  was  exhausted at an overall through-
put of 0.5  bed vol/hr (1 BV/hr/column) showed:
                                                gm COD  adsorbed/gm carbon
                                            CoTTlCol. 2
     Plant A                                  0.26                       0.17
     Plant D                                  0.23                       0.13
One may postulate that selective adsorption of the  "more adsorbable" organics occurs in the
lead bed.  Not only that, when the first bed is exhausted, the second bed has  little
remaining capacity - implying that the  "less readily adsorbed" species, which  are initially
removed in the second bed,  block its  further usefulness  in removing "readily adsorbed"
species (available when the  first bed breaks through).  This phenomenon is shown by the
breakthrough curves  of Figure 2.   Just how "easily  adsorbed organics" may be differentiated
from "less readily adsorbed" species is a difficult question.  The rate of adsorption at the
carbon surface, the rate of diffusion within  the carbon pores,  and the ultimate capacity  of
the carbon for the adsorbates all play a part in affecting "adsorbability."  In isotherm tests
these factors are  reflected in the measured adsorptive capacity and limiting percentage
organic removal.  In  continuous columns they  are reflected in the organic removal achieved
and the rapidity with which  organic breakthrough and exhaustion occur.  The absence,  or at
least minimization, of diffusion  limitations in  powdered carbon isotherms  can give a markedly
different picture  of "adsorbability," in  terms of both extent and capacity, compared to  the
complex inter and intra particle diffusion phenomena which occur in a granular carbon bed.

PURE COMPOUND ADSORPTION STUDIES

     Because of these observations of low percentage removals of TOC and COD,  limited
benefits of increasing carbon bed depth, and rather limited adsorptive capacities (in
continuous columns), an extensive investigation was conducted (2) to quantify some aspects
of "adsorbability" of specific organics.  The effects of functionality, molecular weight and
structure, aqueous phase pH, solubility, polarity,  carbon surface  structure,  and adsorbate
interactions were investigated.  Column vs. isotherm comparisons  were also made  for several
pure chemicals and  simple mixtures.

     In the first phase, 93 organic chemicals from 11 functional group families were
subjected to single dosage adsorption  tests.

     Functional group families:
         Alcohols

-------
                                                                                 347
         Aldehydes
         Amines and alkanolamines
         Aromatics
         Pyridines and morpholines
         Esters
         Glycols and glycol ethers
         Ketones
         Organic acids
         Ethers and oxides
         Halogenated aliphatics
     Test conditions:
         Initial compound concentration = 1000 mg/l (or solubility, if   1000 mg/l)
         Carbon:  5 gm/l of powdered Westvaco Nuchar WV-G (surface  area,= 1100 m /gm)
         Contact time = 2 hours

     Typical graphs of adsorbate loading vs. molecular weight are shown  in Figures 3, 4,
and 5 for the alcohols,  esters, and organic acids tested. The unfavorable effect of branching
and the favorable effect of unsaturation can also be seen.  Figure 6 is a "map" of percent
of compound adsorbed (by 5 gm/l carbon) vs. molecular weight showing the regions which
encompass all  the experimental data.

     Region  A     Aromatics (including ring-substituted compounds)
     Region  B      General trend for aliphatic mono-functional oxygen and nitrogen-
                  containing compounds
     Line C        Saturated organic acid line, shown for comparison
     Region  D     Glycols, glycol ethers
Note that the aromatics,  even those containing polar groups (-OH, -NO~, -Cl) are quite
amenable to adsorption, while the poly-functional oxygenated  compounds (glycols, glycol
ethers) are especially difficult to adsorb.

     Qualitatively, it was concluded that adsorbability, as reflected in constant dosage
tests, is favored by increasing molecular weight, aromaticity, and degree of unsaturation.
Increasing polarity, solubility, branching, and degree of dissociation (for amines and
organic acids) tended to severely limit the extent of adsorption.

     The relative ease of adsorption for simple oxygenated organic compounds may  be
summarized  as follows:
         Number of carbon atoms  4
            acids    aldehydes    esters   ketones   alcohols    glycols
         Number of carbon atoms  4 (straight-chain compounds)
            acids    aldehydes    alcohols    esters    ketones    glycols
The relatively low adsorption, in general, of compounds containing   4 carbon atoms in
these tests is particularly  noteworthy.  The carbon  dosage of 5 gm/l was a rather massive
dosage - in  a continuous column adsorber it would  be equivalent to treating only about
24 gallons of wastewater per pound of carbon.  This poor organic removal at high carbon
dosage implies that carbon adsorption may have limited application to many organic
chemical manufacturing wastewaters, from a cost-effectiveness viewpoint. Low molecular

-------
 348


weight oxygenated organic compounds represent a very large fraction of the production
volume in the industry.

     In the next phase of the study, Freundlich adsorption isotherms were obtained on 1000
mg/l solutions of five different organic compounds (all containing four carbon atoms) using
four different carbons at different aqueous-phase pH levels.  Figure 7 shows that pH  had a
marked effect on the isotherm  adsorptive capacity,  (X/M)- , for butyraldehyde and ethyl

acetate.   Data for Carbon A,  Witco 517, are  shown; similar trends were observed for other
carbons.   Gas chromatographic analyses  indicated that, at acidic or basic pH levels, ethyl
acetate was hydrolyzed  into its less adsorbable (lower molecular weight) components. At
elevated pH, butyraldehyde underwent an aldol condensation into a  higher molecular weight
(more adsorbable) component.  More detailed  discussion of differences between the carbons
tested and the effects of carbon surface  properties are contained in Ref. (2).

     A potential problem in successfully applying activated carbon to multi-component
wastewaters is indicated.  It is not possible to select a pH level  that assures maximal
adsorption of all wastewater constituents. An acidic pH  is required to facilitate adsorption
of organic acids (to avoid ionization),  near neutral pH favors adsorption of esters, while a
highly alkaline pH is necessary to maximize adsorption of aldehydes.

MULTI-COMPONENT ADSORPTION STUDIES

     Interactions in multi-component systems were examined in the third phase of the study
by adsorption isotherm tests on binary and four-component mixtures.  The results are shown
in Table  1.

     In a test (at pH  = 7) of a butanol/ethyl acetate binary mixture (500 mg/l of each
compound), the total adsorptive  capacity was  observed to be 0.237 gm cpds/gm carbon,
while the sum of the  two adsorptive capacities (at 500 mg/l initial concentration) from
pure-component isotherms was 0.27 gm/gm.   By gas chromatographic analysis of the residual
solutions, the butanol adsorbate loading was  1.05 times the loading  in a pure component test
while the ethyl acetate  loading was 0.76 times the  pure component loading.  Even though
more ethyl acetate was adsorbed from the mixture, on a weight basis, than butanol (as
expected from Figure 7), the interaction had  a greater deleterious effect on ethyl acetate,  in
terms of using pure component data to predict  mixture adsorbate loadings.

     By contrast, in the butyraldehyde/MEK binary test,  the total adsorptive capacity from
the mixture  (0.275 gm cpds/gm)  was greater than the sum of the pure component values
(0.270 gm/gm), more butyraldehyde (the "more  adsorbable component") was adsorbed on a
weight basis, and the ratio of  mixture loading to pure component loading was greater for
butyraldehyde, 1.28 vs.  0.60.

     In both four-component mixture isotherms (Table  1), the total adsorptive capacity (0.29
and 0.225 gm/gm for Carbons  A and C, respectively) was significantly less than the  sum
of the pure component capacities (0.575 and  0.312 gm/gm for Carbons A and C,  respectively).
Also, in the Carbon  C test,  less butyric acid  was adsorbed than butyraldehyde or ethyl
acetate,  despite the  previous pure component  conclusion that acids are more adsorbable than

-------
                                                                                349

other organic compounds.

    Thus, while pure component adsorption data are useful in determining which waste
streams are potential candidates for activated carbon treatment, the data cannot be used
to quantitatively predict adsorption from multi-component systems.

CONTINUOUS vs. BATCH ADSORPTION STUDIES

    In the fourth and final phase of the study, continuous column, granular carbon
adsorbers were operated to exhaustion for comparison of ultimate capacities with isotherm
capacities.  Each column  was operated at a throughput rate of 1 bed volume/hour.  The
feed concentration was  1000 mg/l  in the pure component  tests, 2000 mg/l  (500 mg/l each
compound) in the four component test (feed TOC = 1200 mg/l).  The four component
mixture was butyraldehyde, ethyl  acetate, butanol, and butyric acid.  In  all cases, the
ultimate capacity (at exhaustion) was less than that predicted by isotherms (83 to 89%).
                                 Adsorptive Capacity, gm/gm carbon
                                                       Column             Column
Organic                      Isotherm               at Exhaustion       at Breakthrough
MEK                            0.160                0.132                OVK)3
Butyraldehyde                   0.220                0.192                0.141
4-Component Mixture            0.170 (TOC)         0.151 (TOC)          0.124 (TOC)
Note  that the usable capacity in a single-column system - the capacity achieved at
"breakthrough," where significant organic concentrations appear in the treated effluent -
is significantly less than indicated by the isotherms (64 to 72%).  Figure 8 shows the experi-
mental breakthrough curve for the four-component mixture.

    A second important criterion when applying activated carbon in continuous beds is  the
amount of wastewater that can be treated before breakthrough  (often called the "carbon
dosage").  Breakthrough in full-scale practice (where the carbon is removed from service)
would be set at the maximum organic concentration allowed by effluent standards.  For
purposes of discussion, breakthrough was set at the 95% organic removal level in these
lab tests.  Observed dosages were:
    Test                                 Breakthrough "dosage" gal  WW treated/lb carbon
MEK                                                      12.7
Butyraldehyde                                             19.4
4-Component Mixture                                      12.4
Treating large volumes of wastewater at these inlet concentrations will,  thus, lead to
rather short service life (rapid breakthrough), even with the fairly ideal  adsorption wave-
front shape in  Figure 8.

    Hydraulic effects (channeling and axial dispersion) are a partial explanation for the
failure of column adsorbers to achieve adsorbate loadings predicted by isotherms.  The
inherent differences between the static equilibrium attained in batch isotherms and the
dynamic situation in a  continuous column are also limiting factors, particularly when
treating complex multi-component (   4) wastewaters.

    A recent paper (3) by Keinath has shown that competitive adsorption, even in binary

-------
 350


solutions, can lead to chromatographic displacement effects in column adsorbers treating
wastewaters of varying composition.  In this situation, the weakly-adsorbed solute is retained
by the carbon for a time and then is eluted as a rather concentrated "peak" upon prolonged
contact with the feed wastewater.  Keinath further showed that fluidized bed adsorbers are
much less susceptible to this effect  than more  conventional packed beds - a finding that could
have significant impact on adsorber design.  Chromatographic displacement and related
interaction effects are almost certainly key contributors to the  diffused,  spread-out adsorption
wavefronts and  fluctuating organic  removals observed when treating multi-component organic
chemical wastewaters of high component and concentration variability.  The question of
dynamic interactions and effects in multi-sorbate systems  is an  area where further research and
development is sorely needed, not only to  shed further light on which wastes can be efficiently
treated by activated carbon adsorption, but also to ascertain how best to design and operate
continuous adsorption systems to minimize the undesirable effects of interactions.

CONCLUSIONS

 1.   In pure component studies, specific  organic chemicals have been shown to differ widely
     in their amenability to adsorption,  depending on molecular weight, structure, polarity
     and solubility.  Low molecular weight oxygenated organics are particularly difficult  to
     adsorb efficiently.
2.   The relative ease of adsorption of different functional group compounds can vary strongly
     with pH, depending on the chemical nature of the adsorbates.  An optimum pH cannot
     be predicted for a multi-component wastewater of unknown or varying  composition.
3.   While pure component data could be used to predict  binary adsorption  capacity in
     isotherms fairly closely, a four-component mixture isotherm showed  only about 60% of
     the adsorptive capacity predicted.   Mutual solubility effects competition for adsorption
     sites, and  inability to maintain a pH level optimum for all components  contributed to
     this inability to extend pure  component data to more  complex mixtures.
4.   While isotherm capacities were somewhat extrapolatible to continuous  column behavior
     in pure component adsorption and simple  mixture studies, such extrapolations have not
     proved to be possible with real wastewaters (complex mixtures),  particularly with bio-
     treated effluents.
5.   More importantly,  the key parameters of interest in real wastewater treatment situations  -
     percentage organic removal achievable and water volume  treated per pound of carbon
     before breakthrough - cannot be  predicted  from isotherm tests.
6.   The physical differences between an equilibrium  adsorption situation in a powdered
     carbon isotherm and the dynamic, multi-component interactions in a continuous granular
     carbon bed are too great to permit prediction of column performance from isotherms or
     pure component data. The dynamic interplay of adsorption rates, pore diffusion rates,
     hydraulic effects, and pure component chemical  properties in continuous columns is an
     area where further  research efforts must definitely be applied.

REFERENCES

(1)   Lawson, C. T., and  J. A.  Fisher,  "Limitations of Activated Carbon Adsorption for
     Upgrading Petrochemical  Effluents," Water - 1973, AlChE Symposium  Series No.  136
     (1974)

-------
                                                                                  351

(2)  Giusti, D. M., R. A. Conway, and C. T. Lawson, "Activated Carbon Adsorption
    of Petrochemicals/1 JWPCF, 46, No.  5,  May (1974)
(3)  Keinath, T. M., "Design and Operation of Activated Carbon Adsorbers Used for
    Industrial Wastewater Decontamination,"  paper presented at 68th Annual Meeting -
    AlChE, Los Angeles, Nov.  (1974)

DISCUSSION

Frank Manning:  Would you  care to comment on how you would personally combine
activated carbon with biological treatment together, which one is put first?

C. T. Lawson:  Our normal thinking is that we would put carbon before biological treatment
on specific streams where it  has  an application.  In other words, on a stream say that's
noxious for some reason but not biodegradable  or perhaps biologically inhibitory, then we
consider that a very good application for consideration of carbon.  As a post-biological
polishing step,  it depends on how close we are to meeting a permit.  I guess, if a little
more treatment would get us to the permit then I would say a tertiary carbon system might
be justified.  I don't think,  and we are convinced of this fact,  that you can simply add on
a carbon system and make up for deficiencies in your biological system.   If you  are far
away  from meeting a permit, the first think to  do is make the  bio-system work.  Then if you
need carbon to close the gap and it is effective in doing so,  put it on; but I don't think that
carbon can ever be considered as a prescription for a poorly operated treatment  system.

Sterling Burks:  Have you performed any studies on absorption of trace heavy metals?  I
know  this question was asked of  Davis Ford.

C. T. Lawson:  No, we haven't.

Milton Beychok:  As you know,  I have congratulated you once before because I  think it is
long overdue that someone brought some science to this  field.  If we look at the data you
have given us this morning,  the  isotherms in thermodynamic terms,  the isotherms are really
equilibrium.  Your  column data  telling us you  can't approach that equilibrium the way you
think  you should and that involves kinetics, and if we look again and make an analogy
between that and other refining  processes and what has been done in improving  kinetics and
the approach to equilibrium  through catalysis.   Do you intend or do you know of any research
going on this pretreatment of carbon, trying to do something to improve the kinetics as well
as perhaps making kinetic models and seeing what you can learn?

C. T. Lawson:  Well the question of pretreating carbon, I really don't know of  any work
that is going on.  It seems to be the kind of thing that the carbon vendors would be
interested in and they are pretty competitive and  closed-mouth fellows; they don't tell you
much about what they are doing. There is some potential  here, I would say,  for tailoring
carbons through pretreatment or  through activation processes;  but it is not being really
studied much to my knowledge.  On the question of developing a diffusion model or a
kinetic model,  there is quite a bit of academic interest  I think in doing this.  The only
papers that I have seen are really for binary systems and that is not of too much  interest in
a practical application sense.  I know Dr.  Keinath at Clemson University and Dr. Weber
at the University of Michigan have done some  work  in this modelling area.

-------
352
  Anonymous:  It just seems to me from what we all know that the regenerated carbon behaves
  differently than the fresh carbon and you showed us today percentage wise some remarkable
  differences between two different carbons, that  it would be a fruitful area for research work
  whether it be on the part of Union Carbide or the carbon manufacturer.

  C. T.  Lawson:   Yes, I  don't really want to give a commercial for anybody, but  I will say
  that Witco's  carbon in  the studies we  looked  at performed quite well, very good carbon.  I
  think there has been a  lot of reluctance over the years to use anything but bituminous coal
  carbons because its "sturdier,  more rugged, easier to regenerate, not as friable as the other
  carbons."  Looking at waste treatment with some of the speciality carbons, coconut-shell
  carbons, has shown some pretty good results,  - a little bit better than bituminous coal carbons.
  But when  you talk about 75-85$ per pound for carbon, it is really out of the question for
  large volume waste water treatment.

  Nick Sylvester: You mentioned at the end of your talk that you have done some more research
  that you have not  reported on  and you also mentioned this apparent chromatographic type
  phenomena in the  columns.  Have you done more work on that and are you going to report
  that?

  C. T.  Lawson:   That is what we have  tried to do, we have tried to take a  column and take
  samples down through the column at different times and see just where specific compounds end
  up in the  column on a dynamic basis,  to see  if there  is some way that you  can predict when
  something is  going to be spit out as a  big eluted peak, and I  hope that data will be in shape
  for the Cincinnati paper.

  Dave Skamanca:  Davis Ford mentioned that  if you are going to put in an  activated carbon
  system, you are really better off with biological system in front of it to decrease the loading
  and apparently improve the performance of the  carbon.  But as you decrease the loading,
  the pounds in the  concentration to the activated  carbon system, you  also decrease the isotherm
  data,  you don't get as good a  loading on the activated carbon.  Are you getting ahead of
  the game  in most of these systems by trying to remove more of these pounds up stream with
  a  carbon, are you gaining on  the problem, are  you just staying even so to speak?

  C. T.  Lawson:   My own opinion is that the kind of capacities that you see in isotherm tests
  are really an interesting qualitative observation, but they have no practical significance.
  Anything  you can  take out before the carbon column by reasonable treatment means, I  think
  is worthwhile.   From what we  have seen,  running continuous column systems,  this 25% or
  so by weight absorptive capacity (on a COD  basis) to petrochemicals seems to be a pretty
  real number and relatively independent of the feed concentration in  continuous systems for
  the wastes we have examined  thus  far.  The fact  remains that isotherm capacities obviously
  do vary strongly with concentration, you can go  up and down. We really see no way to
  relate  that to a continuous system.  We use our isotherm data just for relative comparison
  of one waste water to another  and one carbon to  another;  we don't use it  in a quantitative
  sense at all.

-------
                                                                                 353
BIOGRAPHY

    Cyron T. Lawson is a Project Scientist in the
Water Quality Development Group of Union
Carbide's Chemicals and Plastics Division,
Research and  Development Department.  He holds
the  B.Ch.E. and M.S.Ch.E. degrees from Georgia
Institute of Technology.  He has taught graduate
courses in water and wastewater treatment as an
Adjunct Instructor at the West Virginia College of
Graduate Studies.  Cyron is a Registered
Professional Engineer (Chemical) in West Virginia
and is a member of the  AlChE and its Environmental
Division.
    Joseph C. Hovious  is a  Group Leader/
Technology Manager in  the Research and
Development Department of  Union Carbide's
Chemicals and Plastics Division.  Mr. Hovious
is an M.S. Environmental Engineer from the
University of Illinois.  He teaches graduate
courses at the West Virginia University
College of Graduate Studies.  He  is a member
of the WPCF, AWWA, and a Registered
Professional  Engineer  in West Virginia.

-------
                                                                                                                     GO
     TABLE 1 "COMPARISON OF MULTI-COMPONENT ISOTHERM DATA WITH SINGLE COMPONENT DATA'
Mixture 1  - Carbon A
  Butanol
  Ethyl Acetate

Mixture 2 - Carbon A
  Methyl Ethyl Ketone
  Butyraldehyde

Mixture 3 - Carbon A
  Butyraldehyde
  Ethyl Acetate
  Butanol
  Butyric Acid

Mixture 3 - Carbon C
  Butyraldehyde
  Ethyl Acetate
  Butanol
  Butyric Acid
Equilibrium Carbon Loading
    in Mixture Test
   g cpd/g carbon (a)

        0.116
        0.121
        0.237

        0.063
        0.212
        0.275

        0.072
        0.075
        0.031
        0.112
        0.290

        0.072
        0.066
        0.023
        0.064
        0.225
                                                         Equilibrium Carbon Loading Fraction of Single Component
                                                          in Single Component Tests,      Equilibrium Loading
                                                              g cpd/g carbon (b)           from Mixture (c)
0.110
0.160
0.270
0.105
0.165
0.270
  165
  160
  110
  140
0.575

0.080
0.080
0.068
0.084
0.312
1.05
0.76
0.60
1.28
0.44
0.47
0.28
0.80
0.90
0.83
0.34
0.76
(a)  Compounds each initially present at 500 mg/l.  Mixtures  1 and 3 dosed at 1  gm carbon/liter and Mixture 2 at
    2 gm carbon/ liter.   Equilibrium loadings (X/M)_  are calculated from total TOC loadings and chromatographically

    determined fractional compositions of removed material.
(b)  Equilibrium loadings  measured at C  of 500 mg/l.
(c)  Calculated by dividing the equilibrium loading of the individual  components in the mixture by the single-component
    equilibrium loading.

-------
                                         FIGURE 1

                   ACTIVATED  CARBON  TREATMENT OF BIOLOGICAL EFFLUENTS

                              BATCH ADSORPTION ISOTHERM TESTS
                                                                                          355
100 ._
               CQ   -   initial  COD (or TOO  concentration before carbon treatment
                          Plant E,  C0 = 288 mg COD/1
                                              Plant A,  C0 = 360 mg TOC/1*
                                                             ~	
                                           D Plant F,  CQ = 2300 mg TOC/1
Plant C
                                                                               C0 = 1363 mg COD/1
                            Plant A,  C0 = 1030 mg COD/1*
                                                   *Plant A tests run at different times,
                                                    several months apart
                            l    i    i
                                                                       i    i    i    i
                                          20
                                  Carbon Dosage, gm/1
                                                              30
                                        FIGURE 2

                  BREAKTHROUGH CURVE FOR PLANT A BIO-TREATED WASTEWATER
                                                                                  40
             hfl
             e
             Q
             o
             u
                1000
                 800
                 600
                 400
                 200
                        Feed = C
                                                      0.5 BV/hr
                                                      0.88 liters/BV
                                                       (2 col.)
                                                      	I	
                              50       100       150       200

                             Wastewater Throughput, bed volumes  (BV)

-------
FIGURE 3

EFFECT OF  MOLECULAR WEIGHT ON AMENABILITY OF  ADSORPTION OF ALCOHOLS
                                                                      FIGURE  4

                                                                      EFFECT  OF MOLECULAR WEIGHT  ON AMENABILITY  OF ADSORPTION  OF ESTERS
        0.200  .
   »
   z~
  -o
  m
  CO
         0.150  _
         0.100
         0.050
 C0* 1000mg/ I at Alcohol
 5gm/l Carbon Dosage
*C0- 700 mg/l
                                               Butanol
                                                          n-Hexanol



                                                        O 2  Ethyl -Butanol


                                                    n-Amyl Alcohol
                                             OIsobutonol
                                             O t- Butanol
                                         Propanol

                                    Q Olsopropanol
                                      "yl Alcohol
                                   Ethanol
                              Methonol
             40               60


                 Molecular  Weight
                                                               2-Ethyl Hexanol
                                                               120
                                                                                              0.200
                                                                             0.150
                                                                        9

                                                                        z"
                                                                        •V
                                                                        •x.
                                                                                              0.100
                                                                                              0.050
C,,= 1000 mg cpdY I

5 gm carbon/ I
  j	L
                                                                                                                                              Butyl
                                                                                                                                             Acrylat*
                                                                                                                    Butyl
                                                                                                                    Acetate
                       Ethyl Acrylcte j/

                      Propyl Acetate
                                                                                                                                               Isopropyl Acetate
                                                                                                                                         n-Amyl
                                                                                                                                           Acetate
                                                                                                                                         Ethyl Acetate
                                                                                                                                  1 Methyl Acetate
                                                                                                         4-
          40              80

            Molecular  Weight
                                                                                                                                                   120

-------
                          FIGURE 5                                                                      357
                           EFFECT OF MOLECULAR  WEIGHT ON AMENABILITY OF ADSORPTION OF  ORGANIC  ACIDS
                     0.200
                     0.130 _
               1
o

t»
 »
a

X
                     0.100
                    O.050
                                C0»IOOO mg cpd/ I
                                Sgm carbon/1
                                                                     O Btnzoic
                                                 Acrylic
                                                   O
                                                    'Propionlc
                                                r Acetic
                                          Formic
                                       40             60


                                          MolMular WilgM
                                                    120
            FIGURE 6

            PERCENT OF COMPOUND ADSORBED VS. MOLECULAR WEIGHT FUNCTIONALITY  EFFECTS
 100
  80
Seo
m
                                            ,Aromatic
s
in
o
  20
                                                                                    Q)  Glycols,
                                                                                        Gly-Ethers
                                                       i     '     '	I	1	1	1	1	1	1	1
             20        40        60        80        100       120

                                                   MOL.  WT.
                                                           140       160       180       200

-------
358
FIGURE  7
                    pH EFFECTS ON ADSORPTION REMOVALS OF  SELECTED ORGANIC COMPOUNDS

                    WITH CARBON A
      0.5
      0.4
      0.3
   I
   o
   u
   Q.

   U





   I
   x  0.2
      O.I
                                          Butyroldehyde •
                         Ethyl Acetate
                                                   wuiuin/i x.
                                               Methyl  Ethyl Ketone'
                                         -L
                                         6



                                          PH
                                           10
12
           FIGURE 8
u
o
    1200
    1000
    800
     600
    400
    200 L
           BREAKTHROUGH  CURVE FOR ACTIVATED CARBON ADSORPTION  OF A

           FOUR-COMPONENT MIXTURE IN  A CONTINUOUS  SYSTEM



                                   Feed
                  TOC adsorbed' 0.151 gmTOC/gm carbon
                                       40             60



                               Throughput , bed volumes (at I BV/Hj/*
                                                  80

-------
           CASE HISTORY:  USE OF POWDERED ACTIVATED  CARBON
                       IN AN ACTIVATED SLUDGE SYSTEM

                                  Joyce A. Rizzo
                    Sun Oil Company, Marcus Hook,  Pennsylvania

INTRODUCTION

    In an effort to expand the performance of existing biological treatment facilities, full
scale  trials utilizing powdered activated  carbon were conducted at Sun Oil Company's
Corpus Christ!, Texas Refinery.  The main objective of the trials was to reduce the effluent
suspended solids loading for compliance with the 1977 NPDES and State permit limitations.
Three separate trials were initiated  over a one-year period comprising five months of
operation with powdered activated carbon addition to the aeration system of the 2.16
MGD biological treatment plant. Although the powdered carbon addition could not
guarantee compliance with the 1977 suspended solids' criteria,  improvement in the existing
system's performance was significant for COD and BOD removal as well as suspended solids.
The treatment costs ranged from 1.7 to 4.3^/10  gallons depending on the influent flow
and quality.  Data have been  compiled showing reductions in the refinery's final  effluent
loadings of up to 56% for suspended solids, 36% for COD and 76% for BOD.

FACILITIES

    A simplified flow diagram of the treatment scheme  at the Corpus Christi Refinery  is
shown in Figure  1.  The system's operating conditions are outlined on Table 1 .  The daily
average charge to the plant is only  half the  design capacity; although the peak design
rate is often operated during periods of rain  and high ballast loading.  The refinery waste-
water flow (averaging about 600 GPM) is equalized with ballast and contaminated storm
water in a 10 MM gallon pond.  The combined flow from the pond (averaging about 750
GPM) is pumped through a dissolved air flotation unit for removal of any excess oil and
suspended solids. The effluent from the DAF is split between two 760,000 gallon
rectangular aeration basins. Air is  introduced into each of the  basins by two 75 hp
floating mechanical aerators.  When operating both basins in parallel at average  flow
conditions, the retention time  is about 22 hours; at peak flow, this is reduced to  11 hours.
The mixed liquor volatile  suspended solids is maintained at about 2200 ppm. With both
systems operating,  the food to mass  ratio  averages only about 0.13 Ib COD/lb  MLVSS or
0.07  Ib BOD/lb MLVSS in each basin; this,  of course,  can  be doubled by operating only
one aeration system.

    Each aeration  system  is coupled with a 55' diameter x 12' deep clarifier.  The over-
flow from the clarifiers combines for discharge. The settled sludge is collected from each
clarifier,  combined and returned to the aeration basins. The return rate usually averages
about 50% of the charge rate.
                                      359

-------
360
  OBJECTIVE

      Basically, the performance of the biological treatment system is satisfactory and the
  effluent readily complies with the refinery's  NPDES and State permits.  With the exception
  of suspended solids,  the limits established for compliance in 1977 are also currently being
  met.  The consistently effective removal of suspended solids in the final clarifier seems to
  be the only area of concern.  Even with substantial  organic removal  in the biological
  system, there appear to be two major factors causing the extra suspended solids carryover
  from the final clarifiers:
          Variability of influent organic and hydraulic loadings.  The biological  population
          is thus unstable with erratic growth and variable efficiency.
          Excessive aerator foaming in the aeration basins.  This foam,  laden with biological
          solids, tends to carryover into the clarifier and out  to the effluent.  The problem
          becomes most severe at high flow periods.

       Interim limitations had been set by the State and NPDES permits  allowing time to
  investigate  in-plant improvements which could possibly permit compliance with the 1977
  TSS criteria without construction of additional facilities.  Full scale trials appeared
  necessary due to the nature of the problems noted — bench scale studies tended to show
  excellent settling characteristics which do not reflect the hydraulics of the operating system.

       Initial  testing was performed with the  addition of polymers to the aeration  basin effluent.
  The polymers did improve the solids settling somewhat but their benefit was overshadowed  by
  the problems still being caused by aeration foaming and shock loads.  Conversely, defoamers
  were effective in controlling the foam, nothing more.

       In the refinery, every precaution is taken to reduce the possibility of shock loads and
  spills.  However, no matter how efficient, there is no way  to completely eliminate the
  variability:  turnarounds,  emergency shutdowns and the like.  This, coupled with the influx
  of highly variable loads of ballast, dictated  that the treatment plant operation  would have
  to compensate for unstable influent conditions to the plant.

  APPLICATION OF  POWDERED ACTIVATED  CARBON

       The addition of powdered activated carbon was initiated in an effort to resolve all the
  operating problems causing the suspended solids carryover — the objective being compliance
  with the 1977 suspended solids'  limitations.  Although this objective was not achieved, some
  interesting data resulted.   A summary of the  three separate  carbon trials conducted is shown
  on Table 2.  Unfortunately, it  was not possible to operate a "blank" system  simultaneously
  with the carbon system.  Therefore,  the  data obtained 30 days prior to the start of each
  trial has been utilized as the base period for that trial.  As can be noted on Tables 3, 4 and
  5, for the most part, the influent loadings of each  base period and respective trial are
  comparable; this is especially true of Trial 3.  During each trial, operation was maintined
  as closely as possible to the conditions observed during the  base period.  No special problems
  were encountered that would lead one to believe that any of the trials  and base periods'
  operation were different,  especially on the average.  Although,  the comparison is not ideal,
  it is relevant, especially in respect to the "real world" effect on the plant's effluent.  The

-------
                                                                                 361


percent reductions shown between the base period and each trial are meant as a guide to
point up the improvement in effluent  quality; the actual values, of course, cannot be
taken as absolute.  From an operation's standpoint, there appears to be no question that
the addition of the powdered activated  carbon  to the aeration system realizes a significant
improvement in the effluent quality and reduces its variability.

     The first trial was considerably longer than the last two — mainly, because the
carbon dose was gradually built up in the system over a four-week period;  whereas in the
other trials, the carbon was batch loaded in a couple of days up to the operating level.
The gradual addition of the  carbon allowed the daily observation of its effect in order to
determine the appropriate loading to  be maintained.  In order to minimize the amount of
carbon needed to load and maintain the system, only one aeration basin was operated
during the  trials; the same was true of the respective base period.  The carbon was
manually batch added  daily to the dissolved air flotation effluent as it entered the
aeration basin.  During the  first trial  it was found that  the optimum operating carbon level
needed in the aeration system was 450 ppm with approximately 1000 ppm resulting in the
recycle sludge.  All three trials were conducted at these levels.   This requires about
6100 pounds of carbon to  charge one  aeration system up to operating levels.  An easy
indication  of the carbon level  in the  system was found to be the foam level in the  basin —
the appropriate amount of carbon would eliminate the foam completely; as the carbon
level dropped, the foam would build  up.  Of course, laboratory analysis utilizing a
standardized comparison procedure confirmed the operating level of carbon on a concen-
tration basis.

     While gradually adding the carbon daily,  it was noted that a distinct  improvement
in the  plant performance did not come about until the 450 ppm level was attained  —
however,  by letting the level drop off gradually,  the peak performance could be main-
tained down to 300 ppm.  It was also noted that more frequent additions of smaller
quantities appeared more  effective than larger  batch additions less frequently.

     Referring to Table 2, you  will note that the carbon addition rate was much higher
during the  first two trials; this  came about for a couple of  reasons:
     '   More frequent  batch additions of larger quantities  ( 1000 pounds at a time) were
        necessary to compensate for shock loadings.  Although the presence of the carbon
        in the system greatly reduced the  chances  of a shock that could inhibit the
        biological activity, it  could  not guarantee that a reduction in activity would not
        occur.   However, in general, it was found that if  caught in time with the  addition
        of a large batch of carbon, a lapse in system efficiency  could be effectively
        brought back up to normal almost  immediately without adverse effects on the
        effluent quality.
     '   At  the time of  the third and most recent trial, the biological system was more
        stable and it took less carbon to obtain the same results.  Better methods of
        equalization had been  put into  operation and ballast was being successfully
        treated at a rateable basis.

     The flow rate during  the third trial averaged 800 GPM versus 625 - 660 GPM for the
first two trials (this was due primarily to increased ballast water treatment).  The higher

-------
362
                                                                       3
 flow with the  reduced carbon addition rate accounts for the lower 1.7^10  gallon treatment
 cost compared to Trial Ts 4.3$.  It is important to note that these costs only reflect the
 carbon usage, the manpower utilized in adding the carbon is not included.  More efficient
 means of carbon addition could, of course,  be employed with minimal capital investment
 but was not deemed necessary for a trial basis.

      On Table 6, the treatment costs  obtained are summarized along with the reduction in
 effluent loading observed during each trial.

      From this data,  one is erroneously led to believe that the efficiency seemed to improve
 because of a lower carbon  dose (and subsequently lower cost).   As explained, the higher
 carbon dosages observed during Trials  1 and 2 are directly attributable to a higher  frequency
 of shock loadings to the system.  The only valid conclusions to be drawn from the costs
 vepsus efficiency data presented is that the carbon treatment costs range from 1 .7$ to 4.3$/
 10   gallons depending on the variability of the influent and the frequency of batch additions
 to solve overload problems.  Again, it is  important to remember that the percent reductions
 are determined from  the trial versus its pre-trial base case and  should only be thought of as
 a guide for comparison purposes.

      Total Suspended Solids Reduction. The  average effluent suspended solids data obtained
 are outlined on Table 3.  The reduction in the effluent suspended solids' loading was
 dramatic, between 49.3% and 55.7%. The  existing permit limit is easily met; however,
 the effluent, at the very low level  of 405 Ibs/day daily average for Trial 3, still would not
 comply with the 327  Ibs/day limit of the 1977 permit.  In addition,  although the daily
 maximum  values while using carbon were significantly lower and under better control,
 several  peaks  still would occur.  Referring to Table  7,  in the case of Trial 3, four  data
 points out of 20 were above the 561 Ib/day 1977 allowable limit, for 80% compliance;
 100% compliance is easily obtainable with the existing  permit.  Of course, the odds of
 being able to  comply with the 1977 limits are better, but still not acceptable.

      Referring to the  frequency distributions for Trial 3  plotted  on Figures 2 and 3,  the
 reduction in effluent suspended solids with carbon addition becomes very obvious.  The
 chance of complying with  the 561 Ibs/day 1977 maximum was increased from 40% during
 the base period to 82% with carbon; of course, these percentages are only relative not
 absolute.  Although the permit is written  on  a mass basis, the effluent concentration of
 suspended solids is a  basis  more easily compared between various facilities and  is there-
 fore, shown in Figure 3.  On each figure, the flat slope of the carbon trial  versus the
 base  period also indicates  the more consistent effluent obtainable with carbon treatment,
 the variability having been significantly reduced.

      Organic Removal Efficiency - as mentioned, the addition of powdered carbon also
 improved  the organic removal efficiency of the system.  A summary of the influent and
 effluent COD loadings of the system are shown on Table 4.  This table also points  up the
 uniformity of influents between the base period and  trial period for Trials 1 and 3.  As
 mentioned,  the operation was maintained as consistent as possible for each case.   During
 Trial 2 the influent COD loading was about 25% higher than during the base period —
 yet,  the effluent loading was 30%  lower showing a significant reduction on a relative
 base.  The removal efficiency during  Trials 1 and 3 was increased significantly up to 70%

-------
                                                                                363


from 62.9% and 54.8% respectively for each trial.  The ultimate  reduction in  COD
loading to the effluent being 35.7% for Trial 3 as compared to base period 3.

     The daily COD removal efficiencies for Trial 3 are plotted in Figure 4.  It can be
seen that not only was the removal  efficiency significantly increased, the variability
of the efficiency was reduced.  Again, resulting in an effluent of consistently  better
quality.

     Figure 5 is a frequency distribution plot of the effect of the carbon addition on the
effluent COD concentration for Trial 3.  Not only does the slope again show more
consistent results but, for example, the 95% point has been reduced from 280 ppm down
to 150 ppm, a 46% reduction  assuming a comparable base.

     The effect of powdered carbon on BOD is summarized in Table 5. It is obvious from
the influent loadings that the  data obtained during Trial 3 are most relevant since the
base period  and trial had  the most comparable influent conditions.  Again, as in the case
of COD, during Trial 2, the influent loading was 38% higher  than the base period; yet,
the effluent was 51% lower.  The reduction in BOD was most sharply noted during Trial
3 when  the effluent BOD  never exceeded 6 ppm — the reduction from the base period
was 76%.

     The frequency distribution comparing the base period and Trial 3 is shown  in Figure
6.  Again the results are significant.  The  curve  for the effluent BOD during the carbon
addition period is virtually flat, showing definite uniformity and consistency of effluent.
Figure 7 denotes  the distribution of the percent BOD removals obtained — again, the
curve is relatively  flat, the variability being virtually eliminated.

     Oil Removal.  No  data were compiled on oil and grease removal since the system is
not required to handle any large amounts of oil.  The effluent oil and grease usually
averages less than 5 ppm with an occasional  peak around 10 ppm; the influent is normally
20-25 ppm.   Any improvement would be difficult to observe at these  low  levels. On
occasion, however, the influent oil level can be higher, as did happen during  the  post
period of Trial  2 when there was an upset in  the Refinery.  The carbon still in the system
readily adsorbed the excess oil, keeping it in the sludge.  Of course, the sludge had to
be removed  from the system, thus spending the carbon.  However, a possible effluent oil
and grease violation was avoided at only the expense of a carbon charge.

SPECIAL NOTE

     Sludge  wasting had always been erratic  and  unpredictable in this system due to the
variable influent and unstable bug growth.  When as stable as possible, about 15,000
gallons per day of recycle sludge would be wasted from one operating aeration  system.
During the carbon trials the wasting was reduced to 10,000 gallons per day under similar
circumstances.  The basis for this reduction being that the sludge was considerably
thicker so less volume needed to be handled  to dispose of the same mass.

-------
364
SUMMARY

     As evidenced by the data presented, the addition of powdered activated carbon to the
aeration system of the refinery's existing wastewater treatment facilities did not reduce the
effluent suspended solids enough to comply with the  1977 permit limitations.  However,
significant improvements in the system's performance were observed as follows:
        Improved  organic and suspended solids removals and reduced effluent loadings
              	% REMOVAL	
     Trial 3    Without* Carbon    With Carbon   % Reduction in Effluent Loading
      COD        5478              7075                  3577
      BOd        93.5              98.5                  75.8
      TSS         -                   -                     49.3
            *Pretrial  base period
        More uniform effluent quality
        Clearer effluent
        Elimination of foam in the aeration system
        More consistent sludge wasting at 2/3 the volume
        Reduced chances of biological upsets -- the use of powdered  carbon does not
        eliminate biological upsets but it does appear to reduce the opportunity of their
        occurance; and when they do occur, it  appears to maintain the effluent quality
        under better control.  When  caught in time,  it was found that a reduction in
        biological activity can effectively be brought back up to normal by massive batch
        addition of carbon to the existing operating level.

     These improvements were obtained by building up the aeration system to a 450 ppm
operating level of carbon with about 1000 ppm  in the recycle sludge.   The system was
maintained at this level by the. batch addition of about 100 Ibs/day of carbon (or 10 ppm)
at an average cost of 1 .7^/10  gallons of water treated. The  amount of carbon necessary
to operate the system for peak performance could be readily determined by observing:
        The aeration basin foam  level -  the proper operating level of carbon eliminates
        foam  in the system.
        Clarity of the effluent - the  presence of carbon in the  system removes the tint
        usually characteristic of biological  effluent.

     It is important to note that the powdered activated carbon utilized  in these trials has
a high bulk density of 44 Ibs/cu. ft.  This becomes an important factor when improved
settleability is the primary objective of the carbon addition to the system.  In addition,
the higher bulk density reduces the opportunity for carbon loss to the effluent.

DISCUSSION

James  F.  Dehnert: We have done considerable work with powdered carbon on a pilot basis
and we have done some of the work that you did.  A couple of problems we ran into and I
wondered if you had  solved them, that is the analysis of the mixed liquor solids.  Did you
determine actual  carbon percentage  or was this a calculated value?

Joyce  Rizzo:  We did it by a standard comparison.  What we did was make up standard

-------
                                                                                365
filters with carbon which we compared.  We took samples of our aeration system before
we added carbon to it.  We added known dosages of carbon and set up a set of standards.
So all the numbers we have obtained are   50 ppm, but we could keep track of it pretty
easily.

James F. Dehnert:  We had not devised a system to determine how much carbon was
actually in our system at any one time.  The second question:  you said you didn't test
very many parameters.  We have a problem  in California with meeting fish toxicity,  did
you ever run any tests on that?

Joyce Rizzo;  Everyone knows that refinery  effluent is not toxic.  Maybe California
is different - they may have a higher breed  of fish down there,  I don't know.  Obviously,
we didn't evaluate toxicity.

Les Lash;  You mentioned holding about 450 milligrams per liter of activated carbon,
Joyce,  in your aeration basin,  I guess.  What is the retention time  in that aeration basin?

Joyce Rizzo:  About 12 hours at average operating  conditions.

Les Lash;  When we ran the pilot plant on municipal waste for the EPA on powdered
activated carbon in Salt Lake,  we used reactive clarifiers and if you didn't allow enough
settling, of course the effluent went over black.  Now that is a little different  than the
clear you were talking about.

Joyce Rizzo:  The only time we ever had a carryover problem with carbon was during a
time when the bug population was upset and they would tend to take the carbon with
them.  If they were going out,  they weren't going to leave the carbon behind.  On a
normal basis we had no problem with carbon carryover at all.  It stayed right with the
sludge.   In fact, it improved the settleability  tremendously.  If you look at the clarifier,
it has the appearance of being  black; of course, the aeration system would turn black
versus the dark rich brown  color and it appeared like  it was all  carbon. The carbon
color would take over completely, and if you  looked  at the clarifier surface, it appeared
black but the effluent water was virtually clear.

Milton Beychok:  I am just a bit curious, I don't know if there is anyone here from
duPont or not,  but something like four years ago, duPont first used this and called it
PACT, powdered activated carbon treatment,  and they were licensing it.  Does the use
of this involve any patent problems or royalty  on  the part of duPont?

Joyce Rizzo:  No, this is quite a different system than what duPont is utilizing  in their
PACT project,  they are also utilizing a regeneration system which I believe is also a
part of their patent on that.  We used  ICI carbon, and the system that we used was out-
lined to us from ICI as to how to operate. I don't think we did anything really  different.
I  have talked to many people who are also using powdered carbon in their aeration
system in a similar manner.

Ed Sebesta:  I note you have a  centrifuge for your waste activated sludge there. Did

-------
366
  you continue to use a centrifuge both before and after carbon addition and what was the
  effect?

  Joyce  Rizzo:  We don't use a centrifuge all the time by the way, only when we have to.
  We have plenty of land out there for land farming and a  lot of times put the sludge out wet
  from the aerobic digester.  We did not see any real difference in our sludge handling
  characteristics on the solid waste handling system - none at all.  Actually, the amount of
  carbon that we used - our aerobic digester is three-quarter million  gallons like our aeration
  system - and the amount of carbon that  we used and the amount that was wasted was
  virtually lost in the system.

  Ed Sebesta:  I have another question and you indicated that "the effluent was very clear.  I
  think the data indicated about  an average of 40 parts per million suspended solids.   This
  is pretty high for a very clear effluent,  do you have any explanation for the concentration
  numbers?

  Joyce  Rizzo:  Very fine solids.  Color really is what I am talking about.  40 ppm does not
  represent optimal operation,  10-20 ppm would be more representative of the best effluent
  obtainable with the carbon.  The solids that would settle out by the way would be ones
  that would settle out after  hours.  You could take a sample  of the effluent and you wouldn't
  think there were suspended solids in  it,  but if you let it sit for a while, the solids that
  would  settle would be very, very fine.  I would say about half the  solids that went out in
  the effluent were carbon solids.  Very,  very fine solids.

  Pat DeJohn: I'd like to make a couple  of comments.   One that Mr. Lash made, the carbon
  that you were using has a density of about 45 pounds per cubic foot,  and consequently it
  settles very rapidly.  I believe what they were using out at  Salt Lake was a carbon that
  had a density of about 15 pounds per cubic foot.  The  higher density is  the reason the
  carbon doesn't go out over the  weir.  The other thing about the patent situation, duPont
  just got a patent issued  about two months  ago on the PACT process and right now there are
  some negotiations going on with respect to that.

  Janis Butler:  You mentioned sometime just spreading  the wet waste activated sludge.  Do
  you experience  any odor problems?

  Joyce  Rizzo:  No - none at all.  If we  put it out wet, the solids settle  rapidly, and there
  is no odor problem at all.  We  are using (I have been gone now from Corpus Christ! for
  about six months), but at the time I left we were using the centrifuge all the time for the
  waste.

  Anonymous:  Do you add carbon continuously?

  Joyce  Rizzo:  Corpus Christ!  has just purchased another load of carbon for usage.  As I
  understand it, their main goal is to use  it mostly for batch addition for shock loadings  and
  things  like that.

-------
                                                                                 367
Fred Goudy; Does the use of powdered carbon have any long range implications with
regard to the operation of the activated sludge plant at Corpus Christ!?

Joyce Rizzo:  They are using it.

Fred Goudy: Continuously?

Joyce Rizzo;  I am not sure, I can't answer that for fact right now.  I know they have
just purchased some more carbon, and I know they have it in stock and I know that their
intentions are to use it.  Now whether they will use it all the time, I don't know at this
time.  I might mention what I think about carbon application in the aeration system.
Like I said, although we really were oriented  to suspended solids  removal, it is obvious
that some of the other things that I  have mentioned about the system are very true.   That
is, if you have a plant that is sitting there hydraulically loaded and you need more
capacity, it seems to help  and gives you that extra capacity that  you  need.  As I
mentioned none of the reductions that I presented, of course, should be taken as absolute
numbers because it was pretrial versus trial period. But there is no question that the
improvement in the plant performance is there and I think the improvement is really
dependent on the variability of your influent.

Dave Story: You mentioned adding polymers;  could you elaborate a little bit more on
that?

Joyce Rizzo:  Yes, we didn't do too much work with polymers because it was the very
first thing we tried to reduce suspended solids. We were adding about 5 or 6 parts per
million of some polyelectrolytes to  our aeration basin overflow and we saw a reduction
in effluent suspended solids and a little bit better settling characteristics, but nothing
very significant.   I think this is mostly due to  our foaming problem because we still  had
a lot of foam carrying over to the clarifier which is a significant problem.

Anonymous:  How do you account for the reduction in sludge wasting?

Joyce Rizzo:  We didn't have to maintain as many bugs in the system and they didn't have
to work as hard.  But the real reduction was in volume not mass.  The recycle sludge
was thicker and more uniform with carbon addition.

BIOGRAPHY

     Joyce A. Rizzo is a staff engineer  in the
Advanced Management and Methods Department of
Suntech, Inc., a subsidiary of Sun  Oil  Company.
She is based in Marcus Hook, Pennsylvania.  Ms.
Rizzo joined Sun Oil  in 1971 as a process engineer at
their Marcus Hook Refinery. Prior  to her  move into
Suntech six months ago, she spent three years  in
Process Engineering at Sun's Corpus Christ! Refinery.
She holds a B.S. in Chemical Engineering from
Northeastern University in Boston.

-------
   368
                                                TABLE 1

                      "OPERATING CONDITIONS OF WASTEWATER TREATMENT PLANT1

FLOW
    Daily average    1.08MGD
    Daily maximum   2.16 MGD
    Design           2.16 MGD

BIOLOGICAL SYSTEM        2 - 760,000 Gal.  Rectangular Aeration Basins
                            2-75 HP.  Floating Aerators/Basin

    Recycle                 50% of Charge Rate
    MLSS                   3000 PPM
    MLVSS                  2200 PPM
    Retention                   11 Hours  at Maximum Design Rate
    (Per Basin)                  22 Hours  at Average Operating Rate
    Sludge  Age                 12 Days

SETTLING                   2-55 Ft. Diameter Circular Clarifiers 12 Ft. Deep

                            Average Rise Rate, GPD/FT2           227
                            Peak Rise Rate,  GPD/Ft               455

SOLIDS HANDLING
    1   760,000 Gallon Aerobic Digester
    1   Solid Bowl Centrifuge
    4  Acres of Land Farm
    Wasting Rate 30,000 GPD
    VSS          7,500 PPM
                                                TABLE 2

                    "SUMMARY OF POWDERED ACTIVATED CARBON ADDITION TRIALS"

    Aeration System:   450 PPM PAC
    Recycle System:   1000 PPM PAC

                                                          Average                            Treatment
                                                        Carbon Dosage ,,                         Cost*
                      Period         Duration          Lbs/D      Lbs/IO  Gal       PPM        $/10 Gal

                                                                   0.20           24             4.3


                                                                   0.16           19             3.5


                                                                   0.076           9             1.7
                 July,  1975           31 Days

    "At 22 <:/Lb
Trial
B>-:se Case
Trial
Base Case
Trial
Base Case
1
1
2
2
3
3
Sept. -Nov., 1974
August, 1974
May, 1975
April, 1975
August, 1975
July, 1975
82
30
28
30
26
31
Days
Days
Days
Days
Days
Days
179
152
87

-------
                                                                                         369
                                           TABLE 3

         "EFFECT OF POWDERED CARBON ON DAILY AVERAGE EFFLUENT SUSPENDED SOLIDS'

                              1977 Permit: 327 Lbs/Day Daily Average
Base Case  1
Trial      1

Base Case  2
Trial      2

Base Case  3
Trial      3
             Flow
             GPM
              626
              629

              645
              657

              839
              799
Effluent
PPM
115
50
163
72
79
42

Lbs/Day
861
381
1262
565
799
405
Reduction

  55.7


  55.2


  49.3
Base
Trial
Base
Trial
Base
Trial
1
1
2
2
3
3
                                           TABLE 4
                            'EFFECT OF POWDERED CARBON ON COD"


Base Case 1
Trial 1
Base Case 2
Trial 2
Base Case 3
Trial 3
Flow
GPM
626
629
645
657
839
799
Influent
PPM
459
457
343
444
367
379
Lbs/D
3445
3446
2658
3500
3698
3632
Effluent % %
PPM
170
135
266
183
166
112
Lbs/D
1277
1020
2059
1445
1670
1073
Removal
62.9
70.4
22.5
58.7
54.8
70.5
Reduction

20.1

29.8

35.7
                                            TABLE 5
                       'EFFECT OF POWDERED ACTIVATED CARBON ON BOD"
Flow
GPM
626
629
645
657
839
799
Influent
PPM
152
213
152
227
188
207

Lbs/D
1144
1607
1173
1898
1895
1981

PPM
15
15
30
13.
12
3
Effluent
Lbs/D
116
114
232
5 113
124
30
Removal
%
89.9
92.9
80.2
94.0
93.5
98.5
Reduction
%

1.7

51.3

75.8

-------
                                                                                             370
                                            TABLE 6

                           'COMPARISON OF CARBON TREATMENT COSTS
                                  AND SYSTEM PERFORMANCE"
                                                                % Reduction*
Trial
Trial
Trial
1
2
3
Dose
PPM
24
19
9
3

-------
                                           FIGURE  I
                           TREATMENT   PLANT  FLOW  DIAGRAM
      BALLAST
         a
     TANK FARM
 REFINERY
   API
SEPARATOR
 EFFLUENT
EQUALIZATION
     a
IMPOUNDMENT
                         POWDERED
                         |  CARBON
                         | ADDITION
                                         AERATION
  FINAL
EFFLUENT

AEROBIC
DIGESTER



^ • v*L.I1 1 l\lr UOC r

^
SLUDGE
TO
LAND a
                                                                     FARM

-------
  372
                               FIGURE  2
EFFLUENT
   TSS
 LBS/DAY
2800

2400

2000

1600

1200

 800

 400

   0
              I
                      EFFECT  OF POWDERED  CARBON
                       ON EFFLUENT  TSS  LOADING
                                             CARBON
                            i	i
I	I
I
I
            10     30   50  70    90 95
             %  OF VALUES LESS  THAN
                99
                                                         TRIAL 3
            240

            200

            160
 EFFLUENT
    TSS     120
    PPM
            80

            40

             0
               I
                               FIGURE  3

                      EFFECT OF  POWDERED  CARBON
                            ON  EFFLUENT  TSS
                     BASE CASE
                                                         TRIAL 3
                                WITH  CARBON
                   j	i
         i
             10     30   50  70    90 95

              %  OF VALUES LESS  THAN
                 99

-------
    COD
 REMOVAL
                             FIGURE  4

              COMPARISON  OF  COD  REMOVAL  EFFICIENCY
                                               373
                                                    TRIAL  3
                     CONTROL	»-H	TRIAL
                     PERIOD           AUGUST
                  JULY,  1975           1975
EFFLUENT
   COD
   PPM
300
280


240


200


160


120


 80


 40


  0
              I
                              FIGURE 5

                     EFFECT  OF  POWDERED CARBON
                           ON EFFLUENT  COD
                            BASE CASE
                                          CARBON
                                                       TRUL 3
                                       J_

           10    30  50  70    90 95

             % OF  VALUES  LESS THAN
                                                  99

-------
                               FIGURE  6
EFFLUENT
   BOD
   PPM
            60
            50
            40
 30
            20
            10
                    EFFECT  OF POWDERED  CARBON
                           ON  EFFLUENT  BOD
                                                       TRIAL 3
                    I
                                              „- WITH
                                              CARBON
                                 _L
                   5  10     30   50   70     90  95

                       %  OF VALUES  LESS THAN
                                      99
  BOD
REMOVAL
100
 98


 94


 90


 86


 82


 78


 74
              -WITH _..
              - CARBON
                              FIGURE 7

                   EFFECT  OF  POWDERED  CARBON
                          ON  BOD  REMOVAL
                            i  i  i  i  i
                                                       TRIAL  3
                      10
                30   50
70
90 95
99
                    o/
                     •'O
            OF VALUES  LESS  THAN

-------
            "PILOT PLANT ACTIVATED CARBON TREATMENT OF
                    PETROLEUM REFINERY WASTEWATER"

           Jack H. Hale, Leon H.  Myers, and Thomas E. Short, Jr.
               Robert S. Kerr Environmental Research Laboratory
INTRODUCTION

     One of the first documented uses of carbon for environmental control was in  1793
when a physician used charcoal to remove the odor associated with gangrene. About
sixty years later a scientist by the  name of Stenhouse recommended the use of charcoal to
remove the odors from sewers. Potable water was "purified" by carbon in 1862. ^  Moon-
shiners in the hills of Tennessee were  using  charcoal, not to mellow their product, but to
remove "hog track" odors.

     Since the mid-1960's, there has been an increasing effort to utilize activated carbon
as a secondary or tertiary treatment system to treat wastewater. Two of the most success-
ful uses of activated carbon were demonstrated at Lake Tahoe and in Pomona, California,
to treat domestic municipal wastewaters. It is,  of course, a natural progression to treat
industrial wastewaters with systems that appear successful in treating municipal waste-
waters.

     The first domestic petroleum refinery to use activated carbon treatment was the
Atlantic Richfield Refinery in Carson, California.  The system  was designed for inter-
mittent use to treat rainfall runoff  and process wastewater during storm periods.  The
second application was designed for the BP  Refinery, Marcus Hook,  Pennsylvania, to
treat process wastewaters.  Unlike  the Arco system,  the BP system was designed to operate
in a continuous mode. Neither system relied on biological treatment preceding the
activated carbon system.

     During  the late  1960's, activated carbon treatment of industrial wastewater was
gaining momentum as the treatment system of the future.  Statements such as "organic
removal" or "removal of dissolved  organics  from wastewater has been demonstrated" and
"activated carbon treatment systems will remove dissolved organic contaminants"  were
prevalent.   These  innocent gross statements were translated rather rapidly to imply that if
you had an organic waste treatment problem, it could be  solved with granular activated
carbon.

     With the complete cooperation of Kerr-McGee Petroleum  Refinery,  Wynnewood,
Oklahoma, a study program  was devised to  investigate the adsorptive capacity of
activated carbon.

ISOTHERM STUDIES

    Adsorption isotherms are used  to indicate the effectiveness of an activated carbon
for a specific wastewater under controlled conditions. When the isotherm is to be used as

                                     375

-------
376

a predictive tool, the wastewater should be as representative as possible, pH and
temperature adjustments should not be made, and sample collections need to be made in
glass containers.

CARBON EVALUATION

     Eleven commercially available carbons were evaluated for this petroleum refinery's
wastewater.  It is necessary to point out that adsorption varies with the carbon being
evaluated and the water sample.  The data presented is based on specific activated carbon
for one petroleum refinery.  Comparable results may or may not  be achieved using the
same carbon at another refinery.

     The controlled conditions used  for comparison purposes of the eleven carbons are
shown  in Table 1  .  Both pulverized  and granular modes were evaluated.

     The waste sample to be evaluated  was collected from the API separator discharge,
settled for four hours, and the candidate water was drawn from the middle of the vessel.
The adsorption capacity of one gram sample of the candidate carbons,  both pulverized
and granular, is shown in Table 2.

     Graphically expressed, the comparison of pulverized and granular adsorption
capacities appears in Figure 1 .

CONTACT TIME

     A major factor relating to adsorption  is the amount of time the water to be treated is
in contact with the activated carbon.  A sample of wastewater from another refinery was
obtained for these studies. Pulverized activated  carbon (Filtrasorb 400) was weighed to
obtain 0.1, 0.5,  1.0, 2.0, and 5.0 grams of sample, and the isotherm procedure was
followed for three hours.  Samples were obtained at 20, 40, 60, and 180 minutes,
filtered, and analyzed for total organic carbon.   Table 3  represents the typical results.

ISOTHERM TESTING

     Figure 2 shows the results of an isotherm carried out on wastewater from the refinery
used in this study and a candidate activated carbon.  Obviously, this curve does not
follow the typical "Freundlich" isotherm.   Instead of the usual straight line,  a curve
resulted.  At the lower end of the curve the amount of carbon added to the wastewater is
increased but  the concentration of TOC does not  decrease as much as would be expected.
In fact, the concentration of TOC changes only  very slightly at high carbon dosage.
Refinery wastewater is composed of a rather complex mixture of  materials.  Some of these
materials are readily adsorbable while  others are  not.  The initial amount of carbon picks
up the easily adsorbed materials and leaves behind  the others.  Therefore,  increasing the
carbon dosage does not reduce the concentration  of TOC to the  same degree.  In fact,
there is a possibility that  this wastewater contains non-adsorbable components that cannot
be removed by carbon treatment.

-------
                                                                               377
     From the "isotherm," the adsorptive capacity of the carbon, tested with this
specific wastewater,  projects to approximately 0.1 Ib of TOC per Ib of carbon.  This was
considered to be a reasonable value and indicated that further investigation of activated
carbon treatment was justified.

     Results from isotherm tests are considered to be valuable  only from an initial
screening standpoint.  Their results are generally not sufficient for predicting results of
full-scale granular activated carbon treatment.  For a better  evaluation of activated
carbon treatment, continuous column pilot plant investigation at the waste stream site
yields much more reliable results.

ANTHRAFILT FILTRATION

     When treating oily wastes by activated carbon, it appears that  a pretreatment system
is needed to assist in the removal of insoluble oils and suspended matter. An anthrafilt
downflow filter system was evaluated for effectiveness, again using  filtrable TOC as the
prime parameter.  The glass  column used for this exercise was five feet in  length and 1 .5
inches in diameter.  A diagram of the system is shown in Figure 3.   Total organic carbon
results obtained at these time intervals are shown in Table 4.

     This particular study lasted 2.5 hours before a pressure drop was noted.  The charge
water contained an appreciable quantity of oil, and the oil percolated into the  anthrafilt
layer during the study period.  When the column was backwashed with distilled  water,
the percolated oil was easily removed from  the anthrafilt,  leaving an apparently clean
bed.

MINI-COLUMN STUDY

     A downflow mini-column system was designed using stock one inch ID glass tubing
that was six feet in length.   The system was designed with an  electronic sampler to
composite hourly samples.  A diagram  of this system  is shown  in Figure 4.

     The primary study involved dividing the six columns into  two sets of three columns
each to evaluate comparison of treatment effectiveness.   Each column was packed with
1,000 cc of Filtrasorb 400 using potable water was the wetting liquid.  The flow rate was
set at 400 ml/minute  and a column pressure of 6-7 psig was used.  Total organic  carbon
was the primary parameter used for treatment effectiveness. Analytical  results of the
primary study are recorded in Table 5.

     Comparing each  column for removal efficiency, using the anthrafilt effluent TOC as
the base,  the percentages obtained are shown in  Table 6.  Columns  1 and 2 compare very
well for the five-day averages, although they do exhibit considerable daily variances.
Columns A-3 and B-3 are not agreeable, and no  reason for this deviation is projected.

    A second study was conducted by  connecting five columns in an upflow mode.  Each
column contained  1,000 cc  of Filtrasorb 400 packed in the same manner as the previous
study.  TOC results of this study are shown  in Table  7.

-------
378

     Percentage reductions of each column based on the anthrafilt effluent are shown in
Table 8.  The average column trend indicates there  is an  increase of treatment effective-
ness with  each succeeding  column, as one would expect.

PETROLEUM REFINERY DATA

     Listed in Table 9 is the pertinent information needed  to characterize the refinery
where the study was conducted.  This refinery was a 30,000 barrel, Class "B" refinery
(using the API classification  system);  processes included fluid cat cracking, HF alkylation,
catalytic  reforming, and asphalt  production.

     The existing refinery eastewater  treatment system is shown in  Figure 5.  Wastewater
from the refinery includes cooling tower blowdown,  boiler blowdown, oily process water,
stripper effluent, and contaminated runoff.  These wastes  flow  to an API separator for oil
removal.  The effluent from the separator is treated  in a "Pasveer  Ditch" activated sludge
treatment  system.  The biologically treated  effluent then  flows through a series of holding
ponds and the effluent from the final  pond is discharged to a small stream.

     Two complete pilot plants  were installed and operated simultaneously, one on the
refinery's  API separator effluent (secondary) and  the other on the clarifier effluent from
the biological treatment system (tertiary).

ACTIVATED  CARBON TREATMENT PILOT PLANT

     Figure 6 contains a flow diagram of the pilot activated carbon treatment systems.
The  wastewater  to be treated first flows through a dual-media filter constructed of 4-inch
PVC pipe.  This filter consisted of an 18-inch layer of sand over pea gravel, topped with
a 6-inch layer of anthrafilt.

     Dual-media filtration  pretreatment was chosen because of  its  reliability  and  effective-
ness, and  because it does not require the use of iron or aluminum salts as coagulants.
This latter point is particularly important insofar  as regeneration of activated carbon  is
concerned,  since aluminum salts  during regeneration can  remain on the surface of the
carbon at  high temperatures.  These salts can become permanently attached to the surface.
Thus, the  effective surface area of the carbon is  reduced  and its adsorption capacity  is
seriously reduced.  Iron salts present  a similar problem .  In addition, these salts  can
catalyze oxidation reactions of the carbon and the gases in the  regenerator.  Thus, the
structure of  the  carbon becomes permanently damaged.

     After pretreatment,  the  wastewater  entered a "Calgon" activated carbon pilot plant.
This plant was set up so that  the wastewater flowed down  three of  the 5-inch ID columns.
The  first column contained an 18-inch layer of granular activated carbon while the
remaining columns had a 36-inch  layer of carbon.

     The flow rate through  each pilot plant  was adjusted to 1/4 gpm.  During the
operation  of  the pilot plant,  samples  of the  API separator  effluent, biological treatment
effluent, and both pilot plants' effluents were taken every two hours.   These samples were

-------
                                                                                379
composited and preserved according to recommended EPA methods.  Twenty-four hour
composite samples were analyzed daily for a spectrum of water pollution control
parameters using EPA analytical methodology and analytical quality control.techniques.

     The dual-media filter and  carbon columns were backwashed whenever the pressure in
the first column exceeded 20 psi.  The pilot plants were  operated over a 10-day period,
at the end of which time the first columns in both plants  were  near exhaustion.

     Figure 7 shows the BOD5 daily composite analysis for the API separator effluent
before and after treatment by the various schemes studied.  Activated carbon treatment
was  not able to achieve the same level of BOD5 reduction as bio-treatment.  In fact,
bio-treatment did a considerably better job. However, carbon treatment following bio-
treatment did  show further reductions in BOD.

     Figure 8 shows the COD daily  composite results for the same wastewater  streams.
Apparently, both bio-treatment and carbon treatment effect about the same COD
reductions.  Carbon  treatment following bio-treatment yielded the best  reduction of COD.

     The TOC results  for the daily composites are shown in Figure 9. Unlike  the BOD5
and  COD results,  the TOC indicates increased removal can be obtained by the API-
carbon combination over the bio-treatment system alone.  There is an erratic behavior
pattern which indicates unadsorbable organics passing through the carbon columns.

     From the  BOD5 and COD plots, it may be concluded that for secondary treatment
alone, on the wastewater studied, the biological treatment system is preferable because
it gives the greatest  BOD^ reduction and gives COD reduction equivalent to  carbon
treatment.  The best levels of reduction were obtained with biological  treatment followed
by carbon adsorption.

     Table  10  gives the results of other parameters evaluated in this study.  As would be
expected, both biological and  activated carbon  treatments are able to produce signifi-
cant reduction in the organic parameters, such as BOD, COD, TOC, oil and grease, and
phenols.  It should be noted that carbon has a particular  affinity for the removal of
phenols.  The color and turbidity are also improved.  Cyanides and ammonia, for all
practical purposes, were not removed by either of these treatments.

     Sulfides are a peculiar  problem for activated carbon treatment.  Apparently, the
sulfide content increases as  it goes  through the carbon s/stem.  This is probably due to
biological activity occurring at the surface of the carbon.  The sulfur containing
compounds which are adsorbed  upon the carbon are reduced anaerobically to  h^S giving
rise to the increase in sulfide content.  Bio-treatment, on the other hand, decreases the
sulfide content very significantly.

     One of the least expected  removals by activated carbon was observed for the metals
—chromium, copper, iron, and zinc.  These metals are significantly reduced by the
carbon.  Whether these removals are due to adsorption, filtration, or other phenomena
has not been evaluated.   However, it must be kept in mind that these removals were

-------
380

 accomplished with virgin carbon. The effect of/some of these metals upon carbon
 regeneration has not been determined. Also, whether regenerated carbon can remove
 equivalent amounts of metals has not been established.

      As mentioned previously, the isotherm study indicated that the carbon could adsorb
 0.1  Ib TOC/lb of carbon.  The  columns on the API separator e/fluent indicated a capacity
 of 0.31 Ib TOC/lb of  carbon, as shown in Table 11 .   The  most probable explanation for
 the difference in the capacity as determined in the isotherm test and the pilot plant study
 is biological activity in the columns.

      In  the case of the API  separator effluent carbon  column, this biological activity
 manifested itself by the anaerobic production of H^S. While the increased capacity is a
 desirable situation, the production of sulfides is not.  Although anaerobic activity was
 not apparent in  the clarifier effluent columns,  it was observed that algae started growing
 in the carbon beds.  In fact, it  was  necessary to cover the columns to prevent sunlight
 from making this growth possible.

      Because of  the considerable cost of granular activated carbon,  regeneration of the
 spent carbon is essential if  the cost  of treatment is to be kept at a moderate level.
 Samples of spent carbon were  taken  from the first column of each pilot plant and sent to
 Calgon Corporation for regeneration studies.  Calgon regenerated these two carbons using
 their standard evaluation techniques in a muffle furnace at 1750° F and with a normal air,
 flue gas, and steam atmosphere.

      As shown in Table 12,  both carbon samples were regenerated to a good activity as
 indicated by the iodine numbers obtained.  The regenerated clarifier carbon appeared to
 have a  little better activity than the API separator carbon.  In general, the iodine number
 can be  related to the surface area of pores larger than 10 A  in diameter.  The molasses
 number, likewise, is related to  the  surface of pores larger than about 30 A.  As  can be
 observed  in Table 15,  there is an increase in the molasses  number with a slight decrease
 in iodine number.  This indicates a  shift in pore size distribution.  Apparently, the
 smaller pores are being destroyed while larger  pores are  being created.  Final clarifier
 carbon  appeared to have  less destruction of the pore  matrix than did the API separator
 carbon.

 GAS CHROMATOGRAPHY STUDIES

      Studies  by Dr. T. C. Dorris and Dr. S. L. Burks at  Oklahoma State University have
 shown that some  organics present in  refinery wastewaters are toxic to fish,  and some
 organics are refractory to biological degradation.    These organic chemicals persist in
 lakes and streams for long periods of time.  It is also noted that a refractory organic can-
 not be detected by the TOD test that has been used in the past to evaluate the efficiency
 of treatment systems.  Tests such as  TOD,  TOC, and COD are much more realistic for the
 indication of organics in a  water sample.  Of these tests,  TOC is the one that measures
 only organics.

      The laboratory researchers  must go further and use sophisticated methods for actually
 identifying and  characterizing the refractory organics so that engineers will know the

-------
                                                                                 381


type of organics that must be removed,  thus designing treatment systems to improve the
quality of water discharged to the receiving streams.

     Carbon treatment  is a  relatively efficient method for removing organics from refinery
wastewater and may be used as a secondary or tertiary treatment system to reduce the BOD
of effluent water to a very low level.  In studies to  learn more about carbon treatment,
carbon columns were installed at a petroleum refinery wastewater treatment plant at a
point immediately after the API  separator and at a point after the final clarifier for the
activated sludge treatment unit.  The carbon removed from these columns was  extracted
with several organic solvents to determine which solvents were most efficient and to
provide samples of the organics for silica gel, gas chromatographic (GC), infrared, and
GC-mass spectrometric analyses.

     About 100 ml of an air dried carbon sample were weighed and  placed directly into
the soxhlet extraction  apparatus with a glass wool plug at the bottom to prevent the
carbon from getting into the siphon tube.  Three hundred ml of solvent were used in the
flask.  The rate of solvent reflux was adjusted to give four cycles per hour, and the
extraction was continued for 22 hours.  After 22 hours, the solvent-oil mixture was
cooled to room temperature and filtered  through a 10 cm depth of anhydrous sodium
sulfate crystals contained in a 20 mm by 25 cm glass tube to dry the mixture and remove
any  carbon particles that might be present. The dried solvent-oil mixture was then
distilled  through a 20 mm diameter column containing 12 cm  of glass beads until the
volume remaining  in the flask was slightly less than 50 ml.  The concentrated solvent-oil
mixture was measured into a sample  vial, and the volume adjusted to exactly 50 ml with
rinsings from the flask.

     A gas chromatograph, with a flame detector and an electronic  integrator,  was used
to determine the amount of oil extracted with each solvent.  A 1/8-inch by 20-foot
stainless  steel column,  packed with  70/80 mesh Chromosorb G, AW-DMCS, with 5%
OV-101  stationary phase,  was programmed to hold at room temperature for four minutes
with the  oven lid open,  then the lid was closed and  the  program proceeded from 35° to
65°  C at 8 degrees per minute and from 65° to 350° C at 6 degrees per minute.  The
temperature was held at 350° C for up to 20 minutes, if  it appeared necessary to elute all
of the sample.  Injector and detector temperatures were  about 370° C.

     A solution of 3 gm of diesel fuel diluted to 50 ml with solvent was used as a standard
since it was available, although an  oil with a higher aromatic content would have been
more desirable.  The total  area of the chromatogram of the standard, excluding  the
solvent peak and correcting for baseline drift near the end of the chromatogram, was
compared with the total areas obtained from the 50 ml samples of extract.  Data obtained
from the  carbon extractions and  GC analyses are  shown in Table 13.

     Chloroform,  benzene, and methylene chloride showed about the same efficiency for
extracting oil from the carbon, while hexane and methanol were less efficient.  Figures
10 through 14 show the chromatograms obtained on the extracts from API  carbon using the
five  different solvents.  Comparison of the benzene, methylene chloride, and chloroform
extracts (Figures 10, 11, and 14) shows that the oils extracted with these solvents were

-------
382

almost identical in composition.  However, chloroform  showed an advantage in the
recovery of one low boiling compound which was an appreciable amount of the sample as
indicated from  the size of the peak.  This peak had a retention time of 11.3 minutes
which corresponds to the retention time of benzene.  This peak could not be seen when
benzene was used as the  solvent, and shows at a much lower concentration in the
methylene chloride extract.

     Chloroform was used as the solvent for extracting a sample of carbon that had been
used to treat effluent water from the final clarifier of the activated sludge treatment
system.  This carbon is referred to as  FC carbon, and the resultant chromatogram is shown
in Figure 15.  The FC carbon contained about 1/3 as much oil as the API carbon.  A
sample of new carbon also  was extracted with chloroform to determine  whether the new
carbon contained any extractable material.  This extract contained no measurable oil, as
illustrated by Figure 16.  This figure also serves as a good illustration of the purity of the
chloroform used as the solvent.  Figure 17 is a chromatogram of the standard mixture used
to quantitate area data in terms of grams of oil extracted.

     The API carbon extract and FC carbon extract were each separated into three
fractions by selective elution and desorption from silica gel. The columns were pre-wet
with hexane and as the last drop of hexane disappeared into the surface of the gel, the
sample-gel  mixtures in the  10 ml beakers were emptied  into their respective columns.
The  samples were then fractionated by eluting the  saturates with  10 ml of hexane,  the
aromatics with  10 ml of benzene, and finally desorbing the polar fraction with methanol.
Each of the fractions was then reduced in volume to 0.5 ml by evaporating the solvents
with a gentle stream of air.

     Gas chromatography was used to obtain an estimate of the amount of saturates,
aromatics, and polar material in each of the samples.  The oil extracted from the FC
carbon contained such small  amounts  of saturates and aromatics that  it  was necessary to
increase the sensitivity of the GC by a factor of 10 to obtain measurable areas. These
areas were then divided  by  10 to get  back to the same basis as the area measurements on
the polar fraction.   The  results are  tabulated in Table  14.

     These data  indicate  that the activated sludge  treatment system has reduced the
saturates and aromatics to very low levels in the final clarifier effluent but has been
relatively ineffective in  removing some of the polar type organic compounds.

     Figures 18, 20, and 22 are chromatograms of  the saturate, aromatic, and polar
fractions obtained by the silica gel  separation of the oil from API carbon.  Figures  19, 21,
and 23 are chromatograms of the saturate, aromatic, and polar fractions obtained by the
silica gel separation of the oil  from FC carbon.

     There was  insufficient  time for complete GC-MS identifications, but Table 15 lists
some of the compounds and types that have been found to date  in the saturate, aromatic,
and polar fractions.  The organic structures shown  in the table are those normally found in
petroleum and petroleum products,  but it is noted  that three sulfur compounds were found
in the aromatic fraction  of  the  FC carbon extract.   It may be that these compounds are not

-------
                                                                               383


biologically degraded as much as the aromatics and are a more significant part of the
aromatic fraction after biological treatment.

     The polar fractions could not be analyzed by GC-MS because of the inability of the
gas chromatograph to separate  the complex mixture. The complexity of the mixture is
illustrated in Figures 22 and 23 by the lack  of baseline separation.  Differences in the
two  chromatograms indicate that some of the components have been degraded in the
activated sludge treatment system.   It is probable that the reduction in phenolic com-
pounds, that js known to occur, would account for some of the differences noted in the
ch rom a tog ram s.

     Infrared scans on the polar fractions were almost identical.  Functional groups noted
were OH and C=O, and CH2 and CH3 adsorption bands indicate aliphatic type structures.
Further work must be done on the polar fractions to identify a number of the individual
compounds in the mixture so that an assessment of their toxicity can be made and treat-
ment methods developed to remove  these compounds.

REFERENCES

1.   Hansler, John W.  Activated Carbon.  Chemical Publishing Company,  Inc.,
     New York, New York, 1963.

2.   Burks, S. L., T.  C. Dorris,  and G. L. Walker.  Identification of Toxic Organic
     Chemical Compounds in Oil Refinery Wastewaters. Technical Completion Report to
     U.S.D.I. Office of Water Resources Research, Project BO-17, Oklahoma, 1970.
     Unpublished.

3.   Hale,  J.  H. and L. H. Myers.   The Organics Removed by Carbon Treatment of
     Refinery Wastewater.  Presented at the Oklahoma  Industrial Waste and Advanced
     Water Conference, Stillwater,  Oklahoma, April 1973. Robert S. Kerr Environmental
     Research Laboratory, Ada, Oklahoma.   Unpublished.

BIOGRAPHIES

     Jack H. Hale holds a BS in Chemical Engineering  from
Oklahoma State University. He is currently a research
chemist in the Industrial Section of the Source Management
Branch at the Robert S.  Kerr Environmental  Research
Laboratory,  Ada, Oklahoma.
    Leon H. Myers holds a BS in chemistry/biology from
Southwestern Oklahoma State University and a MS in
sanitary science from Oklahoma University.  He is
currently Chief, Industrial Section of the  Source Manage-
ment Branch at the Robert  S. Kerr Environmental Research
Laboratory,  Ada, Oklahoma.

-------
 384

     Thomas E. Short holds a BS in Chemical  Engineering from
 Lamar University and a MS and Ph.D. in Chemical
 Engineering from Oklahoma State University. He is currently
 a Chemical Engineer at the Robert S. Kerr Environmental
 Research Laboratory at Ada, Oklahoma.

DISCUSSION

Bill McCarthy: When you used solvent extraction for recovering your adsorbent components,
could you give us an idea of what efficiency or recovery you got?

Leon H. Myers: No I  can't, because we did not attempt a mass balance; however, the quan-
titative value  of three solvents produced similar results.   The quantity recovered agrees with
the projected isotherm loading.

Bill McCarthy:  Do you think solvent regeneration might be a viable process?

Leon H. Myers:  Yes, I do believe it is both  feasible and viable,  and this is determinant on
the mixtures of hydrocarbons to  be removed.   In  some cases, particularly with light hydro-
carbons, steam regeneration might be the most viable regeneration process.   Neither of these
alternate regeneration modes have been thoroughly proven.
                           MYERS                      HALE

-------
                                                                          385

       Table 1. CONTROLLED CONDITIONS FOR ISOTHERM STUDY

       1 .   325 mesh (pulverized)             5.   100 rpm agitation
       2.   200 ml sample                   6.  23° C temperature
       3.   1 hour contact                   7.  75 mg/l TOC
       4.   0.05, 0.20, 0.50, 1.0 gm         8.  pH 7.4
           carbon
     Table 2.  ADSORPTION DATA FOR ELEVEN ACTIVATED CARBONS

Carbon Sample                          mg/l TOC Adsorbed on 1 gm Carbon

                                      Pulverized              Granular
1
2
3
4
5
6
7
8
9
10
11
50
60
56
57
56
52
47
51
49
35
46
38
37
37
48
29
31
41
47
38
13
34
                 TableS. CONTACT TIME ADSORPTION
                                            Time in Minutes
Gm Carbon
(Blank)
0.1
0.5
1.0
2.0
5.0
20
(206 mg/l)*
124
63
57
55
56
40
(260 mg/l)*
96
45
47
52
54
60
(260 mg/l)*
82
51
106
77
50
180
(260 mg/l)*
101
73
45
34
24
* Results are expressed as mg/l filtrable total organic carbon.

-------
386
           Table 4. ANTHRAFILT FILTRATION ORGANIC CARBON REMOVAL
                                               TOC
                                     Turbidity
       Charge
       After 5 minutes (0.3 gal.)
       After 1 hour (3.2 gal.)
       After 2.5 hours (7.9 gal.)
                 255 mg/l
                 220 mg/l
                 210 mg/l
                 206 mg/l
                             11.0
                              2.75
                       Table 5.  MINI-COLUMN TOC RESULTS
Date
3/17
3/18
3/19
3/20
3/21
Final Clarifier
Effluent*
35
33
48
37
35
Anthrafilt
Effluent*
31
34
40
48
38
Columns
A-l*
13
11
13
17
18
B-l*
9
11
16
23
19
A-2*
9
10
14
22
17
B-2*
10
12
14
16
25
A -3*
7
10
8
11
16
B-3*
12
18
12
22
32
  All concentrations are mg/l TOC.
        Table 6.  COMPARISON OF MINI-COLUMN TOC REMOVAL EFFICIENCY
Date

3/17
3/18
3/19
3/20
3/21
A-l

 58
 68
 68
 65
 53
B-l

 71
 68
 60
 52
 50
                                            Percentage Removals
A-2

 71
 71
 65
 54
 55
68
65
65
67
34
77
71
80
77
58
B-3

 62
 47
 70
 54
 16
Average
 62
 60
 63
60
73
 50

-------
                                                                          387
        Table 7.  TOC RESULTS OF MINI-COLUMN UPFLOW MODE

Date
3/24
3/25
3/26
3/27
Final Clarifier Anthrafilt
Effluent*
50
35
41
38
Effluent*
38
33
33
35
1*
19
17
20
17
Columns
2*
12
13
24
19
3*
11
10
18
17
4*
13
10
13
13
5*
13
9
12
10
  All concentrations are mg/l TOC.
      Table 8.  PERCENTAGE REMOVAL OF TOC FOR UPFLOW MOJDE

Date
3/24
3/25
3/26
3/27

]
50
48
39
51

2
68
61
27
46
Columns
3
71
70
45
51

4
66
70
61
62

5
66
73
64
71
Average
 47
51
59
65
69
                   Table 9.  REFINERY PROCESS DATA
       Capacity

       API Class

       Wastewater Volume

       Refinery Processes:
30,000 BPSD

B

18 gal. per minute

1 .   Crude Distillation
2.   Crude Desalting
3.   Vacuum Crude Distillation
4.   Fluid Cat Cracking
5.   HFAIkylation
6.   Hydro Cracking
7.   Catalytic Reforming
8.   Asphalt Production

-------
388
                    Table 10. REFINERY WASTEWATER TREATMENT RESULTS
Median Values


Parameter
BOD5
COD
TOC
Oil and Grease
Phenols
Chromium
Copper
Iron
Lead
Zinc
Sulfide
Ammonia
Cyanides
Turbidity*
Color**

API Separator
(mg/l)
97
234
56
29
3.4
2.2
0.5
2.2
0.2
0.7
33
28
0.25
26
30

Bio-Treated
(mg/l)
7
98
30
10
0.01
0.9
0.1
3.0
0.2
0.4
0.2
27
0.2
17
15

Carbon-Treated
(mg/l)
48
103
14
10
0.004
0.2
0.03
0.3
0.2
0.08
39
28
0.2
11
15
Bio-Carbon
Treated
(mg/l)
3
26
7
7
0.001
0.2
0.05
0.9
0.2
0.15
0.2
27
0.2
5
1
 *  Turbidity given in Jackson Turbidity Units.
 ** Color given in color units.
                        Table 11.  CARBON ADSORPTION CAPACITY
                                        Capacity Lbs TOC Adsorbed Per Lbs Carbon
          Isotherm Study

          API Separator Effluent
              Carbon Column
         0.12

         0.31
                   Table 12.  CARBON REGENERATION ACTIVITY ANALYSIS
                                             Iodine No.
                      Molasses No.
          Virgin Carbon
          API Separator Carbon
          Final Clarifier Carbon
1010
 906
 991
216
405
304

-------
                                                                                           389
                    Table 13. SOLVENT EXTRACTION EFFICIENCY DATA
Carbon
Identification
API
API
API
API
API
FC
NEW
Solvent
Chloroform
Benzene
Methylene Chloride
Methanol
Hexane
Chloroform
Chloroform
Grams
Carbon
49.3008
50.6917
54.5660
56.3179
55.3762
56.6447
44.6169
Grams Oil
Extracted
5.04
5.01
5.44
3.71
4.03
1.95
0.00
Gram Oil/
Gram Carbon
0.102
0.099
0.100
0.066
0.073
0.034
0.000
                    Table 14.  COMPOSITION BY TYPES OF ORGANICS

Organic Type
Saturates
Aroma tics
Polar Material

Oil from
API Carbon
11.1%
24.7
64.2
TOO
Oil from
FC Carbon
0.2%
1.3
98.5
100.0
                   Table 15.  COMPOUND TYPES INDICATED BY GC-MS
                                   API Carbon Extract
                                     FC Carbon Extract
Saturate Fraction
Aromatic Fraction
Polar
 C9H18
 C10H20
 C12H24
 Cg through Cy) n-paraffins

 Ethylbenzene
 Xylenes
 Co Benzenes
 C ]0 Benzenes
 Naphthalene
 Cii Benzenes
 C-\-\ Naphthalene
 C^ Naphthalene
 C13 Naphthalene

 Phenol
Hydroxy toluene
Ethylbenzene
Xylenes
€9  Benzenes
C10 Benzenes
C7H8  S
C7H10S
C9H8  S

-------
      6O
                     PULVERIZED


                     GRANULAR
                                                                        II
o
o-
              FIGURE - I
CARBON SOURCES

-------
ON
500
III
CO
CE
Q 0
< CD
DC
0 <
£ ° 100
U—
u. o
o

£ < 50
< 0
oc
° LJ
_l 0.
i
i

10


K.
-





200ml /min.
/+'
P
/
/
1
Q
/ WASTE - API SEPARATOR, EFFLUENT
6 CARBON-FILTRASORB 400
I

1
o
i
l
1






4.4g/m/ft.2





r
k
















.PUMP
=A

API SEPARATOR
-h
^
^
//
%•
a'4,
0 «
•"*•'
*V
".'*
1'*-


.sl5"ANTHRAFILT




^30" SAND (20 -30 Mesh)
Br
;>i
/«
•^
i
i^



WATER SAMPLE ^
HLl2" LIMESTONE (0.5"Dia.)
^
^
restful K"r:i Aocvi/nrn
           10           50    100           500
                 TOC- MILLIGRAMS PER LITER
                                                                                     TO RESERVOIR
         FIGURE  2 - ADSORPTION  ISOTHERM
                                                           FIG. 3-MULTI-MEDIA FILTER

-------
392
         COLUMN
      234
                                              t
                                    ELECTRONIC
                                      SAMPLER
                   FINAL
                   CLARIFIER
                       FINAL
                     CLARIFIER
                      SAMPLE
        DDDDD
         12345
         COLUMN
         EFFLUENT
                          IVIULTI- MEDIA
                             FILTER
                             SAMPLE
            FIG.4-MINI  COLUMN  SYSTEM
                       SAMPLE
                       POINT-!
                     SAMPLE
                     POINT-2
COOLING TOWER
SLOWDOWN
BOII FR Bl OWDOWN „
OILY PROCESS WATER
STRIPPER EFFLUENT
CONTAMINATED RUNOFF


                                                          TO HOLDING
                                                           POND
              API SEPARATOR
PASVEER DITCH
CLARIFIER
          FIGURE 5 - REFINERY WASTE WATER TREATMENT SYSTEM

-------
   FEED
   SAMPLE
                                                                 393
                                                              Pea
                                                            ^Gravel
         DUAL-MEDIA    FEED

            FILTER     PUMP
                                 "CALGON" ACTIVATED

                                 CARBON PILOT PLANT
            FIGURE 6 - ACTIVATED  CARBON  PILOT  PLANT

                           FLOW  DIAGRAM
a:
UJ
a:
UJ
a.
a:
o
o
o
   200-
   160-
   120 -
                                                 API SEPARATOR

                                                   EFFLUENT
40 -
                                                      CARBON TREATED

                                                        EFFLUENT
                                                   BIO-TREATED EFFLUENT
                                8      10

                            DAYS  INTO STUDY
                                                    14
                                                      BIO-CARBON EFFLUENT
                                                          16
                FIGURE 7  TREATMENT RESULTS, BOD

-------
394
                 280
                 240
              ct
              UJ

              3  200

              IT
              UJ
              Q.
              V)
              5
               I
              Q
              O
                 160
                 120
                80
                 40
                                                        API SEPARATOR
                                                           EFFLUENT
 BIO-TREATED
  EFFLUENT

 CARBON TREATED
  EFFLUENT

 BIO-CARBON
 TREATED
 EFFLUENT

	I	
                                  6     8    10

                                   DAYS INTO STUDY
                                                12
                                                       14
                                                            16
                          FIGURE 8- TREATMENT RESULTS,  COD
     70
     60
  ce
  UJ

  t  50
or
Ul
Q.

c/>   40


cc


H   30
  i
 o
 2  20
      10-
      0
                             I
                                           I
                                                             API SEPARATOR
                                                               EFFLUENT
                                                             BIO-TREATED
                                                             EFFLUENT
                                                           CARBON TREATED'
                                                            EFFLUENT


                                                           BIO-CARBON
                                                             EFFLUENT
                             6       8      10      12      14
                             DAYS  INTO STUDY
         FIGURE 9-TREATMENT RESULTS, TOC

-------
                                                                                 395
                                     TEMPERATURE, °F

      I35°F 260  380  490  610  720 825  895         I35°F 260 380  490  610  720  825  895
                                                          111
UJ
to
z
o
Q.
V)
Ul
cc
FIGURE 10- BENZENE EXTRACT OF
           API CARBON
    8
                             I  I  T
FIGURE II -  METHYLENE CHLORIDE EXTRACT
           OF API CARBON
                                                  \JV
             16  24  32  40  48   56  64      0    8    16  24   32   40  48   56  64

                                      TIME, min.

     FIGURE  12- METHANOL EXTRACT OF       FIGURE 13 - HEXANE EXTRACT OF

                 API  CARBON                              API CARBON
                                       TEMPEATURE, °F

        I35°F 260 380 490  610  720  825  895        I35°F  260 380 490  610  720 825  895
         ~g	fe	24  32   40^48   56   64       08   16  24   32   40   48  56   64

                                       TIME, min.


       FIGURE  14- CHLOROFORM EXTRACT OF     FIGURE 15 - CHLOROFORM EXTRACT OF

                   API  CARBON                             FC  CARBON

-------
396
        Temperature , °F
  tn
  z
  o
  a.
  to
  UJ
  tr
I35°F200 320 440 550 665 780 865
1
1 1

1 1 1 1 1 1 1 I 1 I 1 1 1 1
1 1 1 1 I t 1 1 I 1 1 1 I I
) 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64
                             TIME, min.


      FIGURE  16 - CHLOROFORM EXTRACT OF  NEW  ACTIVATED CARBON
          I35°F 200
320
Temperature °F

 440    550
665
780
865
   to
   z
   o
   a.
   CO
   UJ
   ae
            8   12  16  20  24  28  32  36  40  44  48  52  56 60  64

                             TIME, min.
       FIGURE  17 - STANDARD OIL MIXTURE

-------
                                                                                  397
Ul
CO
z
o
a.
V)
UJ
tr
                                       TEMPERATURE, °F

        I35°F26O 380  490  610  720  825  895      I35°F 260 380  490  610  720  825  895
                                                                   SENSITIVITY x 10
                                   TIME, min


FIGURE   18 -APIC EXTRACT SATURATE   FIGURE 19-FCC  EXTRACT SATURATE

             FRACTION                            FRACTION
                                  TEMPERATURE, °F
                                                                  SENSITIVITY x 10


                                                              I  I  I  I  I   I  I  I  I
           8
                                                     16   24  32  40   48   56  64
         16   24  32   40   48   56  64  0    8

                                   TIME, min.

 FIGURE 20-APIC EXTRACT AROMATIC     FIGURE 21 -FCC EXTRACT AROMATIC

             FRACTION                             FRACTION
     V)
     z
     o
     0.
     CO
                                          TEMPERATURE, °F


           I35°F 260  380  490 610   720  825  895    I35°F  260  380 490  610  720 825  895
            J	L
                     _L
                                       1	1—L
                                                                       i  I  I  I
        0    8    16   24  32  40   48  56  64   08    16   24   32   40  48  56   64

                                           TIME,  min.
        FIGURE 22- APIC EXTRACT POLAR       FIGURE 23- FCC EXTRACT POLAR

                     FRACTION                               FRACTION

-------
398

-------
   ACTIVATED CARBON  TREATMENT OF COMBINED  STORM AND  PROCESS WATERS

                                   M. A. Prosche
           Atlantic Richfield Company, Watson Refinery, Carson, California

   The Watson Refinery of the Atlantic  Richfield Oil Company is  located adjacent to the
Dominguez Channel in Los Angeles County.  This Channel is a non-navigable dredged
tidal estuary which is lined with rip-rap and used specifically for the discharge of refinery
and chemical plant waste  waters and for rain water  runoff.

   In 1968, the Los Angeles Regional Water Quality Control Board made  a study of the
Dominguez Channel and determined that petroleum  and  chemical  plant discharges were
causing a problem due to  the oxygen demand of their waste waters entering the Channel.
The Control Board, in accordance with  these findings,  issued a resolution in February,
1968, which limited the total chemical  oxygen demand  (COD) from all industrial  discharges
into the Channel.  These  discharges also included any contaminated rain  water runoff.
The resolution was  to be complied with  by February, 1971 .

   As defined by  the resolution, the Watson Refinery was limited  to 1330 pounds per day
of COD in its discharge water to the Channel.  Meeting this requirement meant reducing
the COD  in its discharge  waters by 95 per cent.

   Fortunately, the Watson Refinery as  a taxpayer of Los Angeles  County was able to make
arrangements with the Los Angeles County Sewer District to  have  its process waste water
handled in the County's primary treatment unit.  However, due to limitations in the County
unit,  the  County  was unable  to handle rain water runoff.  This presented  a problem for
the Watson Refinery due to the fact that the rain water collection facilities were inter-
connected with the process waste water  collection system.  Therefore, during periods of
rainfall, the process waste water and  rain water mixture could not be sent to the sewer
district facilities, nor could it be sent to the Dominguez Channel  due  to high COD content
of the process waste water.

   To solve  this problem a system was needed which would treat all the process water
plus rain water during the rainy season for the removal of COD and allow its discharge to
the Channel.  A system was  needed which could be  started up easily when rain fell and
then shut down when no longer required.  The system selected was impounding of rain plus
process water during the storm followed  by activated carbon treatment to  adsorb the COD
material.

ADSORPTION SECTION  DESIGN AND DESCRIPTION

   Since it was planned to operate the plant only during the rainy season, it was felt
there was no need to design  for continuous carbon regeneration and changeout of the beds.
Therefore, the design was based on use of only the carbon beds during the rainy season,
and then regeneration of all the beds during the dry summer months in preparation for the
next rainy season.  The significant design criteria for the plant and impounding basin were
as follows:

                                       399

-------
400

                                    Design Criteria
                   Flow Rate                            100,000 BPD
                                                          3,000 GPM
                   Inlet COD                           250 ppm
                   Outlet COD                         37 ppm
                   Impounding Basin                     1 .2 million barrels capacity

   Following the accumulation of 0.1 inches of rainfall, regulations call for diverting
process water from the sewer system.  This  is accomplished automatically by means of a
solenoid operated valve  actuated by a rain gauge.  Process water plus commingled rain
water is then switched to the 1 .2 million barrel impounding basin.  Depending upon the
specific situation, the carbon plant can be started up immediately for discharge to the
Channel or it can be started up later following the rain.  A flow diagram of the adsorption
section of the plant is  shown in Figure 1 .

   Impounded water is delivered to the plant through a  14" line from the basin. The water
is then  delivered to a distribution trough where it is distributed to  12 adsorber cells by
adjustable slide gates.   Each of the twelve cells is 12' by 12'  square and 26"  feet deep.
Each cell originally contained 13' of carbon having a dry weight of about 50,000 pounds.

   The  water passes down through the carbon bed where it  collects in the underdrain system.
Supporting the  carbon  is a one-foot layer of gravel  on  top  of a Leopold tile underdrain
system.  The treated water then flows through 6" lines  from each cell to a 24"  collection
header  leading to the effluent  retention sump.  Each 6" discharge  line has a sample  point
and  an  air-operated pinch valve which can be shut during  backwashing.

   If necessary to further control COD, a chlorine-water solution may be injected into the
incoming treated water stream.  Approximately 15 minutes retention  time is allowed for
chlorine contact in this sump.  From the sump the water flows by gravity to the Channel.

   Each carbon bed must be backwashed whenever it will not  pass  its share of water  flow
due  to buildup  of solids on top of the carbon.  This is indicated by the rise in the level of
the water in each carbon cell.  When the level rises to the height of the backwash  troughs,
the flow to and from the bed is stopped, and treated water from the backwash sump  is pumped
up through the  bed  to expand it and flush out accumulated solids.  The turbid water  overflows
into the backwash troughs to the backwash  effluent sump where it is pumped back to the
reservoir for settling and retreating.

EPA DEMONSTRATION GRANT

   Following completion of the plant, a demonstration grant was received sponsored by the
Water Quality  Research  Division of Applied Science and Technology of  the Environmental
Protection Agency.  The specific objectives of the demonstration project were  as follows:
   1 .  Determine feasibility of activated carbon as a treatment system for storm runoff and
       refinery process waters.
   2.  Evaluate performance of the system.
   3.  Determine operating  costs.
   4.  Assess reliability  of the system.

-------
Length of
Run, Hours
44
48
38
18
95
22
" Table 1
Feed
Rate, GPM
3000
3000-2000
2000
2000
1000
2000
Average COD
Feed
326
360
374
310
237
147

Effluent
43
48
86
67
100
93
                                                                                 401
The report relative to this project has now been published by the EPA.

ADSORPTION SECTION OPERATION AND PERFORMANCE

   The carbon plant was first placed in operation in May, 1971 for some preliminary test
work prior to the rainy season.  At that time the test water was synthesized using process
water diluted with service water.  Operation with rain water and process water was not
required until  later that year in December.  At that time sufficient rain fell to require
placing the unit in full operation at the design rate of 3,000 gpm total to all twelve cells.
Subsequent to that a total of 6 runs were made processing  impounded water  from the inter-
mittent rain storms that followed.  The performance data for all of these runs are shown in
Table 1.
                      Typical  Performance Data First Rainy Season

  Run
Number
~~1
  2
  3
  4
  5
  6

   During the initial runs the COD of the feed was higher than design and the plant was
unable to produce the desired removal.  Therefore, the  feed rate was reduced to 2,000
gpm.   Even  at this lower rate the effluent COD content was greater than the regulations
allowed and feed rate was again reduced to 1,000 gpm.  By the end  of the  season the
reservoir COD had dropped  to 147 ppm due to rain dilution and the  rate was increased to
2000  GPM.  However, performance continued poor even at this low  concentration.

   A  good evaluation of the performance of the plant during this first season of operation
is difficult due to the significant variations in feed COD and the necessity  of having to
reduce the feed rate from 3000 to 1000 gpm.   However, COD data for each cell was
recorded during the course of the run  and some generalizations are possible.  At the end
of the season the COD loading in each cell varied from 0.2 pounds  of COD per pound of
carbon to a high of 0.3 pounds of COD per pound of carbon.  During the season a  total
of 52,000,000 gallons of water were  processed.   The  average feed COD concentration
was 377 ppm while the effluent averaged 67 ppm. The  carbon loading averaged 0.23
pounds COD per pound of carbon.  It also appeared that the potential maximum carbon
loading at a constant effluent COD concentration was quite sensitive to the COD of the
feed.

   Other data collected  during the first season's operation included measurement of the
carbon's adsorptive ability at various depths in the bed  after a period of operation. One of
the cells was taken out of service after 208 hours of operation and the carbon was sampled
at various depths to test its  relative adsorptive efficiency as compared to virgin carbon.
The data are shown in Figure 2.  The  data show that the top two feet of carbon were
completely exhausted while the last five feet were still  65 per cent of virgin adsorptive

-------
 402

ability.

   Prior to the second rainy season, we made several significant changes in operation.  In
order to increase the adsorptive capacity of the unit,  15,000 pounds of carbon were added
to the existing 50,000 pounds in each of the cells.  This increased the bed depth from  13 feet
to 17 feet.  It was also decided to recycle effluent  water back to the feed  in order to
control the feed COD to a concentration of not greater than about 250 ppm.  This change in
operation was made after noting during the first rainy season of operation that the effluent
COD level varied directly with the feed COD concentration and the specification COD
quantity  to the  channel  would be exceeded during periods of high feed COD.  It was also
felt that  controlled feed COD would allow a higher ultimate COD loading of the carbon.
One other modification  made was the aeration of the effluent  recycled back to the feed to
prevent growth  of anaerobic bacteria in the beds.

   For the second rainy  season of operation, we  elected to operate part of  the plant on a
continuous basis, regenerating each bed as it became  spent, and returning  the regenerated
carbon for use again. To accomplish this  five of the cells were placed in continuous
staggered operation along with the regeneration  furnace. This was the largest number  of
cells  operating  in a progression mode that could  be  accommodated by the capacity of the
regeneration furnace.  The remaining seven cells were used for once-through processing as
was done  during the first rainy season.  In all  cases the flow to each cell was  held constant
at 250 gpm which included the waste water feed from  the reservoir plus the recycle dilution.

   Typical performance  data for one  of the five cells operated in staggered mode are shown
in Figure  3.  This figure shows the relationship between feed and effluent COD and the COD
loading on the carbon.  During initial operation of  this cell, it was elected to discontinue
service and  regenerate after the effluent from  the cell  reached a COD concentration of
about 50 ppm.  This corresponded to  a carbon  loading of 0.20 pounds of COD per pound of
carbon and a run length of 22 days.

   The carbon was regenerated and returned to the same cell.   In the second run the cell
was allowed to  operate until the effluent reached a COD concentration of  about  100 ppm.
This corresponded to a carbon loading of 0.40 pounds  COD per pound of carbon and a  run
length of 38 days.

   The performance of the seven cells that were  operated on a once through basis with  no
regeneration is  shown in  Figure 4. Effluent COD concentration was  allowed to reach 50
ppm before all the cells were shut down for  later regeneration.  As shown in the figure, the
corresponding carbon loading attained in all of these cells  was 0.30  pounds COD per pound
of carbon and the run length was approximately 30 days.

   The evaluation of performance of the carbon plant  for this second rainy  season was again
based on  the combined operation of all the cells.  During the  season a total of 102,000,000
gallons of water were processed.  The average diluted  feed COD concentration was 233 ppm
while the effluent averaged 48 ppm.  The carbon loading averaged 0.26 pounds COD per
pound of carbon.

-------
                                                                                  403
   A summary of the performance of the plant during the two rainy seasons of operation is
shown  in Table 2.
                                    Adsorption Data
                                       Table 2
Commercial           Feed COD,  PPM     Effluent COD, PPM   Carbon Loading Lb.  COD/
Operation               Average               Average         Lb. Carbon Average
First Rains                 377                   67                   OS
Second Rains               233                   48                   0.26
As noted previously, problems with varying feed COD and algae growth were experienced
during some of the runs. These problems undoubtedly limited the performance to poorer
levels  than could have been attained under ideal steady state conditions.   However, these
are the real problems that exist when treating refinery waste waters and we therefore feel
the carbon loadings attained are indicative of the true performance of the commercial plant.

CARBON HANDLING AND REGENERATION

   This system is shown  in Figure 5. The spent carbon is removed as a water slurry through
a carbon removal trough and nozzle even with the top of the gravel layer. The slurry flows
by gravity and with the aid of water jets in the carbon removal  trough to the spent carbon
transfer pump sump.  From  here a rubber  lined pump transports the carbon slurry into the
spent carbon tank.  The  tank has capacity  to contain the contents of one cell.  From this
tank the carbon is educted with high pressure water to the dewatering screw above the
regeneration furnace.  A timer valve on  the spent carbon line feeding the eductor opens
and closes at timed intervals to allow close control of the carbon flow to the regeneration
furnace.  The dewatering screw separates most of the water from the spent carbon slurry
and the drained water is returned to the reservoir.  The carbon discharging into the furnace
contains approximately  50  per cent by weight of water.

   The regeneration  furnace is a 56" I.D.  multiple hearth type  with a total of six hearths.
It is gas fired and internal  temperatures are controlled  to about  1600  F.  A center shaft with
rabble arms  moves the carbon across each hearth and downward  through the furnace.  An
elaborate flue gas quench and scrubber system is installed to meet local Air Pollution
Control District requirements.  The furnace is designed to regenerate  8,000 pounds of carbon
per day or one bed in about 6 to 7 days.

   The regenerated carbon drops into a quench tank just below the furnace from where it is
educted into the regenerated carbon tank.   From this tank the carbon  is then educted and
transported through a hose back to the same cell where it originated.

   The carbon handling system has worked  very well.   The original 1" cast iron eductors
were too small  and soon  lost capacity due to erosion.   Larger stainless steel eductors and
a reduction of motive water pressure from 110 psig to 50 psig have remedied this.  Also
originally we had some problem with displaced  support gravel from the bottom of the cells
plugging up the eductors.  This was remedied by constructing a  gravel screen  in the top
of the spent carbon tank to remove  the gravel.  The time required to transfer the carbon
out of the cell  to the spent carbon tank is about 14 hours while  transfer back to the cell
from  the regenerated carbon tank takes about 7 hours.

-------
404


    Operation of the regeneration  furnace has been excellent.  The major effort relative to
 this operation has been to develop a reliable method for determination of the quality of the
 regenerated carbon.  After operational guidelines were developed,  the furnace consistently
 regenerated the spent carbon to near virgin activity.

    Our experience indicates that  about a 5 per cent loss is incurred during carbon handling
 and the  regeneration.  We  feel the majority of this  is due  to attrition  of the carbon and
 resulting production of fines.  These fines are backwashed out of the cells prior to being
 placed in operation and end up back in the reservoir where they settle out.

 OPERATING COSTS

    Actual operating costs for the second year of operation are shown in Table 3.
                                      Cost Data
                            December, 1972 - March,  1973
                                       Table 3
                                     Cents Per                       Cents Per
                                  Thousand  Gallons                Pound of COD
        Cost Areas                 of Water  Treated                   Removed
    Utilities                              9                              4
    Repair Labor                          3                              1
    Operating Labor                      15                              7
    Carbon                               1 1                              5
    Miscellaneous                        2                              1
    Total                                 40~                           f8"
 Updating these costs to 1976 levels would  give about 56 cents per thousand gallons of water
 treated, or about 25 cents per pound of COD removed.

 CONCLUSIONS

    The plant has proved to be easy to maintain, easy and quick to start up, simple to shut
 down and leave in a standby stage,  and very  reliable in its performance.

    For  the unique intermittent type operation required at the Watson Refinery, we feel the
 Carbon Adsorption Process has been  succes