&EPA
            United States
            Environmental Protection
            Agency
            Robert S. Kerr
            Environmental Research
            Laboratory
            Ada OK 74820
EPA-600/2-78-058
March 1978
            Research and Development
Proceedings of the
Second Open Forum on
Management of
Petroleum Refinery
Waste water
            Environmental Protection
            Technology Series

-------
                                              EPA-600/2-78-058
                                              March 1978
      PROCEEDINGS OF THE SECOND OPEN FORUM
                       ON
   MANAGEMENT OF PETROLEUM REFINERY WASTEWATER
                  Presented by

    The U.S. Environmental Protection Agency

        The American Petroleum Institute

   The National Petroleum Refiners Association

             The University of Tulsa
                 Project Officer

                 Fred M. Pfeffer
            Source Management Branch
Robert S. Kerr Environmental Research Laboratory
                  Ada, Oklahoma
           Editor and Project Director
               Francis S.  Manning
               University of Tulsa
                 Tulsa, Oklahoma
ROBERT S.  KERR ENVIRONMENTAL RESEARCH LABORATORY
       OFFICE OF RESEARCH AND DEVELOPMENT
      U.S. ENVIRONMENTAL PROTECTION AGENCY
               ADA, OKLAHOMA 74820

-------
                                  DISCLAIMER


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

-------
                                   CONTENTS

Disclaimer                                                                  ii
Acknowledgements                                                            vi
Welcome Address                                                               1
Keynote Address                                                               3
SESSION I  Regulatory and Research Direction in U.S. EPA                       12
           Lamar Miller "Toxic Strategy Overview"                             13
           Marvin L. Wood "EPA's Research in the Refining Category"            20
SESSION II Panel Discussion: Research Supported in the Private Sector            35
           Robert T. Denbo "Discussion of Research Related to Petroleum         36
                           Refinery Wastewater Sponsored by the API
                           Committee on Refinery Environmental Control"
           Kent G. Drummond "Water Quality Committee of API"               43
           Francis S. Manning "Overview of Research on Petroleum Refinery      48
                              Wastewaters at U.S. Universities"
           David C. Bomberger "Overview of Research on Petroleum Refining     53
                              Wastewaters at Independent Contract Research
                              Organizations"
           Judith G. Thatcher "Current API Studies of Residuals in Refinery       56
                              Effluents"
SESSION III Open Question and Answer Session - Individual Problems in           64
            Management of Refinery Wastewater
Banquet Proceedings and Address                                               80
SESSION IV Origin and Interpretation                                          85
           Ridgway  M. Hall, Jr.  "Regulation of Problem Pollutants Under the     87
                                 Federal Water Pollution Control Decree"
           W. M. Shackelford "Evolution of the Priority Pollutant List From      103
                              the Consent Decree"
           R. W. Dellinger "Incorporation of the Priority Pollutants into        112
                           Petroleum Refining"
           Leon Myers  "Generating Problem Pollutants Data for the EGD        120
                       Document:  Refining"
                                       m

-------
           A. Korim Ahmed "Considerations for Defining Substances
                            Hazardous to the Environment"
SESSION V Problems                                                        155
           Dwight Bellinger "EPA's Analytical  Development Program  for        157
                            Problem Pollutants"
           Fred T. Weiss "Fates, Effects and Transport Mechanism of            1°6
                         Pollutants in the Aquatic Environment"
           Donald I. Mount "Measuring Aquatic Impact of Toxic Con-          185
                            taminants"
           Davis L. Ford "Overview of Advanced Treatment Systems"          193
SESSION VI Future Considerations in Biotreatment                             215
           J. F. Grutsch "Design and Operations: Bases for an Activated      217
                         Sludge Route to BAT (1983) Water Quality Goals"
           Robert W. Griffin "Consideration in Reuse  of Refinery Waste-       281
                            water"
           Milton R. Beychok "State-of-the-Art in Sour Water Stripping"       293
           Irv Kornfeld "Refinery Discharges to a Large Municipal              304
                       Sewerage System"
SESSION VII  Powdered Activated Carbon                                      323
           Francis L. Robertaccio "Combined Powdered Activated Carbon       325
                                Treatment-Biological Treatment: Theory
                                and Results"
           Colin G. Grieves "Powdered Activated Carbon Enhancement of      344
                            Activated  Sludge  for BAT Refinery Wastewater
                            Treatment"
           Paschal B. Dejohn "Case Histories:  Application of PAC in          369
                            Treating Petroleum Refinery Wastes"
           James F. Dehnert "Case History: Use of PAC With a  Biodisc-        389
                            Filtration Process"
SESSION VIII Add-On Granular Activated Carbon                             401
           Fred M. Pfeffer "Pilot-Scale Effect on Specific Organics            403
                          Reduction and Common Wastewater Parameters"
           R. H. Zanitsch "Granular Carbon  Reactivation-State of the  Art"     419
           L. W.  Crame "Activated Sludge Enhancement: A Viable            440
                        Alternative to Tertiary Carbon Adsorption?"
SESSION IX Costs/Benefits                                                  470
           W. Wesley Eckenfelder "Overview of Costs/Benefits"               472
           Lial  Tischler "Treatment Cost-Effectiveness as a Function of         481
                        Effluent Quality"
                                      IV

-------
             Carl E. Adams, Jr. "The Economics of Managing Refinery       503
                                Sludges"

             Melville Gray "Compliance Monitoring Costs for the           526
                           Priority Pollutants"

             Leo J.  Duffy "Analytical Costs in the Problem Pollutants"    531


List of Participants                                                      542

-------
                            ACKNOWLEDGMENTS

      The Second Open Forum on Management of Petroleum Refinery Wastewater was
held at the Sheraton Inn-Skyline East in Tulsa, Oklahoma.  This four-day (June 6-9, 1977)
symposium was sponsored by the U.S. EPA in the form of Grant No. R804968-01-0
to the  University of Tulsa.  Fred M. Pfeffer and Francis S. Manning served as Project
Officer and Project Director respectively.   In addition, the American Petroleum
Institute, the National Petroleum Refiners Association and the University of Tulsa con-
tributed matching funds.

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

      Success of any project is aided immeasurably by support "from the top. "  Such
support was provided enthusiastically by Arne E. Gubrud,  API; HerbBruch,  NPRA; Bill
Galegar, EPA,  Ada, Oklahoma; 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 to Katie Whisenhunt, Shirely  Clymer,  Cathy Whisenhunt,
and Cathy Clymer for managing the registration and typing the proceedings;  Ed and Dan
Andrews for tape recording the proceedings; Nelda Whipple for transcribing  the tapes and
typing  the proceedings; and John Byeseda for photographing the speakers and session
chairman.
                                       vi

-------
                               WELCOMING REMARKS

                                  William C. Galegar
                 Director,  R.S. Kerr Environmental Research Laboratory
                     U.S.  Environmental Protection Agency

Honored Guests, Ladies and  Gentlemen...

        As we assemble here for the opening of the Second  Open Forum on "Manage-
ment of Petroleum Refinery Wastewaters," I am reminded of a saying attributed to the
French author,  Alexandre  Dumas.  It is ... "Nothing Succeeds like Success!"

         In a sense, we will be challenging  the truth of that statement during  the
next few days.  As those who participated in the first Open Forum just over a year
ago will recall, it was a very successful gathering.  It clearly established the  advantages
of the exchange of technical information among the industry, academic, all levels of
government agencies and the interested public.

         The question is, what do we do  for  an encore? In welcoming you here, I
want to sketch for you some  of the efforts made by the hard-working committees putting
the conference  together to improve on the earlier product.

         We have maintained the firm base on which  the first meeting was built  —
the co-sponsorship of the Environmental  Protection Agency, the American  Petroleum
Institute,  the National Petroleum Refiners Association and the University of Tulsa.

         We have again brought together a group of outstanding speakers, each with
specialized knowledge,.who will maintain the high level of technical excellence and
the presentations last year.

         We; have directed our emphasis  to a more specific objective, that of
"Environmental Conservation in Petroleum Refinery Wastewater Treatment."

         Finally, we have solicited the views of the Natural  Resources Defense
Council and the Environmental Defense Fund on the scientific rationale behind the
differences which led to the  courts and the consent decree.  This, we believe, adds
a new dimension to our forum.

        Along with expressing the formal welcome to you.. .many attending for the
second time.. .allow me to include the hope that  through  your active participation

                                         1

-------
in the sessions we will prove beyond a doubt that Mr. Dumas was correct...
that we will succeed in taking another big step toward our mutual goal of
improving enviornmental quality.
BIOGRAPHY
William C. Galegar
      William C. Galegar holds a B.S. in
Chemical Engineering from Oklahoma State
University and an M.S. in Chemical Engine-
ering from the University of Oklahoma.
His career has included 11 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

                      "WHAT DOES JULY 1, 1977, MEAN NOW?"

                              Joe G. Moore, Jr.
              Head, Graduate Program in Environmental Sciences
                      The University of Texas at Dallas

     It is good to be with you.  I spoke to you in your prior conference as
the Program Director of the National Commission on Water Quality, attempting
to predict whether or not the Commission would recommend changes in Public
Law 92-500.  The Commission did make such recommendations.  A year has
elapsed since those recommendations were delivered to the Congress, and we
now approach one of the critical deadlines in that statute.  So that I have
some knowledge of my audience, how many of you are engineers and consultants
to oil companies?  I see, mostly that.  Alright, how many of you are lawyers?
Any lawyers?  Interesting.  How many of you are oil company employees as
opposed to consultants involved in water pollution control?  Okay.  How many
of you will claim to be representatives of the Environmental Protection
Agency besides Bill?

     I would like to focus primarily upon the question, "What Does July 1,
1977 Mean Now?", now that it is the first week in June.  For some reason,
there is a general feeling that that date is no longer significant since the
deadline is almost gone.  How many effluent limitations for best practical
control technology currently available are final four and one-half years
after the passage of Public Law 92-500?  How many permits required to have
been issued under P.L. 92-500 are now coming to the end of their five year
terms?  What has been the effect of the extensive litigation with regard to
the concepts involved in P.L. 92-500, particularly as related to July 1,
1977, as the answer to that question might bear upon the requirements of
July 1, 1983?  You see, this statute is reaching the mid-point in its first
ten year plan, and the question is, "Where are we, and where are we going?"

     I would like to review first of all the fundamental concepts that
92-500 requires effluent limitations as the basis for permits under the
NPDES (the National Pollution Discharge Elimination System).  Effluent limi-
tations are the basis for permits for every single point source discharge in
the nation.  The backbone of the Act is really in sections 301 and 304 which
prescribe the promulgation by the Environmental Protection Agency of "best
practicable control technology currently available" and "best available
technology economically achievable,"  the effluent limitations upon which
permits are supposed to be based.  The first U.S. Supreme Court decision,
du Pont I vs. Train,has been decided with the opinion that the approach of
the Environmental Protection Agency in developing single-number effluent

-------
 limitations is valid.  The court also bridged the gap between the different
 appeals procedure for sections 301 and 304 in such a way as to" validate what
 EPA had thus far done with their effluent limitations approach.   In addition,
 the court held that it is appropriate to issue single-number effluent limi-
 •tations as opposed to a range of numbers as had been argued is actually re-
 quired by section 304.  The effect of this ruling is to validate the ap-
 proach used by the EPA in the promulgation of effluent limitations.  It was
 also held  appropriate that  the agency  issue not only  the guidelines under
 section 304  but  the  effluent  limitations under section 301 simultaneously.
 In  the  final analysis, EPA  has won one of  the most crucial issues to the
 application  of P.L.  92-500  even  though their record of litigation in most
 cases has  not been that outstanding.   The  other sections that do require
 effluent limitations  or provide  some basis for permitting conditions are
 section 306  for  new  source  performance standards, section 307 for toxic and
 pretreatment standards, and a section  that has been often overlooked, sec-
 tion 302 which requires that  the Administrator issue effluent limitations
 for water  quality standards achievement for those waters which have been
 characterized by the  EPA as water quality  limited.  To my knowledge, the
 Administrator of EPA  has issued  no effluent limitations for the achievement
 of  water quality standards  described in section 303 with the effluent limi-
 tations detailed in section 302.  The  reason I mention that particular pro-
 cedure  is  because, in my view, that may come to be one of the more critical
 provisions of P.L. 92-500 as  we  move past July 1, 1977.

     Let me  digress for just  a moment, to mention the Congressional issues,
 or  the  legislative issues,  with which  the Congress has contended since the
 National Commission on Water  Quality made its recommendations.  I shall men-
 tion the ones that I  regard to be crucial in judging what Congress may do.
 First of all,  the construction grants program is running out of money; that
 is, the construction  grants program for municipal wastewater treatment
 plants.  There are several  states which will have, before this calendar year
 is  ended,  committed or obligated all of their shares of the $18 billion
 originally authorized in P.L. 92-500.  That means that some action either
 must be taken by the Congress to provide additional funding for construction
 grants  under  P.L. 92-500, or  the publically owned treatment works construc-
 tion grant program will grind to a halt in some states.  One of the crucial
 states  in  which the construction grant funds will be exhausted is Maine.
 Section  404, which allows the Corps of Engineers to issue permits for dis-
 posal of dredged materials  into navigable waters and adjacent wetlands con-
 tinues  to  be  controversial in the Congress, with a wide divergence between
 the viewpoints of the Senate  Subcommittee of the Environment and Public
 Works Committee and the House Public Works Committee; that issue remains
 unresolved.   On the House side, the Members have expressed continuing inter-
 est in decentralization of  the program, or what has come to be known as
 certification to the states for various activities required under the Act.
 I think  the House will continue to be  interested in that particular issue.
 There was  an  effort to extend the July 1, 1977, deadline so that is an issue
 or was an  issue in the legislation that has been considered twice by the
 Congress in the last 15 months.  There is also the continuing question about
what to do with the 1983 deadline and whether or not that deadline should be
 extended.  These represent what  to me  are  the major issues; probably the two
 critical ones are the question of any  extension of deadlines and the money

-------
for construction grants.  I regard section 404 as a peripheral issue  to  the
general thrust of the water pollution control program; nevertheless,  it  does
provide a basis for disagreement between the Houses.

     When Congress adjourned in the fall of 1976 there was pending legisla-
tion to amend the pollution control act.  It was in conference committee
between the two Houses, and there was rather a stormy session at which there
was substantial disagreement between Congressman Jim Wright, then the ranking
majority member on House Public Works, and Senator Muskie, Chairman of the
Subcommittee on Environmental pollution.  At that time, apparently the con-
ferees adjourned and decided not to meet again, or at least did not have an
opportunity or occasion to meet again.  The bill was stalled, in any event,
by the filibuster then occurring on the Senate floor over the amendments to
the Clean Air Act of 1970.  Thus the congressional session ended with major
disagreements over the general thrust of both air and water pollution control
legislation.

     With the change in Administrations and an emphasis on the need to stim-
ulate the economy, the Senate Public Works Committee again, through a device
designed to provide money just for construction grants, precipitated a par-
limentary issue in such a way as to frustrate the legislation amending
92-500.  The sequence of events is complicated because the water pollution
control program became intermingled with the question of economic recovery
in such a way that it was sometimes difficult to determine just exactly what
was happening.  Senator Muskie succeeded in having added to the Administra-
tion's jobs bill an appropriation to continue the construction grants pro-
gram for a year.  Since this issue represents one of the major questions in
the amendment to P.L. 92-500, had that appropriation remained in the jobs
bill, then some of the leverage which the House felt was essential in the
consideration of other amendments would have been lost to the House.  When
the jobs bill was returned to the House where it had originated, the appoint-
ment of conferees from the House was delayed.  In other words, the jobs bill
which had originated in the House had gone to the Senate where this amend-
ment had been added and then the bill had been returned to the House for the
appointment of conferees.  Had the House gone immediately to conference,
they would have been confronted with the situation in which all they could
attempt to do would be to eliminate the $4 billion authorized and appro-
priated for construction grants for publicly owned treatment works.  Faced
with that alternative, the House decided that they would proceed in such a
way as to precipitate amendments to 92-500 into the conference on the jobs
bill; thus, the appointment of conferees was delayed until the House passed
a bill amending 92-500 and covering some of the points I mentioned a moment
ago, as well as others, that are in controversy between the two Houses.  The
House then appointed conferees with both the jobs bill and the House amend-
ment of 92-500 in the same conference.  So the conference began.  Initially,
the House proposed that the question of construction grants authorization be
separated from the jobs bill and the jobs bill be agreed upon and sent back
to both Houses for passage and thence to the President.  The Senate declined.
After several meetings Senator Muskie, apparently concluding that there was
no chance of the construction grants appropriation surviving in the jobs
bill, finally agreed to the original House offer.  The House dropped  consid-
eration of any amendments to the jobs bill that would relate to 92-500,  and

-------
 the Senate accepted the deletion of  the appropriation for construction
 grants.   Where that leaves the water pollution issues is  that the House-
 passed bill is now theoretically pending in the Senate, but  there is a
 question as to whether the Senate will actually consider  that piece of
 legislation.

     Senator Muskie called hearings beginning this week in several cities
 across the U.S. and extending them throughout the month of June, culminating
 in  a series of  hearings  in Washington late in June on the provisions of  92-
 500.  Unfortunately, the hearing cities do not afford real opportunity for
 participation.  For example, one in Portland, Maine; one in Greeley, Colo-
 rado;  one in Duluth and  another in Alexandria, Minnesota;  one in New Orleans,
 and one  that was initially scheduled in Berkeley but has now been resched-
 uled,  I  understand, for  San Francisco, with one other hearing at a place
 listed in the ammouncement as "undetermined," Iowa.  Whether these hearings
 will produce a  great deal of testimony remains to be seen.  There will then
 be  several days of hearings in Washington.

     I have reviewed that legislative history, because I want to return  to
 the fundamental question, "where are we with July 1, 1977?"  For soroe^reason
 everyone seems  to think  that we're so near the deadline that it doesn't  make
 any difference.  One of  the reasons I asked whether there were lawyers here
 is  because I think the lawyers are going to be very busy in the month of
 July.  And if your clients or your employers are not in compliance with  per-
 mit conditions  on July 1, 1977, I would suggest that you call your lawyer's
 attention to the fact that they should be prepared to respond to litigation
 shortly  after July first.  I'm in a minority in what I think is likely to
 happen,  I know.  It is very difficult to predict; quite frankly the closer
 we  get to this  date, the more difficult it may be to predict, exactly what
 will happen.

     Just let me mention some choices that are available to the Adminis-
 trator of the EPA.  When I talked with Bill Galegar about this appear-
 ance,  I  told him that the title of my remarks was going to be, "What Now,
 Brown  Cow?" Because the question is, "What now, Mr. Administrator of EPA?"
 Of  course,  one  alternative is to do nothing and just let things proceed  as
 they have been  proceeding, with no change.  Another possibility is to
 selectively institute enforcement proceedings in the courts against indus-
 trial dischargers.  Pick out the horrible examples	and the chances are,
 if  this  is  the  course that EPA has under consideration, undoubtedly U.S.
 Steel will be at the top of the list.  There may be some oil companies close
 behind.   What I'm saying is that the Administrator could selectively choose
 against  whom the law would be enforced and, in this alternative, institute
 no  actions  against publicly owned treatment works, that is, municipalities.
 Everyone sort of shrugs and says,  well, what can you do about the cities?
 Would it help to put Abraham Beame in jail?   It might help Bella Abzug,
 but I don't know that would help anyone else.  Another alternative would be
 to  selectively enforce,  that is, pick some flagrant Industrial violators
 or  recalcitrants and institute suit against them; do the same for some
 selected municipalities, hopefully municipalities with Republican mayors or
 Republican administrations, so that you could get even with those actions
which were instituted under the Republican Administration against some of

-------
 the  larger cities  in the country.   Here again, this is a selective alter-
 native;  the Administrator picks those against whom he would enforce.  An-
 other alternative  would be to continue the procedure of ECSL letters.  If
 you  haven't seen an ECSL letter,  or don't know what this means, it is, in
 effect,  an administrative device for selective extension of the July 1, 1977
 deadline - - an administrative device, you will notice.  This procedure was
roundly criticized publicly by  Senator Muskie  and staff of  the  Senate  envi-
ronmental Pollution  Subcommittee as being  illegal under 92-500  and might now
be characterized as  a Republican dodge of  the  intended  effect of the statute.
Quite frankly, I think  it's illegal.   I do not believe  that it  has support
in the statute, and, therefore, I think those  who go this route may find
difficulty  if the courts, as one court already has, hold it an  invalid  appli-
cation of 92-500.  By the way,  EPA may still be thinking that the ECSL  letter
offers a means of doing something about the July 1, 1977 deadline.

     Of course the real alternative that, in my view, is mandated by the
statute is  the filing of suits  against every point source discharger in the
nation that  is now,  or  will be  on July 1st, in violation of the law.  Regard-
less of whether it is a practical solution or  whether it offers any real
chance in terms of being seriously considered  administratively, in my view,
this is what the law requires.  The objective  would be  to secure as rapidly
as possible  consent  decrees in  the courts spelling out  the  conditions under
which dischargers would comply with the requirements for July 1, 1977.

     You should be aware that the EPA  is, however, approaching  industrial
dischargers with an  unusual question as a basis for determining penalties,
"how much money have you saved by not  installing 'best  practicable control
technology  currently available'," or,  to put it differently, "what would you
have had to  spend had you timely complied with the requirements by July 1,
1977?"  This concept of the appropriate penalty is one  that was embodied in
the proposed revisions  to the Clean Air Act of last year and will undoubt-
edly be, in my view, proposed by the Administration in  any  recommended
changes to 92-500.    The argument is that collecting a penalty equivalent to
what the discharger would have had to  spend will somehow restore equity be-
tween those who have complied and those who failed to comply.   If you are
now involved in litigation, you may as well tell your lawyers to be pre-
pared - - and you might as well begin  to try to figure what the cost would
have been if you had complied with the statute - - because  I think this will
be an issue, at least in the discussions of settlement  of pending litigation.

     The last alternative that I have  identified is the Administrator could
ask for Congressional relief.  Frankly, this appears to be very unlikely at
this point in time.  The controversial issues  in 92-500 are interlocked in
such a way that it  would be very difficult to  solve any  single  one of them,
as the Senate learned by trying to put construction grants money in the jobs
bill; it is very difficult to take any single  one of them and isolate it for
statutory amendment.
                 ?'-.• ;      :
     How would yoii -like to be faced with the alternatives I've  suggested?
For some reason, thei impression has grown that if we can just get past July 1,
1977, it will go away.  I don't think  so.

-------
     There  are  a  series  of statutory provisions  that  are  interlocked  in  such
a way  that  you  cannot  plan with  very much certainty what  may happen in an
individual  case.   I'm  just going to  read a sentence or two  from the series
of  these.   Beginning with section 301 -  - and sometimes we  overlook these
kinds  of provisions because they're  stated right in the front  of the  section
and we don't  get  around  to them  because  we focus upon the substance of a
provision 	 section 301 starts out by saying,  "Except as in compliance with
this section and sections 302, 306, 307, 318, 402, and 404 of this Act, the
discharge of any pollutant by any person shall be unlawful."  Now, the sec-
tions cited are section 301, the one under which the Administrator prescribes
general effluent limitations for 1977 and 1983;  302 is the section that re-
quires effluent limitations for water quality limited waters; 306 is the
section that prescribes new source performance standards; 307 requires toxic
and pretreatment standards; 318 contains a special provision on  aquiculture;
402 is the  permit section; and 404 governs the disposal of dredged spoil to
which  I referred earlier. I'm going to read the  first sentence of 402,
"Except as  provided in sections 318 and 404 of this Act,  the Administrator
may, after  opportunity for public hearing, issue a permit for the discharge
of  any pollutant, or combination of pollutants, notwithstanding  section 301
(a), upon condition that  such discharge will meet either  all applicable
requirements under sections 301, 302, 306, 307,  308, and  403 of  this Act, or
prior  to the taking of necessary implementing actions relating to all such
requirements, such conditions as the Administrator determines are necessary^
to  carry out the provisions of this Act."  (Emphasis added.)  What that pro-
vision says is  that if the Administrator for some reason has not done what
he's required to do under those sections 	 301, 302, 306, 307  and so on 	
he  may require  "such conditions as the Administrator determines  are necessary
to  carry out the provisions of this Act."  Since the Administrator has not
prescribed  effluent limitations under section 302, and even though the ef-
fluent limitations under  section 301 and 304 have not become final - - nor
have those  for  new source performance standards  or toxic  pretreatment stand-
ards - - the Administrator may apply such conditions as he deems appropriate
to  achieve  the  objectives of this statute.  At the end of that same section
402, there  is a provision, subsection  (k), which  says, "compliance with a
permit issued pursuant to this section shall be  deemed compliance, for the
purposes of sections 309  and 505, with sections  301, 302, 306, 307, and  403,
except any  standard imposed under section 307 for a toxic pollutant inju-
rious  to human  health."   Thus, if you have a permit that  has been issued by
the Administrator or the  state acting under delegation of the NPDES author-
ity, then that  does comply with the law.  The permit contains the bottom-
line conditions.  It is  true that the permits may expire  as late as 1979, so
that your client or company may not be confronted with a  permit  to be re-
issued this year or during  the next year or so.  Remember, however, that a
violation of the permit  is  a violation of the statute, so,  if there is any
evidence that a permit is being violated, either as to the completion of the
waste  treatment facilities  or the actual parameters contained in the permit,
the discharger  is subject to enforcement action  under  the statute.

     One of the "hookers" about which no one  is  certain is  the  application
of  section  505, the first provision of which  reads, "Except as  provided  in
subsection  (b)  of  this section,  any  citizen may  commence  a  civil action  on


                                      8

-------
his own behalf,  (1) against  any  person  (including (i)  the United States,  and
 (ii) any other governmental  instrumentality or agency  to the extent per-
mitted 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 a State with respect to such
standard or limitation,  or  (2) 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."   The  effect  of  that
provision is to say that even if  the Administrator  of  the EPA determines  that
he will not, on behalf of the U.S.,  institute  enforcement action against  a
violating discharger, any citizen may institute action not only against the
violator but against  the Administrator for  failure  to  execute the provisions
of the statute in those cases where  he does not have discretion.

     I haven't read the statutory provision describing enforcement, but that
provision very clearly says  that, when the Administrator  becomes aware of a
violation, he shall 	 not, he may  	 but he shall institute proceedings.
At the risk of being  proved wrong in less than a month, I say again, you  had
better advise your lawyers to be  prepared for  litigation,  if you are in
violation of any part of an  existing permit.

     I will outline what could be, I think, a  realistic chain of events.
First of all, any Administrator subject to confirmation has to commit himself
to enforcing the law.  Can you imagine a question in a confirmation hearing
where a Senator asks  a prospective administrator, "what do you intend to  do
about the July 1, 1977 deadline?"  If the Administrator wishes to "pass
muster," he almost has to say, "I will enforce the  law."   Of  course there is
a difference between  a confirmation  hearing and July 1, 1977, but the issue
will remain on July 1, 1977.  The Administrator has taken an oath to enforce
the statute.  If the  EPA decides  that they will selectively enforce, either
against industrial dischargers or against a selected group of industrial  and
municipal dischargers, in my view, it will merely be a question of time until
they are compelled to file suit against every discharger  in violation of per-
mit conditions.  Which is to  say,  they may begin with the  intent that they
will selectively enforce the statute, but it will merely  be a question of
time until events will lead  them  to  be compelled to file  litigation against
every person they know to be in violation of 92-500.   No  one  could publicly
argue that somehow o-r another it  is  alright to enforce the law against one
discharger that is in violation and  not against another discharger that is
in violation; there would likely be  a public outcry.   Remember also that  it
merely takes one person - - wherever the violating  discharges are occurring,
it only takes one person to decide that they will litigate against the dis-
charger and the Administrator simultaneously and, presumably  in the same
litigation, if they wish to bring a  violating discharger  to account.  Thus,
you are confronted with that possibility from  the outset.

     I've concentrated primarily upon what I regard to be the major regula-
tory issues that not only the petroleum industry but also all industries
will face on July 2nd, that is, what do you do about your permit conditions
if you are in violation?  Now of course, you can solve this problem by say-
ing we are 100% in compliance and then you don't have  to  worry about the
possibility of litigation.

-------
     I'm supposed to say something about research, so let me just add one or
two words about that topic.  First, nobody yet knows how much sludge we are
going to generate or how many residuals we're going to have all across the
U.S. from all these water pollution control efforts.  If you're involved in
a research area, sludge is one of the areas on which you should concentrate.
The next major issue, in my view, will be the identification and control of

non point sources.  Another is,  if we can reduce the volume of water being
used in such a way  ss to close the cycle, I think more and more in  the future
the emphasis will be on closed cycle systems.  Water reuse will continue to
be critical and is  highlighted by such episodes as are now being experienced
in California.  Probably, the most critical approach to research, however,
is to design production processes to minimize the generation of pollutants.
If you really want  to be on the  leading edge of the future in water pollution
control, concentrate on the production process rather than the end-of-the-
pipe, or minor modification at the end-of-the-pipe, processes for waste con-
trol.  Those firms  which manage  to design processes that are specifically
intended to minimize pollutants will be the ones that have the business in
the future.

DISCUSSION

J. Dewell - Phillips Petr. Co. -  In those states where they have require-
ments more restrictive than EPA where those states do not in themselves have
delegated authority but those restrictive requirements have been written into
the EPA permit, to  what extent will EPA act to enforce those more restrictive
state requirements?

Moore -  Let me see if I can restate the question as I understand it.  In
those states in which there are water quality standards which require strict-
er levels of treatment than the  effluent limitations prescribed by EPA and
the conditions are  in EPA permits, to what extent is EPA likely to move to
enforce those permit conditions?  I'll hazard a guess.  I don't think EPA
has yet seriously considered the problems of enforcement generally so I
don't think that they fully appreciate the question yet.  My guess would be
that EPA will not move initially to enforce to achieve water quality stand-
ards fixed by the states so much as they're likely to move to enforce a-
gainst those who have violated permit conditions under the effluent limita-
tions of 301 and 304, which is to say that I'm not sure that EPA will assign
very high priority  to enforcing water quality standards.  You may recall
that in the summer  of 1976 the then EPA Administrator went to the Congress
with a recommendation that federal assistance for publicly owned treatment
works for levels of treatment beyond those required for secondary treatment
in 92-500 be withdrawn.  That is to say, they were suggesting that the
Federal government's interest should be restricted to a minimum level as
prescribed at the federal level rather than that prescribed by the states.
So, I don't think,  offhand, that they'll move to enforce in those areas
where they have water quality standards.  Does that answer your question?
                                      10

-------
BIOGRAPHY

     Until May of 1976, Joe G. Moore, Jr. was Program Director of the
National Commission on Water Quality, the agency created in the Federal
Water Pollution Control Act Amendments of 1972 to analyze the technological,
economic, social, environmental and institutional imparts of PL 92-500. He
is presently Head, Graduate Program in Environmental Sciences, The University
of Texas at Dallas.  Mr. Moore has had some 20 years of governmental service
including Chairman, Texas Water Quality Board, Executive Director of the
Texas Water Development Board, Administrative Assistant to Texas Governors
Connally and Daniel, and Commissioner of the Federal Water Pollution Con-
trol Administration, Department of Interior, directing the national water
pollution control program before the creation of the Environmental Protec-
tion Agency.  He is a graduate of The University of Texas at Austin with
bachelor's and master's degrees in government.
                                     11

-------
                   SESSION I

REGULATORY AND RESEARCH DIRECTION IN U.S. EPA


            Chairman

            William C. Galegar

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


            Speakers

            Lamar Miller
            "Toxic Strategy Overview "


            Marvin L.  Wood
            "EPA's Research in the Refining Category"
                      12

-------
                          "TOXIC STRATEGY OVERVIEW"

                                 W. Lamar Miller
                            Effluent Guidelines Division
                          Environmental Protection Agency


          Good morning,  it is very nice to be here  today.  I am not sure whether I  am a
designated hitter or a pinchhitter.  I guess the effect is really the same.  Mr. Schaffer
does indeed send his apologies for being unable to be here.  He is participating in a
command performance this  morning and tomorrow on  some of the very topics which Joe
Moore has just been discussing with you.  Most specifically, the agency's suggested input
for midcourse corrections which the agency would like to recommend to the Congress in
the very near future, as they relate to P.L. 92-500.  I think that may be a sufficiently
good excuse for his inability to be  here today.  He did want to be here.  I'll try to do my
best this morning to try to tell you  the same story that he would tell you if he were here.
I'll exempt most of the humorous remarks from that qualification. I  don't think that Bob
would use the same stories  that I would tell you,  but the serious part of the speech will be
close.

          I'd make one observation on a remark that the keynote speaker just made.  It
somewhat puts me in mind,  Bill, of a cartoon that I keep pinned to my  wall to remind me
of our actions.  Joe Moore's remarks about the approach used to evaluate the cost for
implementation of BPT.  The cartoon on my wall is two cavemen talking to one another.
One says to the other in a  very philosophical note,  "Our future is shaped by our past, so
be careful what you do in your past."

          Be mindful that  the estimated cost for the installation may also become the
amount of the fine if the installation is not made. Very soon after the  Supreme Court
decision was announced I called up a friend  in industry and said, "congratulations Bob,
your corporation's name will go down in history as the first major loss in the water fight.
That case will be known forever as the du Pont case. And, by the way, what did industry
win on the appeal? Hadn't you really rather that you had lost some of them and there-
fore kept the old rules instead of the directions that  you hope  I'll tell you about today?"
I  think we ought to keep some of these thoughts in mind as reminders and go back and
look at the 1976 meeting out here at Tulsa.  I looked at some of the remarks which were
made by the keynote speaker last year, as a matter of fact,  Bill Galegar said that
participation and communication made for a  successful meeting; and, he asked this
morning, how are we going to improve on that success.  The keynote speaker last year
said that effective communication is a two-way street and cited Joe Moore on effective
communication in saying,  "we must not only listen we must hear what is said."  And then
a number of speakers proceeded to  tell everyone at the meeting that comments should be
solicited from industry.  Industry should be heard.  EPA should listen to the industry.
EPA should know industry's viewpoint, then reminded the audience  of what the  courts had
told EPA about the errors of their direction at that time.
                                        13

-------
          Folks, at this meeting, this morning,  I would like to point out to you that Mr.
Moore was right about communication.  We must not only  listen but we must hear what is
being said.  Mr. Beychok was also right; that communication is a two-way street; but it
needs to be pointed out again; it really  is a two-way street. We have to listen to you.
We want to listen to you.  We're trying to  listen to you.   And  we're scheduling a lot of
time to sit around and talk to you and listen to you. But,  it is a two-way street. While
we're listening, please hear what we're saying to you. It works both ways.  Nobody in
this audience is any more tired,  and I'm glad Joe asked how many lawyers were here.  I
didn't see but one hand go up, I might have been wrong, maybe there are more, but we
are all technical people,  let's think that way for a minute.  None of the technical
people in this room are any more tired of the adjudicatory processes than are the technical
people at EPA.  And the people  in the effluent guidelines division are the people who are
catching the bulk of that work effort.  We don't encourage the adjudicatory system; we
don't want it; and, the directions are sufficiently different at this time that I don't think.
it is going to be as necessary. Communication, and every time I think about the topic I
am reminded of a good  Polish joke,  a good  Polish joke is a joke told to me by Polish^
people,  I'm not one of the government employees who wants to get hung by the ethnic
joke problem.  I've no interest in telling that type of joke.  But, it does reflect both on
politics and is a reasonable observation of the problems of communication.  It seems that
in that great hall up in the sky,  once upon  a time,  Lenin, the Czar of Russia and
Napoleon were reviewing history as they saw the problems that they had in  their political
life and the problems in Poland today.  Poland's problems were very much in a political
discussion last fall if you remember,  Lenin observed that if he had had the ability to
control communication and what people understood as a result of mass communication the
same as the First Secretary of the Party in power during the 1960's uprisings in  Poland,
that he would have won the revolution in Russia  in 1908, he wouldn't have had to wait
until 1918. The Czar responded that if he  had had  the ability  to control communications
to the people the same as he had observed had been happening today in  parts of the
world,  that no  one would ever have heard of Lenin, and that he would still be the Czar.
Napoleon responded that  if he had the ability to control communications and what people
heard and what people understood as was done by the Polish First Secretary, that he would
be convinced today, that after all,  he had, in fact, won at Waterloo.  I sometimes think
that most of us  hear and most of  us listen to just about what we want to hear, and that
means both  EPA and industry.  We're really trying to change that in the effluent guide-
lines division.   We're really trying to change it  at EPA.  Remember that if we both go
that extra mile it will make it a  lot easier to make it a two-way street.  We've got to
both make the extra effort to listen and  to talk.  I believe that this  group, represented by
refiners and people associated with them, can really make that possible. I  don't believe
that it is necessary for either of  us to prove that  we are willing  to go to  court, we've
already done that.  I hope that it isn't necessary for us to  prove to each other that we are
failing by having to go to court. We've both done  it enough times to know that we're
willing to go back if necessary but I hope that we can avoid it by learning that it is
unnecessary. Technically we can resolve our next series of issues.  We're both going to
have to learn to listen to both sides and we're going to have to both learn to go both ways
on a two-way street. Let's not spend all of our time worrying any more  about categori-
zation, let's not spend it worrying about the semanic differences between limitations and
guidelines. It  was very interesting to me to hear Joe tell  you this morning,  that's settled.
As I understand it,  it is settled.  The problem is, and our direction is, (this is very
important to us  and this is literally the most important thing in terms of the  costs and
economics of the effluent guidelines division right now),  is to determine how best to do
the regulations.  By spending our money, and yours, in such a manner as to derive the
greatest benefit for the  environment for  every dollar that is spent by us or by you. But


                                        14

-------
we're not interested In spending money for the sake of spending money and we're not
interested in deriving regulations for the sake of deriving regulations.  We are interested
in implementing the law as we understand it.  It was once said by a famous English
politician, Gladstone, the only thing "worse than a politician who thinks he has the
truth up his sleeve is one who thinks God put it there."  I want you to know that that
particular observation is to me somewhat like the best advice that my father gave me
(since Bill digressed into the good old boy storeis I think I will too). Dad looked at me
once when I was just ready to finish college and he said, "son never argue with a fool
in public, passersby can't tell the difference."  I didn't recognize for twenty years that
my father's remarks were really in fact a paraphrase of Justice Oliver Wendell Holmes'
remarks published as an essay which he entitled,  (I think fittingly for this particular
group), "The Hydrostatic Paradox of Controversy", in which Mr. Holmes starts off by
reminding us that if we had two pools of water, one the size of a soup bowl and one the
size of an ocean, connected by a pipe stem, the  levels of the water in each would be
the same.  This is very similar to the status of both wise men and fools engaged in contro-
versy. The  major difference being, the fools know it.  The same is true of most of us;
when we all get reduced to a common level  of thinking that we have a noble exercise to
go about, and no one can challenge or have any  interest in adding to what we do.  Now,
you can put the last remark in perspective.  I'll tell one more joke,  current from the
Washington  scene about five men on an airplane.  The pilot comes back and he says,
gentlemen I have put this  plane on automatic, I've come back to tell you personally that
we have a real problem.   This plane is not going  to make it to the airport, and we're too
far from any other place to satisfactorily land it.   I know nothing else to do but tell you
that we have four parachutes on board,  and, unfortunately, there are five of us.  You
might think, as a general  rule, since I am the captain of this ship, that I would give you
each a parachute and allow you to jump while I stay behind with the plane.  That, hoW-
ever, is not to be.  I  think that you should know  that I work for the CIA and  I have some
very important information up in my mind that must get to the president tonight.  So with
that I am going to take my chute and jump out the door. I'll leave the problem to you.
Another guy stood up and  said;  "gentlemen you may not think it's very  important, I'm
only an ex-president but it's very important that I survive because  I still give continuous
advice to the new administration, and I think that none of you would deny that I ought
to have one of those chutes." With that he straps on a chute and jumps out the door.
Another fellow stands up and he says; "I have been the Secretary of State and undoubtedly
have been recognized overwhelmingly as the outstanding intellect of the world.  I must
be rescued for the benefit of all the world.  My very being is very important; and with
that I will leave"; and he jumped out of the airplane.  There was left a hippie student
and a minister.  The student looks at the minister and he says; "father, I think that you
should take  the remaining chute." The minister  looks at him and says, "no, son, you
take it, you take it."  The student looks around and he stirs around a little bit and says,
"well I don't really think  that we need to engage in this conversation."  The minister
looks at him and says, no, this is my life's work,  to serve my fellowman, this is all I
have to do, to make other people happy,  you should take it and leave  and I must make
the sacrifice.   I've had a  long and healthy life and I've enjoyed it.  You go and you
save yourself."  And then the boy looked back at him and he says; "but father,  it isn't
necessary for either one of us to make this decision; you see,  the smartest man in the
world just jumped out the  door with my knapsack."  I want you to know that the effluent
guidelines division of EPA does not take the attitude that we are necessarily the smartest
people in the world.  We  are neither going to jump out of  that door with a knapsack,
nor are we going to drive  it straight into a wall.  Joe pointed out, and I think well,
that our litigation record was not too good.   He's right.  We intend that it's going to
get a whole lot better.  The reason for that  is that we don't intend to issue regulations


                                        15

-------
that we can't sustain.  And If we can sustain them, I think you will probably be reason-
ably ready to agree with them.  We may not have to have too much litigation^in the
effluent guidelines area.  Now, to give you a reasonable overview of our toxics
strategy, which is presumably what some of you wanted to hear from me.

           On June 7, 1976,  we entered into a settlement agreement which settled
four cases brought against the agency by environmental groups.  These cases were
brought against the agency  to force us to expand the list of toxic substances under
section 307, to promulgate  final standards  for toxic substances already designated under
section 307, and to finalize 8 pretreatment standards.  I'm not going to discuss at length
the details of the settlement agreement in as much as at least two other times during this
conference it will be discussed by people who are more competent than me to discuss it.
The materials that are dealt with under section 307a for which the final standards must
be promulgated are pretty well known to most of you; aldrin, dieldrin, benzidene,
endrin, DDT, DDE,  DDD.  The other materials which are identified in the settlement
agreement have come to be known as the list of 65.  Sometimes also referred to as the
list of 129.  Sometimes also known as the list of 543.  Sometimes referred to as the list
of 43,000. We will specifically delineate later in the  meeting for you by another
speaker from our division, exactly what the list really means to this industry.  IHs our
attempt to try to focus your interest on specific materials. It's not  going to continually
change.  We have identified  129 today.  We hope that they  will be satisfactory to do
the job.  The effluent limitations will be established on highly specific, technology
based,  limitations for 21 industrial categories identified in the agreement.  At the risk
of confusing this particular  issue, also the  21 are now in fact 28, because miscellaneous
chemicals which is one of the 21 is made up of 8 different categories.  It's no longer 28,
because one of them as of last week is settled and will be no discharge.  The carbon
black industry.  That puts an  end to one of the 28.  The effluent limitations will be
promulgated in the form of best available treatment economically achieveable for
July  1, 1983.  This particular work activity has become known as the BAT revision in
the terminology of those developing effluent guidelines.  We also sometimes refer to it
as BATox, or BAT combined with toxic controls or priority pollutants.  You may hear
that term in the future.  What that really means is best available treatment with priority
pollutant control. These standards will also include pretreatment for existing sources,
pretreatment for new sources and new source performance standards.  The settlement
agreement reflects the agency's strategy for handling toxic materials and other problem
pollutants discharged into the nation's waterways.  A number of control options are
available to the agency and standards and  regulations will be developed taking many of
these options into account. The first step will be to establish technology based
effluent limitations in a similar fashion as we did  for previous effluent limitations, we
don't expect to put any out that are not technically sustainable.  If it is determined that
these technology based effluent limitations are not sufficient to control the problem
pollutants, the agency has the option of designating a particular material  for coverage
under section 307 and will not hesitate to  do so.  Under this section more  stringent
standards may be established without the detailed economic considerations as required
in establishing  effluent guidelines under sections 301, 304, and so forth.  There is also
the option of developing and  upgrading existing water quality standards to resolve
potential  problems with specific materials  of concern. You heard Joe refer to a little
used section of the Act this morning.  That particular sentence refers to the same little
used section.  It is an option  available to  us for the control of toxic materials.
Specifically, the use of water quality standards to resolve specific problem areas.  Our
studies  will proceed along slightly different lines than they have proceeded in the past.
In fact  the studies will proceed along  lines which reflect the organization of the Office


                                        16

-------
of Water Planning and Standards.  We have four divisions and we have four study areas
ongoing at the moment.  Technical contractors will establish alternate levels of
technology; and that is an important, perhaps new, consideration. -We expect to have
more than one level of possible treatment technology designated  in thfe development of
the regulations.  These studies will look at the specific pollutants, i.e., the 129 priority
pollutants and the traditional  pollutants.  Secondly, economic impact analysis as was
done in the past will be a major factor and a major input to the overall study.  Thirdly,
studies to provide facts with regard to the impact of the designated priority pollutants on
the aquatic environment will be conducted.  Fourth, a study of the environmental mass
balance of each of the materials will be conducted.  This effort will look at the manu-
facture of the material, the movement of the material  into various products and ultimately
into the aquatic environment, and the source of the materials which do move into the
aquatic environment.  Each of the above factors will be weighed in determining the
ultimate choice of technology and/or control options for preparation of regulations, not
just technology and not just economics, but a mass balance and the effect on the aquatic
environment.  It is our intent, because of the complexities of all of these studies, to
begin by screening specific  industrial discharges and identified subcategories to deter-
mine the presence or absence  of the designated priority pollutants. I made a short
presentation to one committee of API, not very long ago, and within two days  the Deputy
Assistant Administrator got a telephone call that said;  that I said at the meeting, we're
going to have 129 regulations for every refinery.  I don't know what I  said that led the
man to believe that, but I want to make sure that you  don't go away believing that.  The
first thing that we are  going in our technology based studies is to determine how many of
the 129 are in fact a problem, and how many are not present in significant quantities,
i.e., how many can we eliminate from being a concern from a particular industry or a
particular category.  This, hopefully, will allow us to narrow down the number of
materials with which we have to be concerned in developing treatment technology.  Once
this has been done, we will proceed to quantify on a more extensive basis the significance
of those materials that we found to be present, that is, how well  does existing  treatment
handle those that are present and how many of those are no longer a concern.  There will
be some cases,  for some categories, for some processes, where BAT is equal to  BPT for
one of several reasons. Either priority pollutants are not present, or the existing treat-
ment renders them dischargeable in insignificant quantities, or they do not constitute a
major source to the total aquatic environment, or mass balance studies indicates that
the point sources under study do not constitute a major portion of the environmental
impact of the pollutants.  There are many reasons why any of these may not cause you a
problem.  Once we have quantified and made the determination that a material is
present,  and that it is  present in an environmental significant quantity, then and only
then will our efforts be concentrated on the selection of appropriate control strategies.
As we  proceed to develop these standards we will be opening up  the process and will be
inviting more active participation of the industrial  community.  We will again be
soliciting data,  and their assistance in areas of technology and its application  to the
problem.  As a matter of fact, we have already begun soliciting assistance in technology
application. We intend to let everyone know as best we can what we're doing and what
levels  of technology are under consideration at all  times.  We're going to play the entire
scenario right on top of the  table in the public view for all interested  parties.  We intend
to use  fully the authority that we have been granted under section 308 of the Act to gain
information with regard to the various industrial categories on the specific materials of
concern.  We're also working very closely with industrial associations, not only this one,
but others, in developing working groups, particularly in the area of analytical methods
development and technology assessment.  Of paramount importance to us is orderly and
timely progress.  In reading the settlement agreement one will  notice that there are


                                        17

-------
specific dates by which time certain portions of the job must be completed. The
schedules are very tight.  We will not have the time to provide leisurely reviews.  Since
we are opening up the process, soliciting cooperation and allowing everyone to know
what we're thinking as we proceed, we feel that we can eliminate long review periods
as well as a lot of litigation.  Everyone is concerned with the discharge of problem or
toxic pollutants and as good citizens everyone wants to control them to the maximum
degree possible.  We believe that as long as we proceed in a rational, cost effective
manner,  we will  make progress together.  We believe this to be the proper strategy for
the control of the problem.

DISCUSSIONS

Pete Foley, Mobil Oil Co.  - Can you give  us an example of how economics will  play a
role in choosing the appropriate technology?

Miller - Yes, I can give you an example. Let me just try to describe a very specific
case.  Our Analysis and Evaluation Division will  take  the cost data associated with each
technology level and evaluate it. There  may be for instance a number of technology
levels and an associated cost with each one. Obviously one level is no treatment.
Another  level may be, for instance, just an API separator.  Another level may be an
API  separator followed by an extended aeration activated sludge system.  Another level
may be a bioplant combined with a filter, or combined with a  carbon column within the
process,  or a carbon column at the end of the bioprocess. Each of those  levels will be
costed.  Then the impact on the industry and the economy as a whole will be determined
prior to determining the trade off or the benefit which  we think is most cost effective.
We will  do this for a  number of treatment levels for each regulatory action.

Milt Beychok, Consultant - I was interested in hearing you say that EPA would again use
contractors to gather  a great deal  of the background information.  They do that, and
whether  or not the EPA later, as they become more knowledgeable in the field, depends
to a great extent on those contract reports I'm not sure.  But they]re at least preceived
by those of us in the field as being a data base that must be  contested.  Industry often
ends up hiring their own contractor to prepare a rebuttal to the first contractor or to
prepare cost  estimates to rebutt your cost  estimates.  Has EPA ever considered opening
up the process at the  front end and gaining the review  and approval of all interested
parties in the selection of that first contract?

Miller - I'm  not sure  I really understand the question.  I think that the answer is, no.  I
don't think that our administrative procedure allows the government to allow someone
else to determine how they will contract for their business.  There are some pretty
stringent rules on what you go through to  select contractors.  Let me make one remark in
regard to what you said about industry's general feeling or your general feeling about
contractors.  Let it be clear, at least from this day forward, no contractor writes
regulations.  I've seen that  suggested in the past.  The policy is pretty specific.  All
regulatory actions will be the specific responsibility of the appropriate project office
and  his supervision.   The first level of regulatory recommendation we also changed as
you  will  hear in a subsequent presentation by Bobby Del linger. The system by which
we review various levels of  technologies, the costs, the associated economic review,
the mass  balance in the environment, and the rest of it has been put into what we call
a working group.  This group has some highly specific  responsibilities.  Bobby will dis-
cuss those with you in more  detail, but let me just tell you that it's not going  to be the
work of a contractor that puts out a draft  regulation for you  to have to worry about.


                                        18

-------
It's going to be very easy for you to point a finger and say, this man is responsible for
this exercise.  We subscribe to that policy and it is the state  policy of the new adminis-
tration. You ought to be able to identify who it is and we're going to help you know
who it is.   It will not be a hidden contractor as it has been sometimes alleged to be in the
past.

Milt Beychok, Consultant -  I appreciate that, and I wasn't suggesting hidden contractors.
I also appreciate that no one can participate in your selection of the final contractor in
terms of the contract terms of price or scheduling. It might be well for the EPA to
solicit from industry whom they at least consider competent enough to be on the original
bidding list and who really does have experience in that particular industrial  category.
That was really all that I was talking of.

Miller - I have been in that position before; most specifically in the inorganic chemicals
and the organic chemicals industries*  Where industrial representatives said,  "why don't
you ask us who you should hire?" Which is effectively what  you have said.   I looked at
the particular  company that  made the remark and asked, "is your engineering division
available?"  "We  know they are competent, we know they know how to design these
processes, will you bid?" And the answer was,  "no." It's not always that we don't get
the people you want for us to have because we don't try.  Mostly, it's because they don't
want to get into government audit problems.  The most recent case that comes to my mind
is an extensive carbon column evaluation study where I asked a lot of people, everyone I
know, who was qualified, please bid this time, please quit criticising us for the quality
of our contractors when you  won't even bid.  And the bid  list was not very long, and it
was for quite a number of dollars.  Yes, we have trouble getting good contractors; but
the contractors we have, if we don't hold them to the line, don't blame it on the
contractor.  It's our fault.  We have a job to supervise them and if we don't do it, it's
our division's responsibility to come up and say,  it's our ball. And we will this time.  I
think the one we have on this project is doing a  very good job right now.

BIOGRAPHY             W. Lamar Miller

            W. Lamar Miller is the Chief of the
Organic Chemicals Branch of the Effluent Guide-
lines  Division. He has a B.S. degree from
Mississippi College and an M.S. and Ph.D.  in
Chemistry with minors in Engineering from the
University of Florida.  He spent twelve years in
teaching and industrial work prior to joining the
Environmental Protection Agency in 1972.
                                         19

-------
                   EPA'S RESEARCH IN THE REFINING CATEGORY

                            Marvin L. Wood,  Chief
                          Source Management  Branch
              Robert S. Kerr Environmental Research Laboratory
                             Ada, Oklahoma 74820

     Actually, this is going to be more of an historical reminiscence than a
current commentary on research the EPA is doing on refinery wastewater.  For
one thing, we haven't done any research since November;  we've been working
for Effluent Guidelines on a technical assistance project.  Secondly, all
the projects that we have any current interest in are listed later on in the
program; thirdly, rather than chance saying  something that would be corrected
by a person who actually knows what they are talking about, I'm just going
back in history and bring you up-to-date, partly from a philosophical angle,
about what went on in the formulation of the research program and some of the
results of our earlier projects.  Actually,  the reminiscing may be apropos
since a Harvardian, a fellow by the name of  Santayana, once observed that
those who do not learn from the lessons of history are condemned to repeat
them.  Do we have the time or money to afford to repeat some of the lessons
of history that we should have learned already, particularly in wastewater
treatment and control in this specific industrial category?

     For those who expected a more western philosophy here in the oil capital
of, at least, the United States:  there was  once an Okie (not from Muskogee,
but a little closer to here—the town of Claremore) who observed that all he
knew was what he read in the newspapers, and he became rather famous for his
comments based on that observation.  Wouldn't Will Rogers have had a field day
reading some of today's headlines in the newspapers?  Especially with regard
to aspects to the energy "crunch," "crisis," or whatever tag you want on it.
Depending on which expert is being quoted, we may expect within the next one
or two years to buy our gasoline in small goatskins, or at the other extreme,
we might still be driving our gas guzzling V-12's by the year 2077 at a modest
increase of X dollars per gallon for fuel.  The truth, as usual, probably lies
somewhere in between.

     Where there are problems, there are ways to solve them.  A technology as
advanced as ours in a country that lives, moves, clothes itself, and produces
its food from liquid hydrocarbon products can afford to give up this form of
energy.  We are not then in the position of  holding a wake rather than forum
or pondering over, for example, the wastewater treatment problems associated
with buggywhip factories.  Regardless of whatever source these liquid hydro-
carbons are to come from in the future, they're going to have to be refined
and their changing characteristics will cause new wastewater treatment prob-
lems.  I hope to cite herein some of the lessons of history so that we may
avoid being condemned to repeat them in our future research and our future

                                      20

-------
wastewater treatment and control efforts. With respect  to the  soothsayers of
doom or Utopia, some of the fairly recent economic projections were  that there
would be no shortage of crude because shale oil would  be available in large
quantities when the price of crude reached five dollars a barrel.  Crude is
not cheap, not plentiful, and not coming from shale.   So much for soothsayers
and oracles—one would almost be tempted to put them into a mythological
category with demons, ogres, virgins, and dragons.

     What is known at present, for example, is that there is  more oil in known
locations than we have consumed since Colonel Drake drilled his  first well in
Pennsylvania.  Admittedly, there are problems in  recovering this oil, since
it has been subjected to primary production methods, and in some cases,
secondary production methods.  It is difficult to believe that a nation can
afford to mine rocks on the moon but cannot develop technology to extract a
little more of the crude that we know is there  (and to a goodly  degree, how
much is there).  Hopefully, we should be able to  extract enough  to give us
some latitude until adequate stocks are available from exploration and new
production, coal conversion or shale oil or even  such  "way out"  things as
hydrocarbon production from sunlight and cellulosic materials to make a new
 (and continuously renewable) source available.  Admittedly, everything is
finite and so are these potential sources of the  "new  crude." Nevertheless,
until the little handy-dandy portable nuclear fusion energy generator comes
along, we're going to have to learn to utilize them, and utilize them in a
more wise fashion than we have in the past.  America does live on liquid
hydrocarbons; we have for seven decades, since the automobile and internal
combustion engines were developed; and we are going to continue  to,  because
we  cannot replace overnight the transportation network that moves our fuel,
our food, our fiber, our raw materials, and our finished products.

     The situation does suggest that any new treatment options be regarded
carefully since changing feedstocks are well known to  have changing  effects
on  wastewater effluent parameters  (physical as well as chemical  and  biological)
and the treatability there of.  The answers to today's problems, as  we see
them now, are certainly not expected to be absolute.   Until Woehler's syn-
thesis of urea, chemists regarded organic chemicals as incapable of  production
by  man—they were produced only by nature.  Until 1903, the oil  industry
"knew" that there wasn't any market for aviation  gas or jet fuel—there were
no  airplanes—how could there be a market?  Until relatively  recently, some of
us  who call ourselves sanitary engineers—and before I offend more people in
the audience, I'm registered as one—anyway, some sanitary engineers just
"knew" that the Imhoff tank or Emscher tank which removed the big pieces was
going to solve water pollution problems.  Then when biological treatment was
developed, we "knew" that it would solve all the  effluent treatment  problems.
We  have been remiss in being so proud of the percentage removal  figures; it's
not what we took out that counts, it's what we left going back into  the
environment that has hurt us in the past.

     If I may leave general philosophy, we can, with one dirty word, start
concentrating on the background of refining wastewater research  done at  the
Robert S. Kerr Environmental Research Laboratory  (RSKERL).  The  dirty word
is  "matrix"; that word was "out" not too long ago, maybe it will be  "in"
again by tomorrow; but it fits this situation.  If you will consider the


                                      21

-------
number of control and treatment options that are available for effluents in
general, you will find that it presents to a rather complicated listing.
For example, select a number of simple headings for a series of columns—such
things as chemical, physical, biological, and reuse/recycle.  Now picture any-
one of these headings and list the number of past and present treatment options
that are available.  Let's pick biological.  Most of us are principally con-
cerned with nominally aerobic processes, but there are also anaerobic pro-
cesses, and in addition, there are some facultative treatment schemes which
in nature work both aerobically and anaerobically.  How many nominally aerobic
processes can we list under one of the three subcategories?  How about slow
sand filters, high-rate trickling filters, low-rate trickling filters, contact
beds, Hayes process, activated sludge, extended aeration, Passvier ditches, and
a few more?  The matrix, for that is what we have, depending on how closely
one defines each simple single unit process or unit operation, can extend
to nearly 100 possibilities.  The combinations and permutations possible for
a treatment plant are somewhat horrific to contemplate, especially if operat-
ing from a zero based budget.  Incidentally, a few people have observed that
insofar as research on treatment of refinery wastewaters, we have been operat-
ing from a zero base budget for the last eight years, so anything would be an
improvement—I didn't say that; I'm just repeating it.

     One of the most apparent, logical, rational approaches in winnowing out
the more promising treatment options for any given effluent is the good old
state-of-the-art approach.  Investigate what has been done; compare the results
to what is being done; what has been dropped; what appears capable of refine-
ment; and concentrate on the most promising.  Admittedly, a state-of-the-art
project is highly attractive to many of us bureacrats; after all, it costs
relatively little money; it seldom fails to yield a nice thick pile of paper
as a report, and it cannot be attacked technically except on the grounds of
illiteracy, if the author has failed to research thoroughly all previous
references and include them in his document.  Rest assured that the state-of-
the-art, which initiated research in refinery wastewaters at the Kerr Labora-
tory was not predicated on this "safety first basis," but the fact that we
had very little money, a small staff, and as it appeared from the recommenda-
tions of the finished document, a lot to do.

     It is generally conceded that it is wise to be aware of the physical,
chemical, and biological characteristics of any wastewater which you wish to
treat.  Since treatment plants involve a sizeable investment of time, money,
construction materials, and possible untoward affects downstream (should they
fail to perform as designed), a companion report to the state-of-the-art was
prepared to delineate wastewater characteristics of the various unit processes
and unit operations which existed in a modern refinery, and which, together,
represented the composite wastewater stream.  Actually, it was not until
several years after these two reports were published that we obtained a good
statistical basis with respect to only the most common physical and chemical
parameters for the composite stream.  This basis was developed during the
EPA/API Wastewater Characterization Survey which happened to be another
Effluent Guidelines Division technical assistance project.  The present EPA/
API study for the Effluent Guidelines Division will develop similar information
with respect to many, but not all, of the organic chemicals and their charac-
teristics in refinery wastewaters.  No such study, however, is currently


                                      22

-------
planned with respect to the characteristics of individual unit process or
unit operation effluents.  Yet, we know that they vary widely, on the basis
of singular or limited observations, from one process to another, from one
unit operation to another.  Since it's generally easier to find the needle in
a bale of hay rather than the entire stack, I would commend to your reflection
that this is a fertile area for some future research.

     From these two cited documents, plus some black magic or legerdemain
known as a planning system, we developed an in-house document which laid out
on a PERT-type chart our research program in terms of where we had to go and
how to get there—a plan for obtaining the most widely usable results for our
limited research dollars.  Although barred from including air pollution con-
trol and solid waste control research, the program document itself did recognize
the obvious inter-relationships between the three media and did attempt at
all times to prevent creating a new problem in the course of finding a solution
to an existing problem.  We did this without the usual standing jokes about,
"what the hell, let's airstrip it and let the air polluters worry about it."
Built into the research program, and continuously upgraded as technology and
techniques allowed, was the concept of quality control from initial research
project design through the sampling procedures through the analytical proce-
dures to the data evaluation procedures.  Those of you who have been assoc-
iated with projects which EPA has funded are aware of EPA's analytical and
quality control manuals—the "CuSum" statistical quality control procedure in
the latter manual was developed originally at the Kerr Laboratory.

     It is germaine to our forum to review the findings of the original state-
of-the-art report and compare them to the projects which have been developed
since the start of the program in 1968, nearly one decade ago.  Particularly
of interest, are those recommendations which were not acted upon through lack
of applicant, lack of funding, or lack of analytical and engineering technology.
Some are specifically pertinent to the potentially painful problems that we
are facing now regarding the "priority pollutants."  With your permission I
will read these 20 recommendations, instead of depending on memory, because
some of them are so beautiful in their simplicity and yet farsightedness:

     "1.  Develop sampler to obtain representative samples of floating oils,
dissolved oils, emulsfied oils, and oily sludges."

     Has anyone yet developed a sampler that would do all that?  I think not.
There has been progress but a representative sample still remains one of the
most difficult and probably one of the impossible things to obtain from any
wastewater stream.

     "2.  Conduct internal refinery studies to reduce waste volumes and
strengths for old and new refineries."

     Back to unit process and unit operation waste streams.  Has anyone done
any definitive study of these?  In a few instances, that I am aware of, yes.
We are beginning to approach the area of this recommendation, but not exten-
sively.
                                      23

-------
     "3.  Extend biosystem studies to optimize treatment efficiencies and
handle shock loading."

     Admittedly, there has been a great deal of work on optimizing biosystems
firom an operational standpoint.  A project in an allied field of research may
have developed a monitor that, while not eliminating shock loading, will at
least give adequate forewarning to divert and hold spills or shock loadings
and toxicants that might upset a biological system until they can be handled
separately, diluted, or bled back through the system at an acceptable concen-
tration.

     "4.  Devise a continuous monitor for hydrocarbon detection in wastewaters
using common refinery laboratory equipment."

     One of our laboratory scientists has developed a procedure using a gas
chromatograph, which is now common equipment in most laboratories, to determine
hydrocarbons in wastewater.  As opposed to the freon, or petroleum ether, or
carbon-tetrachloride or hexane or whatever your solvent preference is for oil
and grease extraction procedures, this method appears to have repeatability
and reproducability characteristics.

     "5.  Design original wastewater treatment systems for the petroleum
industry."

     In the case of new grassroots refineries, scarce as they are, this is
being accomplished.

     "6.  Perform chronic (long term) toxicity studies on treated effluents."

     Our Laboratory in conjunction with Dr. Mount's Laboratory in Duluth,
Minnesota, has started some screening studies in this area.  Additionally,
some work has been done at Oklahoma State University.

     "7.  Identification of toxic components in petroleum wastewaters."

     Refer to the forum program for further information.

     "8.  Develop efficient devices and techniques to remove oil spills on
diverse waters surfaces (i.e., swamps, rivers, and turbulent seas)."

     One of Murphy's laws; if you're going to handle the stuff, pump it, or
process it, sooner or later you're going to spill some of it.  A great deal of
money has been expended in developing spill clean-up equipment, procedures,
processes, and techniques.  But a total answer approaching a respectable
efficiency for all problems of all types of spills is still not within our
reach.

     "9.  Assess environmental effects of spilled oils (i.e., volatile,
soluble, emulsified, floating, etc.,) and oil products."

     My God, that's many life times of work for many biologists in one simple
statement.

                                      24

-------
     "10.  Investigate use of cooling towers for treating selected refinery
wastewaters for recycling."

     How much has been done?  Well, some—we're far away from covering all
aspects though, nearly a decade later.

     "11.  Study water reuse within refinery."

     Now certainly a good deal has been accomplished in this area.  Not all
of it in the last 10 years—some of it was done prior to that time—and yet
there have been significant accomplishments in the last decade.

    ."12.  Explore feasibility of phenol removal from wastewaters using
phenol-soluble oils."

     There is more than one plant in the country today that is extracting phenol
using solvent extraction processes.

     "13.  Perform economic studies of brine treatment and disposal on land
and sea."

     This is still a very sticky problem from the standpoint of what do you
do with  it regardless of how you manipulate inorganic salts that constitute
brine,.whether they're from a producing well or a desalter, you still have
the salt left over, and you can't burn it, you can't eat it, you can't sell
it, you  can't use it, and you can't dump it—unless you are lucky enough to
be on the coast.  Most of the brines approximate seawater; admittedly indi-
vidual component concentrations range widely, but the components do approximate
seawater in terms of numbers and types of elements.

     "14.  Develop remote sensing techniques for detection of oil and brine
pollution."

     Supposedly, the CIA can read a license plate from 400 miles out in orbit
via satellite, but nothing approaches that degree of precision and accuracy
with respect to this simple little recommendation of 10 years ago.

     "15.  Perform feasibility studies on by-product recovery from refinery
wastes."

     The phenol recovery technique referred to earlier is, in fact, one of
these by-products recovery operations in actual practice.

     "16.  Devise a monitoring program to prevent subsurface pollution from
abandoned oil wells."

     There has been a considerable expenditure of EPA funds on a GE project
called Tempo having to do with monitoring techniques of groundwater sources
and the presence, direction of flow, and source of groundwater pollution.
Yet, there remains a great deal to be done to put these into effect with
respect to individual plant areas; tank bottoms fail and buried transfer
lines will leak.


                                      25

-------
     "17.  Examine pollution problems associated with extreme cold in the
Alaskan oil fields."

     There's one where they really put some money into.  Considerably more
apparently than any of these other 20 recommendations.

     "18.  Determine proper measures to collect and reuse waste oils from
U.S. vehicular and boat service stations."

     This is something that has probably been beaten to death in terms of
rerefining of lube oils; nevertheless, from the amount produced each year
versus the amount that is used, even in cars that are as smoky as mine there
are volumes of lubricating oil that are not reclaimed and not reused—that
are dumped in places damaging to the environment.

     "19.  Design antipollution devices and management controls to insure
proper storage of crude oil."

     As mentioned earlier, those tank bottoms will fail, transfer lines will
leak, and hoses will rupture.  I cannot say that there hasn't been a great
deal of progress in 10 years in this particular area; but these excursions
still occur from time to time with great fanfare and headline publicity
usually connected with a tanker loss.  Finally, this is the real "catch 22"
that will occupy much of your time this week—

     "20.  Assess toxicological aspects to man and to warm-blooded animals
ingesting oil and oily substances."

     That was a rather obvious one, wasn't it.  Looking back now it's rather
obvious to see the wisdom in that recommendation, but we're just now starting
to scratch the surface of it in reality.  Admittedly, some of these may have
been limited by the lack of analytical equipment.  Some 20 or 25 years ago,
the mass spectrometer was the answer to "on line" process control equipment
for refineries, but that didn't work too well.  Then the gas liquid chromato-
graph came along and everybody dumped their mass spectrometers and bought
GLC's and hooked them up as process control instruments, and they worked, but
not too well.  Finally, we have third generation interfaces between the two
instruments which we are utilizing to separate and identify these relatively
minute traces of organics in treated wastewater effluents.

     A few examples of the conclusions in completed extramural reports include:

"Fluid Bed Incineration of Petroleum Refinery Wastes" - American Oil Co.

     1.  The fluid bed incineration process has been demonstrated to
     be practical and effective for the disposal of petroleum refinery
     generated spent caustic and oily sludge.

     2.  The process creates no atmospheric pollution problems,
     emitting only carbon dioxide nitrogen and water vapor.  The odor
     of the off gas has been described by various observers as being
     slight to non-existent.


                                      26

-------
    3.  The ash produced  contains sodium carbonate,  sodium sulfate,
    other  soluble  inorganic salts and inert material such as sand,
    clay,  rust, etc.

    4.  Operations are frequently prematurely terminated by the
    loss of bed fluidity  caused by excessive bed particle size.
    The particle size growth rate can be controlled  by including
    salient design features and by exercising proper operating
    techniques.

"Oily  Waste Disposal By Soil Cultivation Process" - Shell Oil Co.

    1.  Disposal of oily  sludges (hydrocarbon) by microbial
    action in cultivated  soil has been demonstrated  at prevailing
    soil and  climatic conditions at Deer Park, Texas.

    2.  Three simultaneous experiments with three oils, i.e.,
    crude  oil, bunker C fuel oil, and waxy raffinate oil, indi-
    cated  decomposition rates for the three oils to  be approxi-
    mately equal and averaged about 0.5 pounds of oil per cubic
    foot per  month without adding nitrogen and phosphorus
    nutrients and  about 1.0 pound per cubic foot per month when
    fertilizers were added.  This is equivalent to about 70 bar-
    rels of oil per acre  per month.

     3.   Cost  of the soil  cultivation process based on the demon-
     stration  project expenses and a disposal rate of 70 barrels
     of oil per acre was $7.00 per barrel of oil.  Assuming oily
     sludges and waste materials contain 33 percent oil, the
    disposal  cost  by the  soil cultivation process would be about
     $3.00  per barrel.

     4.  An optimum fertilization program appears to  be a) the
     initial addition of chemicals, if needed based upon soil test
    results,  to attain a  slight excess of nitrogen,  potassium,
    and phosphorus, and b) test at regular intervals, once per
    month, for ammonia and nitrate contents of the soil and add
     small  dosages  of ammonium nitrate as needed to maintain a
    positive  test  result  (10-50 ppm) for ammonium and/or nitrate
    contents.

    5.  The major  species of microorganisms present  are members of
     the genus Pseudomonas, Flavobacterium, Nocardia, Coryne-
    bacterium, and Arthrobacter.  The nature of the  hydrocarbon
    substrate did  not appear to influence the type of organisms
    present but did affect the number of bacteria in the soil.
    Crude  oil tank bottoms produced the highest count, waxy
    raffinate oil  produced an intermediate count, while bunker C
    fuel oil  exhibited the lowest microbial population.  Temp-
    erature appeared to have no effect upon the microbial count
    and distribution.  Addition of fertilizer did not affect the
    microorganism  distribution but appeared to be directly re-
    lated  with the total  aerobic count.

                                      27

-------
     6.   Oil decomposition rates were low when the concentration of
     oil in the soil approached the starting condition of 10 percent
     oil in the soil.  Also, the low reaction period coincided with
     the winter months and low temperature period.

     7.   Both aromatic and saturated hydrocarbons were reduced with
     time in the soil for crude oil tank bottoms and bunker C fuel
     oil.  Only the saturate fraction of waxy raffinate oil appeared
     to be reduced by soil microbial action at conditions existing
     during the project period.

     8.   Infrared and gas chromatographic analyses of the oil added
     to and extracted from soil indicated a) the absence of organic
     acids in oils added to the soil and the presence of organic acids
     in each of the extracted oils (oils from plots that were ferti-
     lized showed higher concentrations of organic acids), b) the
     organic acid increase coincided with a decrease in total satu-
     rates, and c) the percent weight boiling less than 500°C gener-
     ally was lower for the oil extracted from the soil at the finish
     of the project than for either the oil added to or the oil ex-
     tracted from the soil at the start of the project (the lowest
     percent weight boiling up to 500PC was extracted from soil which
     had received the largest quantity of fertilizer materials).

     9.   Oil and fertilizer chemicals did not infiltrate vertically
     into the soil at the test location and condition.

     10. Rainfall runoff water contained 30 to 100 ppm oil.  This
     oil appeared to be essentially naphthenic acids based upon
     infrared inspection of oil fractions.  Also, rainfall runoff
     water contained ammonia (nitrogen nutrient) approximately pro-
     portional to the excess ammonia content of the soil.  Phosphorus
     and nitrates were not found in runoff water.

     11. Oil and nutrient contents of rainfall runoff water from
     the soil cultivation process can be relatively high, and this
     discharge water should receive treatment before entering public
     waterways.

"Refinery Effluent Water Treatment Plant" - Atlantic Richfield

     1.   Reduction of Chemical Oxygen Demand (COD) content in re-
     finery wastewater effluent has been demonstrated to be feasible
     by using activated carbon as an adsorbent.

     2.   The system performed well in that it demonstrated an
     ability to start up and shut down without delay or difficulty.
     This gives the process a distinct advantage over biological
     units for use in handling intermittent rainfall.

     3.   During the two-year project the unit was operated at an
     overall average cost of 49 cents per 1000 gallons of water
     treated or 24 cents per pound of COD removed from the

                                      28

-------
    effluent wastewater.   The  first  year's operational costs were
    62 cents per 1000  gallons  of water treated,  or 34 cents per
    pound of COD removed.  After improvements over the first
    year's operation,  the  costs during the second year were reduced
    to 40 cents per  1000 gallons of  water treated, or 18 cents
    per pound  of COD removed.

    4.  The plant  demonstrated excellent reliability.

    5.  The carbon adsorption  plant  has demonstrated that when the
    feed COD is controlled to  an average of 233  ppm, the plant can
    treat refinery water using 8.5 pounds of carbon per 1000 gal-
    lons treated and reduce the effluent COD to  an average of 48
    ppm with a high  level  of 95 ppm.

"Chemical Coagulation/Mixed-Media Filtration of Aerated Lagoon Effluent"
American Oil Co.
    1.  Chemical coagulation/mixed-media filtration has been demon-
    strated to be  an effective tertiary water treatment for accom-
    plishing further reductions in suspended solids, oil content
    and biochemical  oxygen demand  following a refinery end-of-pipe
    treatment  sequence consisting  of an API separator and aerated
    lagoon.

    2.  The most significant factors influencing the treatment
    effectiveness  in the scalping  mode operation are incident con-
    taminant concentration,  the applied chemical treatment level,
    and seasonal aerated lagoon conditions.

    3.  Backwashing  with lagoon water will keep  the filter media
    satisfactorily free of oil and waste accumulation, and return-
    ing the backwash waste to  the  lagoon can be  handled in a way
    to avoid unmanageable  sludge buildup.

    4.  The filtration facilities  require a minimum amount of
    operator attention and generate  no objectionable wastes or
    odors.

    5.  The unit operation can be  demonstrated easily, appearance
    of the effluent  is greatly improved and the  facility there-
    fore has considerable  public relations appeal.  Both the
    process and the  results are easily visible and comprehensible.

    6.  Unless adequate hydraulic  controls are provided, the filter
    media can  be physically disturbed, rendering it less effective
    than designed.  Safeguards are necessary to  prevent such
    distrubances.

    7.  Optimized  operation to achieve near potable water quality
    clarity requires response  to the required water chemistry
    for destabilization of colloidal material.  This response may

                                     29

-------
     require taking any, or all, of the following into account:
     1) pH control; 2) two or three chemical destabilization
     systems, and 3) reduced hydraulic loading at water tempera-
     tures less than 15.5°C (60°F).

     8.  Brackish water has a major impact on the effectiveness
     of destabilization chemicals.

     9.  Brackish water interferes with direct usage of zeta
     potentials for determining optimal chemical pretreatment
     because the reduction of zeta potential by double layer
     repression must be detected.  This is readily achieved by
     determining zeta potentials on diluted samples.

     10. Colloid entrapment and double layer repression are
     destabilization mechanisms to be sorted out and avoided for
     direct filtration.

     11. Charge neutralization and bridging are required destabi-
     lization mechanisms for optimal filter performance.

     12. Weakly anionic polyelectrolytes are much more effective
     than nonionics for filter aids in a three chemical system.

     13. Even with optimized chemical pretreatment, filter loading
     must be decreased with decreasing water temperatures.

     14. Recommended filter hydraulic loadings are proportional
     to the viscosity of water.

     15. Incorporation of powdered activated carbon up to 150 mg/1
     had no favorable impact on the destabilization chemistry of
     biocolloids.

     In-house reports and papers produced by program personnel include:

"Analytical Variability' of Five Wastewater Parameters - Petroleum Refining
Industry" - Petroleum-Organic Chemical Wastes Section, Robert S. Kerr Environ-
mental Research Laboratory.

     1.  The chemical oxygen demand (COD) test had a repeatability
     expressed in terms of standard deviation of 9.5 milligrams per
     liter (mg/1) for petroleum,refinery wastewater which had a COD
     average of 134 mg/1.   Reproducibility for this same refinery
     wastewater exhibited a standard deviation of 15.0 mg/1.

     2.  Suspended solids with an average concentration of 19 mg/1
     had a standard deviation for repeatability of 1.8 mg/1 and a
     standard deviationifor reproducibility of 5.2 mg/1.
                                      30

-------
     3.  Results  obtained for the ammonia test with an average
     concentration of 8.5 mg/1 exhibited a repeatability standard
     deviation of 0.1 mg/1 and a reproducibility standard deviation
     of  0.9 mg/1.

     4.  The  repeatability standard deviation for phenolics was
     0.2 mg/1 for a sample containing 5.5 mg/1; the standard
     deviation for reproducibility of phenolics was 0.8 mg/1.

     5.  Oil  and  grease standard deviation for repeatability was
     2.3 mg/1 and reproducibility was 2.9 mg/1 for a sample contain-
     ing approximately 11 mg/1.

     6.  There does not appear to be any major differences in con-
     centration between the results of hexane extraction and freon
     extraction procedures.

     7.  The  variance of the analysis for oil and grease appears
     to be less for the freon method than the hexane method.

     8.  A better than 95 percent spike recovery for COD and
     ammonia  was  achieved.

     9.  A comparison of results between Phases I and II indicate
     the instruction seminar, which was held to achieve uniformity
     of analytical procedures accomplished:

        a. A significant reduction in arithmetic and extreme
        outlier value errors;

        b. Enhancement of uniformity of laboratory technique;
                                             '•'<*.  -j
        c. Minimizing the COD mean values between intra-
        laboratory and interlaboratory results;
        d. Improvement of spike recovery for COD and ammonia.

     10. The  standard deviations for COD, ammonia, and phenolics
     were  decreased between Phase I and Phase II.
                                                     >,

     11.  Environmental Protection Agency (EPA) methodology for the
     parameters studied appeared applicable for petroleum refinery
     wastewater when the analysts were properly instructed.

"Acute Toxic  Effects of Petroleum Refinery Wastewaters on Redear Sunfish"
John E.  Matthews  and Leon H. Myers, Robert S. Kerr Environmental Research
Laboratory.

     1.  Short-term static bioassays of 24-hours' duration can be
     an effective tool for screening industrial process wastewaters
     to locate sources of toxic agents; these tests can also be
     used  to  evaluate effectiveness of industrial waste treatment
     processes.
                                     31

-------
2.  Static bioassays cannot be used to obtain reliable 50 per-
cent tolerance limit (TL5Q) values for low toxicity wastes
that exert a high oxygen demand although a range can often be
established by an experienced observer.  Dissolved oxygen (DO)
usually becomes critical at about 12 hours; low DO plays a maor
role in a mortaility after this time.  Activated sludge treat-
ment processes tend to reduce the oxygen demand beyond the
critical stage.

3.  Raw wastewaters from different oil refineries vary greatly
in their toxic characteristics; wastewaters from different
processes within a single refinery also vary greatly in their
toxic characteristics.  Toxicity of oil refinery wastewaters
varies considerably at different treatment stages in the
activated sludge process; toxicity appears to decrease follow-
ing each stage of treatment.

4.  Toxicity of oil refinery wastewaters cannot always be
predicted from results of chemical analyses; the toxic effect
of the waste is dependent on the synergistic or antagonistic
activity of toxicants present.

5.  The most common toxic constituents of untreated oil
refinery wastewaters are:  ammonia, sulfides, phenolic
compounds, and cyanides.  Raw wastewaters also may contain
other toxic compounds including various hydrocarbons.

6.  Due to the volatile and unstable nature of some toxic
components of oil refinery wastewaters which may have led
to a reduction in concentrations during sample transporta-
tion, storage, andhandling, TL^Q values obtained during these
tests may be higher than the actual value.

7.  Acute toxic effects of raw wastewaters from oil refineries
are generally exerted within the first 12 hours of the static
test; therefore, a 24-hour test will provide good positive
results under static conditions.  If samples containing toxic
compounds, other than those mentioned above, toxic effects may
be exerted over an extended period.  Tests should then be
continued for at least 48 hours to obtain more positive
results.

8.  Toxicity of the final clarifier effluent from oil
refineries with activated sludge treatment systems is depen-
dent on toxic constituents present in the influent and their
concentrations.  Results of chemical analyses conducted dur-
ing these tests indicate that concentrations of most toxicants
other than ammonia are reduced by activated sludge treatment
systems and at least a four-fold decrease in toxicity can be
expected after treatment.
                                 32

-------
     9.   Although activated sludge treatment systems have very
     little effect on ammonia concentrations, the toxic effect
     of  ammonia is lessened due to a decrease in pH of the treated
     effluent.

     A listing of some of the current funded research may be of interest:

"Identification of Refractory Organic Compounds from Treated Refinery Waste-
waters"  - ERDA, Division of Environment & Safety, Washington, D.C.  Project
Site - Argonne, Illinois

"Cyanide Removal from Petroleum Refineries" - Illinois Institute of Technology
Research Institute, Chicago, Illinois.

"Refinery Sour Water Stripper Studies" - American Petroleum Institute,
Washington, D.C.  Project Site - Ponca City, Oklahoma.

"Powdered Carbon-Activated Sludge-Filtration Processes for Petroleum Refinery
Wastewater" - Atlantic Richfield Co., Harvey, Illinois.
                                                 . <*
     In closing, the lesson which I hope we^have,learned from this recap-
itulation of the Laboratory's research program history is that the list of
priority pollutants will not be an end in itself.  We should apply the
available technology to its utmost to determine what is in an effluent,
where it came from, and how to remove or control it, if it is present in
environmentally significant amounts.  Such approach may avoid or at least
ameliorate future crises.

     Despite having disposed of soothsayers earlier, the urge to predict is
too  strong to resist.  Try these with respect to the priority pollutants
list.

     1.  Some compounds on the list ain't gonna be found in refining effluents

     2.  Some compounds not on the list will be found in refining effluents

     3.  If you continue to search long enough and hard enough for some-
thing you may find it anywhere—even though it ain't really there.  Now
ponder on that one a while.
                                      33

-------
 DISCUSSION

 Paul Mikolaj,  Lion Oil Company;  Is there any research work being done on radiation
 techniques to treatwastewater?

 Marvin Wood;   There is, of course, some being done, but none that I am aware of or
 at least none that we are funding in this specific area of refinery wasfes.  There  is
 some being done in Hie case of other wastes—domestic and possibly a few cases of
 industrial.  It might involve come coblat 60 or something sources or  radiation to change
 the physical characteristics to make the wastes more amendable to handling.

 BIOGRAPHY

           Marvin L. Wood holds a B.S. in Chemical
 Engineering from the University of Arkansas and an
 M.S. 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.
                                     34

-------
                            SESSION II


'PANEL DISCUSSION: RESEARCH SUPPORTED IN THE PRIVATE SECTOR"


                      Chairman

                     * Milton R. Beychok

                      Consulting Engineer
                      Irvine, California


                      Panel Members
                      Robert T. Denbo
                      Exxon
                      Baton Rouge, Louisiana
                      Kent G. Drummond
                      Marathon Oil Company
                      Findlay, Ohio
                      Francis S. Manning
                      University of Tulsa
                      Tulsa, Oklahoma
                      David C. Bomberger
                      Stanford Research  Institute
                      Menlo Park, California
                      Judith G. Thatcher
                      API/DEA
                      Washington, D.C.


                     *Biography on Page 300

                                35

-------
           "DISCUSSION OF RESEARCH RELATED TO PETROLEUM REFINERY
                WASTEWATER SPONSORED BY THE API COMMITTEE
                    ON REFINERY ENVIRONMENTAL CONTROL"

                                  Robert T. Denbo
                     Coordinator of Environmental Control, Exxon
                                Baton Rouge Refinery

       Mr. Drummond's and Ms.  Thatcher's papers discussed the organization of the
Division of Environmental Affairs of the American Petroleum Institute and research
sponsored by the Division of Environmental Affairs.  This paper will cover the research
program sponsored by the Committee on Refinery Environmental Control (CREC).

       First, some added background on the API.   The API was incorporated  in 1919.
The objectives were set forth as follows:

       -  afford a means of cooperation with the government in all matters of national
          concern.

       -  foster foreign and domestic trade in American petroleum products.

       -  promote, in general, the interest of the petroleum industry in all its branches.

       -  promote the mutual improvement of the members and the study  of the arts and
          sciences connected with the  petroleum industry.

       Next,  it may be appropriate to put the CREC Committee in proper prospective in
the American Petroleum Institute.  The first chart shows the  organizational setup in API
down to the divisional level.

       The second chart shows the relationship of the CREC Committee to the General
Committee of the Division of Refining.   The CREC Committee was set up  to have the
following responsibilities:

       - the development and publication of the API Manual on Disposal of Refinery
         Wastes which covers atmospheric emissions, solid  wastes and liquid wastes.

       - the sponsorship of research related to refinery pollution control.
                                       36

-------
AMERICAN PETROLEUM  INSTITUTE
               BOARD OF
               DIRECTORS

1
ADMINISTRATION


MANAGEMENT
COMMITTEE
1

INDUSTRY
AFFAIRS



BUDGET ADVISORY
COMMITTEE

1
COMMITTEE ON
PUBLIC AFFAIRS



COMMITTEE ON
POLICY
DEVELOPMENT
1
DIVISIONOF
ENVIRONMENTAL
AFFAIRS

1
DIVISIONOF
.REFINING
ICOSH





DIVISIONOF
PRODUCTION


1
DIVISIONOF
MARKETING


1
DIVISIONOF
FINANCE AND
ACCOUNTING

1
DIVISIONOF
TRANSPORTATION


1
DIVISIONOF
MEDICINE AND
BIOLOGICALSC.
                37

-------
DIVISION  OF REFINING
INTERDIVI SIGNAL
COMMITTEE ON
OCCUPATIONAL
SAFETY AND HEALTH


	
GENERAL
COMMITTEE

COMMITTEE ON
LIQUEFIED
HYDROCARBON
GASES

COMMITTEE ON
REFINERY

ENVIRONMENTAL
CONTROL

COMMITTEE ON
REFINERY
EQUIPMENT

COMMITTEE ON
RESEARCH, DATA
AND INFORMATIO
SERVICES
N












OPERATING
COMMITTEE

COMMITTEE ON
TRAINING AND
DEVELOPMENT

OPERATING
PRACTICES
COMMITTEE

PETROLEUM
PRODUCTS
COMMITTEE


          38

-------
       -  the dissemination of information on legislative developments concerning stream
          and air pollution and solid waste disposal.

       It is within this context that the R&D related to petroleum refinery waste water is
carried out.  During the past few years, CREC has worked closely with the API Division
of Environmental Affairs and participated actively in many of the projects of the Water
Quality Committee.  CREC's own research and development and related studies have
resulted In the publication of the following reports:

       -  the volumes on liquid wastes including:

          o  separator design parameters
          o  biotreatment design parameters
          o  stripping
          o  extraction
          o  adsorption
          o  and other areas

       -  a filtratfon volume

       -  a sour water stripper volume.

The following projects are currently in the planning stage tentatively for 1978:

       -  Enhancement of biological treatment -  Phase III

       -  Studies of handling of storm runoff from petroleum refineries

       -  Studies of variability of treated effluents

       -  Benthic assays of treated refinery effluents.

       The following two projects will be discussed briefly in this paper for purposes of
illustration,  how research and related studies are developed in the CREC Committee:

       -  Sour water stripping

       -  Bioenhancement

Sour Water Stripping

       Sour or foul water in petroleum refining is process water containing organics that
also contains more hydrogen sulfide and ammonia than can be handled effectively in a
biological treatment system.  The types of organics present in this stream include phenolic
type compounds and other objectionable materials.  This  type of water  is probably the
most difficult to handle of all process waste waters in most refineries.  It is usually

                                         39

-------
produced in such operations as a steam condensate from catalytic cracking, coking, hydro-
cracking and other hydroprocessing.  Effective removal of the hydrogen sulfide and
ammonia are essential to allow meeting limits on these two compounds and to permit
effective biological oxidation of the phenols and other organics.

        Ron Gantz of Continental Oil has been the guiding hand in the CREC-sponsored
studies by contractors to improve the efficiency of removal of these two constituents by
steam stripping which is the existing state of the art.  The current program of study has
extended over a period of more than three years and has been, in part,  a joint effort by
EPA and API CREC. The EPA Water Labs at Ada, represented by Leon Myers, played a
significant role in the development of this work. The program has consisted of extensive
surveys of actual plant operations to determine state of the art in the industry, plant
testing to obtain  actual operational data, the operation of pilot facilities in laboratories,
and  the correlation of all the results into a comprehensive, new equilibria calculation
procedure.

        The results from these studies are to be published soon. The most significant
findings include:

        -  the identification  of ammonia that cannot be removed by stripping from the
          water as produced - so called "fixed ammonia". This fixed ammonia
          apparently will not come out of solution by stripping if it exists as ionic
          ammonia in combination with carboxylic acids and other such compounds when
          slightly acidic.  It was found that the presence of a slight excess of sodium
          hydroxide over the stoichiometric amount will allow stripping to proceed
          efficiently.

        -  another significant result has been the development of a new equilibria calcu-
          lation  procedure.  New equations have been developed and tested which give
          much more accurate results and can be used  more flexibly than existing
          accepted approaches.  The procedure exists  as a computer program which will
          permit calculations including the effect of pH adjustments at selected points in
          the stripper tower. This is not possible in the existing sour water stripping
          computational approach.

Bioenhancement
       The CREC study of bioenhancement will be covered in some detail in Session VIII
of this open forum by Lyn Crame of Texaco.  Consequently, it is not appropriate to go
into the findings of this study in this paper.

       However, it is pertinent to say that this study was started about a year ago prior
to the availability of initial results from other API studies which were designed to deter-
mine the presence of residual organic compounds  in refinery effluents after biological
treatment.  By "residuals" is meant organics present in the effluent which were introduced
in the raw waste load feed to biological treatment.  Initial results of the study indicate

                                         40

-------
only a small amount of residuals are present in biologically treated effluent.  Apparently
the vast majority of organics in the treated effluent are metabolities, humic acids, and
other compounds associated with the biodegradation process.  Almost no objectionable
compounds have been found in biologically treated effluents. At this point it would not
appear that it is practical to require further treatment on the basis of results of con-
ventional  analyses or  even results on the basis of analyses for the more exotic list of toxic
compounds.

Conclusion

       The CREC program for the past few years has been consistent with the topics dis-
cussed at  this open forum. The agenda of this open forum suggests the high degree  of
sophistication that is  employed in the treatment of refinery wastewater.  It can also be
said that the  performance of treatment facilities for petroleum refineries suggest that the
current quality of treated wastewater from  these refineries is no longer a significant
problem in water pollution abatement.  It appears that technology and research in this
area has advanced to the point where it is  well ahead of other areas of environmental
concern for the industry.
BIOGRAPHY            Robert T. Denbo

        Robert T. (Bob) Denbo is the coordinator of
environmental control at Exxon's 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 University  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 environmental control for all domestic
refineries.  He is chairman of the American Petroleum
Institute Committee on  Refinery Environmental Control,
member of the API Division of Environmental Affairs,
former president of the  Louisiana Water Pollution
Control Association,  member of the Technical Advisory
Committee to the Louisiana Air Control Commission,  and
member of the Air Pollution Control Association.  Mr.
Denbo has served on a number of panels related to develop-
ment of information on  target values for U.S. refinery
water effluents.  He  has served on a joint  U.  S.-Soviet
planning group for cooperative effort in refinery water
pollution  controls.

                                         41

-------
                         WATER QUALITY COMMITTEE OF API

                                  Kent G. Drummond
                 Technical Coordinator, Environmental Control Div.,
                                Marathon Oil Company


       The Water Quality Committee is one of eight standing technical  committees under
the General Committee on Environmental  Affairs (see Figure 1).  The General  Committee
Is composed of 28 oil company environmental coordinators.  Under the General Committee
are 8 standing committees which have over 50 task forces involving more than 300 oil com-
pany personnel.

       These committees meet on a regularly scheduled basis of twice a  year; once  in
the spring to review current projects and to generate new projects for the forthcoming
year and again in the fall to review current projects.   The spring meeting  is the normal
time for proposing the next year's budget.

       In the case of the Water Quality Committee, if a project is accepted by our
committee, it  is  then proposed as part of the committee's budget, to the Planning
Budget Advisory  Committee which is under the General  Committee on Environmental
Affairs.  Incidentally, each standing technical committee has a liaison representative
from the  Planning Budget Advisory Committee.

       If PBAC approves the project, it is then submitted to the parent Environmental
Affairs Committee for consideration.  The project then becomes part of the proposed
budget of the Environmental Affairs Committee to be considered by the API  Finance
Committee and,  if approved,  becomes a funded project  for the new year's budget.

       Figure 2  lists the members' company affiliation on the Water Quality Committee,.
Of the 18 members on the Committee, only 2 companies have 2 representatives and this
is because their respective representatives are either chairmen of  more than one task
force or are a representative from the Committee on Refinery Environmental Control.

       There are 3 main project areas of interest as shown in Table 1.  The Water
Quality Committee is primarily interested in the quality of water as it leaves the
premises.  We also become involved in projects which need immediate attention.  At
other times, we get into projects which are beyond the monitory scope of the various
divisions.
                                        43

-------
       Later on this afternoon, Mrs.  Judith Thatcher will discuss 2 of the projects which
come under the Water Quality Committee.

       In 1977, we had 10 active projects.  These projects are listed in Table 2. Their
names and the budgetary allocations are given.  Our original budget of $240,000 was
approved in December, 1976, however, because of the changes in government regulations,
we came back with  the supplementary budget of $450,000 in March, 1977.  Project W-14
and the last 4 items on Table 2 were approved at that time.

       Our 1978  proposed budget came to a total of $515,000.  This is shown in Table 3.

       For the past five years, the budgetary allocations for the Water Quality Com-
mittee have continued to grow.  Back in 1974, we had a budget of only $60,000; whereas
in 1978, our proposed budget is $515,000 plus 40% of $450,000 which is $180,000;
making a total of $695,000 for the year of 1978.  This is all shown on Figure 3.

       In the past six years, the Water Quality Committee has attempted to learn
more about our own refinery operations.  With this better understanding, we have been
able to defned ourselves against unreasonable regulations and to discuss with EPA regu-
lations which we feel attainable.

       The five projects which we have proposed for 1978 will help us to better under-
stand some of the  problems within our refineries and hopefully, this will lead to guide-
lines in 1983 which are attainable and will give best valuable  treatment economically
achievable.
 BIOGRAPHY      Kent G. Drummond

       Kent G. Drummond  is Technical Coordi-
 nator in the Environmental Control Division of
 Marathon  Oil Company, Findlay, Ohio.  He
 has a B.S. degree in Civil Engineering and a
 M.S. degree in Sanitary Engineering from Iowa
 State University.  He is currently Chairman of
 API's Water Quality Committee and  Chairman
 of the 1979 Oil Spills Conference.
                                        44

-------
                                  TABLE 1

         WATER QUALITY COMMITTEE PROJECT AREAS OF INTEREST

         1.  Quality of wafer leaving premises
         2.  Projects which need immediate attention
         3.  Projects beyond monetary scope of divisions



                                  TABLE 2

                WATER QUALITY COMMITTEE 1977 PROJECTS

 No.                               Title                          Budget $k
 W-12          Bioassays of Refining Effluents                        $  90k
 W-14          Toxic Pollutant Effluent  Standards                    $  25k*
 W-15          BPT Vs BAT Petroleum Refining Guidelines            $100k
 W-19          Non-process Effluent Standards
 W-20          Analysis of Refinery Effluents (BPCTCA)               $  50k
 W-21          Amendments to 92-500
 W-22          Sampling & Testing Protocol for Toxic Pollutants       $200k*
 W-23          Refinery Questionnaire (EPA) Analysis                 $100k*
 W-24          Refinery Effluent Pollution Contribution in Perspective  $100k*
 W-25          Economic Study  BPT to BAT                          $  25k"
,*
                                            Total                 $240k $450k*

               *Approved March '77



                                  TABLE 3

            WATER QUALITY COMMITTEE 1978 PROPOSED BUDGET

 No.                               Title                         Budget

 W-12P (cont'd)  Bioassays on Refiner/ Effluents                      $ 40k
 W-22P (cont'd)  Sampling & Testing Protocol for Toxic Pollutants      $200k
 W-23P (cont'd)  Refinery Questionnaire Follow up                   $100k
*W-26P (new)    Removal of Toxics from Refinery Effluents - Pilot      $ 75k
                Plant (Joint Study)
*W-27P (new)    Evaluation of Selected Hazardous Refinery           $100k
                Chemicals to F.W. Organisms                      	
                                            Total                 $515k

                * Projects to be carried out if significant quantities
                are found in effluents.
                                     45

-------
    COMMITTEE ON ENVIRONMENTAL AFFAIRS
          I
        PBAC
GOVERNMENT
   LIAISON
                  COMMITTEES

        1.  Fate & Effects of Oil
        2.  Oilspill Prevention & Control
        3.  Mobile Source Emissions
        4.  Environmental Economics
        5.  Stationary Source Emissions
        6.  Solid Waste Management
        7.  Air Quality
        8.  Water Quality

ENVIRONMENTAL AFFAIRS ORGANIZATION CHART
             API Environmental Affairs

                    Figure 1
           WATER QUALITY COMMITTEE
           MEMBERSHIP BY COMPANY

Chairman - Marathon

Vice Chairman - Standard of Indiana

Members

Chairman of CREC - Exxon Refining

Chairman of Liquid Wastes - Union Oil

Task Force Chairman
  1. Exxon Research & Engineering
  2. Union Oil Co.
  3. Gulf Oil Co.                    Others
  4. Standard Oil of Ohio               1.
  5. Shell Oil Co.                     2,
  6. Shell Oil Co.                     3.
  7. Sun Oil  Co.                       4.
  8. Mobil Oil Co.
    Atlantic Richfield Co,
    Getty Oil Co.
    Texaco, Inc.
    Chevron Oil Co.
 5. API Representative
                    Figure 2
                        46

-------
$600k-i
$500k
w $400k
DC
5
0 $300k
O
$200k
$100k
0









o I
to I


to
U)
M




S
CM





|
(O
O
CM







O
U)
u> to
5 U)




1974
1975
1977   1978P
           1976
           YEAR
WATER QUALITY COMMITTEE
 PAST BUDGETS 1974 - 1978P
       Figure 3
           47

-------
                OVERVIEW OF RESEARCH ON PETROLEUM REFINERY
                       WASTE WATERS AT U.S. UNIVERSITIES

                                 Francis S. Manning
                          Professor of Chemical Engineering
                                 University of Tulsa
      This overview addresses research at United States universities not covered by the
previous papers.  In other words, research sponsored by federal agencies such as the
E.P.A.; the A.P.I.; and research institutes has been summarized elsewhere and will not
be repeated.  Obviously it is impossible to canvas every potential department in every
university and hence the following list will be incomplete.  The author apologizes to
those researchers whose efforts have been omitted.
                                        i
      This overview will not discuss the  following studies because they are described else-
where in this symposium: -

      o        the Brigham Young study of liquid-vapor equilibrium of H«S-NH«-CO«
               which is reviewed by Milton Beychok in Session VI

      o        the University of California at Berkeley work on stripping of organics using
               volatile solvents - see Marvin Wood in Session I

      o        the University of Texas at Austin investigation of the effect of temperature
               on critical sludge age as this is described by Davis Ford in Session V

      o        the University of Delaware experiments on PAC in biological treatment
               which is covered by Francis Robertaccio in Session VII.

      This survey uncovered three long-term  research programs which, surprisingly, have
many common features. All three are sponsored by groups of petroleum and petroleum
related  industries and all three address very specific research objectives. These three
programs at Louisiana State University, Oklahoma State University, and the University  of
Tulsa are now described.

                           LOUISIANA STATE UNIVERSITY

      The Petroleum  Refiners Environmental Council of Louisiana (PRECOL) group has a
long history of financial support for environmental research in the Zoology and Physiology


                                        48

-------
Department at LSD.  Initial support occurred in 1946 when the Louisiana Petroleum
Refiners Waste Control Council provided two fellowships for the determination of
immediate and accumulative toxicity of refinery wastes to fish.  Participating companies
included Chalmette Petroleum Corporation, Cities Service,  Continental Oil Company,
Pan American Petroleum Corporation, Shell and Standard Oil.

      Bioassay techniques were developed in the 1950's and pesticide toxicity work
became the focal point of the laboratory during the 1960's.   Emphasis shifted to toxicity
studies on the effects of crude oil, emulsifying agents and interactions between them on
aquatic fauna during the early 1970's.  The organizational name was changed to PRECOL
in 1973.  The community structure of the  Calcasieu estuary began  in 1974 along with an
investigation of heavy metal movement through the estuary.  Two new projects were
initiated  in 1977.  L.S.U.  is investigating the accumulation of heavy metals and organic
pollutants by water snakes and benthic invertebrates from the Mississippi River near the
Baton Rouge industrial complex.  The effects of low oxygen tension on the survival and
metabolic rate adjustments of aquatic fauna are also being  investigated.

      Financial support  for  research efforts has continuously grown as has the number of
oil companies providing it.   The PRECOL group now consists of: Cities Service Oil
Company, Continental Oil  Company, Exxon Company, Gulf Oil Company, Murphy Oil
Corporation, Shell Oil Company, Tenneco Oil Company and Texaco, Inc.  The L.S.U.
laboratory has always enjoyed a close working relationship with the Louisiana Division of
Water Pollution Control  which provides technical support for graduate students.

                          OKLAHOMA STATE UNIVERSITY

      Since 1956 the Oil Refiners Waste Control Council of Oklahoma has sponsored bio-
logical investigations on petroleum refinery waste waters in Oklahoma. This work,
originally directed by Professor Troy  Dorris and directed since 1970 by Professor S. L.
Burks, is currently studying:

      o       The Effects of Residual Volatile Toxins in Oil Refinery Waste Waters

      o       The Biological Evaluation of BPTCT and BATCT for Refinery Waste Waters

      o       The Development of Benthic Bioassay  Techniques.

At present the Oil  Refiners  Waste Control  Council of Oklahoma includes the following
members:  APCO,  Champlin, Conoco,  Hudson Oil  Company,  Kerr-McGee, OKC
Refining Company, Sun  Oil, Texaco and  Vickers.

      The recent biological  evaluation of activated sludge, sequential activated sludge-
dual media filtration (equivalent to Best Practicable Treatment Control Technology,
BPTCT) and sequential activated sludge-dual media filtration-activated carbon  adsorption
(equivalent to B_est Available Treatment Control Technology) at the ETU refinery showed
that;

                                        49

-------
      1.       Activated sludge and BPTCT would not produce a non-toxic effluent.

      2.       BATCT produced an effluent which did not cause any significant deleterious
              effects upon fish, benthic macroinvertebrates, and periphyton during 32-
              day exposures.

Chemical analyses of the treatment effluent streams indicated that most of the toxicity was
associated with the organic substances as measured by TOC and COD analyses.

      A new bioassay method for determining the effects of environmental contaminants on
populations of benthic macroinvertebrates has been developed. Colonized Hester-Dendy
samplers were  transported from a natural stream to artificial streams and exposed to
industrial waste water.  Species diversity, number of taxa, and density of the aquatic
organisms were measured before and after selected time intervals of exposure.  A 30 and
a 32-day continuous-flow exposure test with the benthic macroinvertebrates showed that
activated sludge treated petroleum refinery wastewater caused a greater decrease in
species diversity than the sequential activated sludge-dual media-activated carbon
treated effluent.  The effect upon number of taxa and mean density of individuals was
even greater.  This procedure has permitted exposure of several species of aquatic
invertebrates to the test solution and thus measured the effect upon pollution sensitive and
tolerant organisms.

                               UNIVERSITY OF TULSA

      The University of Tulsa Environmental Protection Project (UTEPP) began January 1,
1974.  The UTEPP program, which is directed by Professor Nicholas D. Sylvester of
Chemical Engineering, has received support from the following companies: Ameron,
Amoco Production Company, Aramco, Bechtel Corporation, Calgon Corporation, Chevron
Research Company, Crest Engineering, Compagnie Francaise des Petroles, Dow Chemical,
U.S.A., Exxon Production Research Company, Foster Wheeler Corporation, Getty Oil
Company, Iranian Oil Services, Ltd., Lummus Company, Marathon Oil Company, Mobil
Research and Development Corporation, and Pullman Kellogg.

      Currently the program includes the following projects:

      1.       Chromate Removal From Cooling Tower Slowdown
              Commercially available ion exchange resins for chromate removal from
              cooling tower blowdown are being evaluated.  The effectiveness of
              chromate removal, the regenerability of the resins and chromate recovery
              are being studied.

      2.       Oil Removal From Wastewaters by Induced Air Flotation
              The effects of oil type, concentration, drop size and size distribution;
              process  residence time, air flowrate, bubble size; and polyelectrolytes on
              the performance of induced air flotation is being studied.
                                       50

-------
     3.       Oil Removal  From Wastewaters by Ultrafiltration
              The removal of gulf coast crude oil from wastewaters by ultrafiltration is
              being investigated.  The effects of oil type and  concentration, flowrate
              and salt concentration on oil removal efficiency are being studied.

     4.       Pollutant Removal by Adsorption and Reaction in Aqueous Slurries of
              Powdered Activated  Carbon
              The adsorption  and oxidation reaction characteristics of pollutants (SOo
              and organics; e.g.,  benzene) in aqueous slurries of high surface area
              powdered activated carbon is being studied,  In conjunction, the effects
              of the mass transport processes attendant to slurry reactors on adsorption
              and oxidation are being determined.  In addition, the catalytic activity
              and mechanisms of activated carbon towards oxidative pollutant removal
              will be elucidated.

     5.       Urea  Removal From Industrial Wastewaters
              A state-of-the-art review of urea  removal  from industrial wastewaters is
              being prepared.  Although urea is synthesized on a large scale for use in
              the manufacture of urea-formaldehyde resins,  the primary emphasis of
              the review will be to its manufacture for use as a fertilizer.

     6.       Activated Sludge Enhancement With Powdered Activated Carbon
              An experimental study is being initiated to investigate the mechanisms
              involved in activated sludge enhancement  with powdered activated carbon.
              In particular, answers will be sought to questions such as:
               (i) How much does the blocking of the carbon surface bacteria and  their
                  by-products reduce the adsorptive capacity of the carbon?
              (ii) What is  the mechanism of carbon surface renewal by the action of
                   microorganisms?
              In addition,  appropriate mathematical models of the process will be
              developed.

                                     SUMMARY

     It is hoped  that more universities will  become involved in the treatment  of refinery
wastewaters.  While  considerable progress has been made due to encouragement and
financial support  from the  U.S. EPA, API, etc.,  several questions remain to be answered.
Some key topics are: -

     o       the effect of temperature on biological degradation operating  performance
              during the  severe winter of 1976-77 has questioned the accuracy of the
              traditional Streeter-Phelps equation

     o       the effect of PAC on nitrification

     o       identifying inhibitors of biological kinetics

                                        51

-------
             dewatering sludges especially when the contributions from refinery API,
             DAF, and ASP units are combined.
                            ACKNOWLEDGEMENTS

      Professors Bud Burks; Bill Stickle and Nick Sylvester provided the descriptions of
the major programs at O.S.U., L.S.U. and T.U. respectively.

BIOGRAPHY

      Francis S. Manning is the Director of the
Petroleum Energy Research Institute (PERI) and
Professor of Chemical Engineering at the University
of Tulsa.  He holds the following degrees in
Chemical Engineering:  - B. Eng. (Hons) from
McGill University and  M.S.E., A.M., and Ph.D.
from Princeton University. He is a  professional
engineer,  registered in Oklahoma,  Pennsylvania
and Texas.  Frank taught at Carnegie-Mellon
University for 9 years before  Joining the University
of Tulsa in 1968. The author of one book and over
60 papers, Frank's current research  interests lie in
thermodynamics, reaction kinetics and industrial
pollution control.  In 1969 he received the R.  W.
Hunt Silver Medal from the AIME.
                                       52

-------
          OVERVIEW OF RESEARCH ON PETROLEUM REFINING WASTEWATERS
               AT INDEPENDENT CONTRACT RESEARCH ORGANIZATIONS

                           David C. Bomberger
            Chemical Engineer, Environmental Control Group
                           SRI International
      The major contract research organizations in the United States
were  surveyed  to determine what research is being done on the waste-
waters from petroleum refineries.  Companies that were engaged primarily
in engineering and design were not contacted.  Any research relating
to identification of priority pollutants identified in the 1976 Consent
Decree is not reported here because it will be covered by other panel-
ists.

      Very little research relating directly to refinery wastewater is
being done at contract research organizations.  Only ten ongoing or
recently completed projects were identified, and only four of these
were concerned directly with refinery wastewater.  The other six proj-
ects covered aspects of synthetic fuel production and utility water
usage.  The results of these projects could be utilized by petroleum
refiners through technology transfer.
      Most of the research funding is from government agencies, princi-
pally ERDA and EPA.  Only two projects were funded by private industry,
one by a single company and the other by the American Petroleum Insti-
tute.

      The ten research projects cover four general topics:  the cost
of effluent treatment, wastewater treatment technology, measurement of
wastewater components, and reduction of effluent volume, as summarized
below.
Cost of effluent treatment:

      "  Battelle has studied the economic impact of environmental
         regulations on the petroleum industry, for the American
         Petroleum Institute.
                                  53

-------
Treatment technology:

      •  SRI International has studied ammonia fixation in sour
         water strippers, for the American Petroleum Institute
         (EPA grant).

      •  Midwest Research Institute has examined a petroleum
         refinery treatment system, for a refinery.

      •  Gulf South Research Institute is studying the technology
         for removal and destruction of organic compounds from
         wastewater, for the EPA, the State of Mississippi, and
         private industry.

Measurement of wastewater components:

      •  Radian is conducting two projects on the development of
         sampling technology and on analysis of effluents from
         synthetic fuel production, for ERDA.

      •  Battelle Northwest is studying the wastewater produced
         in shale oil retorting, for ERDA.

      •  Radian is studying fugitive hydrocarbon emissions from
         petroleum refining, for the EPA.

Reduction in effluent volume:

      •  Radian is investigating optimization of water use in the
         utility industry, for the EPA.

      •  Radian .is also studying saline water use in energy
         facilities, for ERDA.

these last two projects are applicable only indirectly and would re-
require technology transfer.
                                  54

-------
BIOGRAPHY

       David C. Bomberger is a chemical engineer
in the Environmental Control Group at SRI
International.  He has a Ph.D. in chemical engineering
from Princeton University.  Before joing SRI  Inter-
national, he worked on environmentally related
projects for Bechtel Corporation and The  Shell
Development Company.
                                       55

-------
          "CURRENT API STUDIES OF RESIDUALS IN REFINERY EFFLUENTS"

                                 Judith G. Thatcher
                 Environmental Associate, American Petroleum Institute

         Good afternoon, ladies and gentlemen.

         Mr. Drummond has described the general  structure of API's Environmental Affairs
Department and the type of research carried out under the Water Quality Committee.  I
would like to discuss two current projects relating specifically to one of the major issues
being discussed at this forum, that is, toxic pollutants.

         In late 1974, a widely publicized study focused national attention on the
presence of a number of allegedly toxic organic compounds in New Orleans' public
drinking water.  Although the validity of the studies attempting to relate New Orleans'
drinking water to disease has been questioned, the studies understandably generated much
publicity and public concern. There are many possible sources of organics  in public water
intakes, including effluents  from industrial, chemical,  and municipal sewage plants, in
addition to surface run-off.

         Although much was known in 1974 about refinery effluents with respect to con-
ventional waste water parameters such as BOD and COD, little was known  of the nature
of the organic residuals  contributing to the concentration of these parameters in the final
discharge of treated refinery effluents. In  order to begin to fill this data gap, funds were
made available to begin work in this area.  The project was entitled "Analysis of Residuals
in Refinery Effluents"  and is under the direction of API  staff and the W-20 Task Force.

         The objective of the first phase of the work was to determine the types and
amounts of residuals in the effluent from a refinery meeting BPCTCA, and the potential
effects of these residuals on  public drinking waters. During Phase I,  samples of the intake
water and the effluent water from a Class B refinery were obtained, along with samples of
the effluent from  a municipal sewage treatment plant.  It was felt that obtaining and
analyzing samples from the treatment plant would lend some perspective in  evaluating any
residuals found in the  refinery effluent.  The types of analyses that were run on the
samples were metals, polynuclear aromatics,  organo-halides and volatile and nonvolatile
hydrocarbons.  In addition,  classical parameters such as COD, BOD, and total suspended
solids were run.  In order to determine whether any precursors of organo-halides were
present  in the refinery effluent and intake waters,  portions of the refinery intake and
effluent samples were  chlorinated and then analyzed.  The analyses of these samples were
then compared to the chlorinated municipal plant samples.

                                         56

-------
        The objective of Phase II is to determine the types and amounts of residuals in a
second BPCTCA refinery, and the removal of these residuals across various pilot plant
treatment trains.  Samples from a second  Class B refinery were taken as well as samples
from another municipal treatment plant.  The Class B refinery is the same refinery sampled
for the EPA-Argonne study.  This study will be discussed in detail on Thursday by Mr.
Fred Pfeffer of EPA's R.S. Kerr Laboratory.  In this study,  a pilot unit mixed media filter
and activated carbon column were set up at the refinery to treat a slip-stream from the
refinery activated sludge unit.  Samples of the refinery's intake water, DAF and ASU
effluents were taken, along  with samples of the two pilot units, over a four-day period.
API took samples along with EPA at the same time and place. Our analyses were primarily
for polynuclear aromatics and organo-halides, whereas  EPA determined trace organics
present in the samples using  gc-ms techniques.

         During the planning of the Phase  II study, the task force became aware of a pilot
study  that was to be carried  out by Texaco under contract to API's Refining Department.
This study consisted of evaluating the performance of several different pilot scale treatment
trains — including powered  and granular activated carbon — on refinery wastewater.  The
W-20 Task Force decided to take samples across the various treatment  trains and analyze
them for the presence of polynuclear aromatics and organo-halides.  Conventional
parameters were measured by Texaco.  Details of this study will be given later in the
program by Mr. Len Crame of Texaco, Inc.

         About the time the scope of the Phase II  study was being  finalized,  the toxics
settlement agreement was signed by EPA and several environmental organizations.  Under
this agreement, EPA is undertaking a program to determine which contaminants out  of a
list of 65 allegedly toxic compounds and classes of compounds are  present in the effluents
of 295 industries in 21 major categories,  including petroleum refining. As a result  of this
agreement, a new task force, W-22, was formed at API.  This task force was given the
objectives of (1) reviewing, evaluating,  and critiquing analytical methods selected by
regulatory agencies for  qualitative and quantitative determination of the presence or
absence of trace toxic pollutants in refinery waste waters;  and (2)  conducting sampling
and analyses for these trace toxic pollutants at selected refineries.

         This task force has been extremely active since it was formed early last fall, and
members and API staff have  met with EPA personnel on several occasions. We  have
reviewed and critiqued  the sampling and analytical protocol being used by EPA in its
analyses of refinery and other industrial wastewaters.  The task force at the present time
has grave concerns about the capability of the sampling and analytical protocols to detect
-- and perhaps quantify — the  presence of these toxic compounds at exceedingly low
levels, and of the accuracy, repeatability and reproducibility of the methods being
employed.

         As Marvin Wood mentioned this morning, Kerr Laboratory personnel are presently
sampling and analyzing waters  from 12 refineries.  These include intake waters, separator
or DAF effluents, and final  effluents.  In addition,  EPA is conducting pilot scale powdered
activated  carbon enhancement  of activated sludge units, and granular activated carbon

                                          57

-------
end-of-pipe treatment, at some of these 12 refineries.  The W-22 Task Force is
monitoring this effort and has also selected  three of the 12 refineries for its own parallel
study.  Samples are being taken at the same locations and times as those taken by EPA.
The samples are being analyzed for the 129 priority pollutants using analytical techniques
similar to those being used by EPA contractors.  In addition,  more detailed and specific
analyses of PNA's are being run.  We feel that  the results of this study will  give us
valuable information  concerning sampling and analytical  variability as regards the
priority pollutants in  refinery wastewaters.

         Phase I of the W-20 study is presently  in final review stage, and the last of the
analyses for Phase II are now being carried  out.  As a result, no reports have been issued
and many of the findings are preliminary.  In addition, I  should emphasize that in view of
the limited number of facilities involved (two refineries and  two municipalities), the
study results will not  necessarily be typical  of either the refining industry  or of all
municipal  treatment plants.  Rather,  they were  carried out to build  a data base where
none  existed.

         Preliminary  results from both the W-20 efforts and the W-22 refinery analyses do,
however, look very encouraging.  Although a few of the  so-called priority pollutants
were  found to be present at levels greater than  10 ppb in  separator or DAF effluents, even
these appear to be significantly reduced in  the refineries' biological treatment systems.
Available  results indicate that there are no significant quantities of the allegedly toxic
pollutants  in the final effluent samples of the refineries analyzed to date.  In addition,
the W-20 study revealed that there were no significant quantities of organo-halide
precursors  in the effluents of the two  refineries sampled.

         One very interesting result of both studies  is that a  very poor  carbon material
balance is obtained when comparing ppm total organic carbon (as determined by a TOC
analyzer) to the sum of the ppb concentrations of identifiable, extractable organics.  One
would expect  to find,  however, high molecular weight compounds such as cell metabolites,
proteins, carbohydrates, and oxygenated aromatics in biologically treated effluents and
these would not be identified or measured by the analytical methods used.  It is quite
possible that these types of compounds account for much of the TOC in refinery effluents
meeting BPCTCA and  'in well operated municipal treatment plants.

         It would appear logical, then,  that expensive, cost-ineffective technologies
should not be  required by regulation just because considerable TOC removals are obtained.
Rather, it should first be demonstrated that the TOC removed is actually harmful to man or
to the aquatic environment before removal is required.
                                          58

-------
BIOGRAPHY             Judith G. Thatcher

        Judith G. Thatcher holds a B.S.  in Chemistry
from Southeastern Massachusetts University.  She is
currently an Environmental Associate with  the Environ-
mental Affairs Department of the American Petroleum
Institute.  Prior to joining API  in February of 1976,  Ms.
Thatcher worked for 10 years in the Research and
Technical  Department of Texaco Inc.
                                         59

-------
DISCUSSION

Ned F. Seppi  - Marathon Oil Co. - I wondered if Mr. Denbo would discuss bio enhance-
ment a little more thoroughly.   "

R. T. Denbo- I  think the subject will be discussed thoroughly in a later formally-developed
session.  However,  let me just mention generally what we had in mind when we set up a
research project. PL 92-500 includes requirements for BAT - Best Available Technology.
As time has passed, there have been a number of developments that indicate that toxics
have become the important thing as far as water quality was concerned.  This began to
become obvious to us a couple of years back.  The first question is - do we have a toxics
problem  in BPT treated refinery effluent? And the first step in attempting  to ascertain
that was the program that Judy Thatcher and Kent Drummond talked about  and that was
the analysis of BPT  treated refinery effluents for residuals.  But before the  results became
available we began to look for ways to get  rid of compounds of concern if  they are, in
fact,  present. We  decided to look at what can be done along the line of upgrading
existing bio systems.  Jim Grutsch and others had done work on higher sludge ages and
how effective this is in further reducing organics in treated effluent.  There had been
work done by  many people on powdered carbon. We decided to undertake the controlled
experiment that  Len Crame will talk about in detail.  That's what is meant by bio enhance-
ment or enhancement of biological treatment systems.

Paul Mikolaj - Lion Oil  Co. - You mentioned that you're monitoring three refineries
along with the 12 that EPA is monitoring.  Are  EPA's results coming out the same in terms
of the removal of these priority pollutants?

Judy Thatcher -  We have not seen any of EPA's results to date,  Paul, and we  only have
some  limited results from our own studies right now.  I was basing my discussion on what
we had found  in the two  phases of W-20 and on our limited results to date  on  these three
refineries. We're obtaining samples from the third refinery this week.  We would hope
that EPA's results come out the same.

E. A.  Buckley - Lion Oil Co. - I would like to ask this question of either Mr. Beychok
or Mr. Denbo.  First, Mr.  Beychok, I assume that you will cover the injection of caustic
into sour water strippers.  My question is: at the present time,  do we have enough infor-
mation to use  caustic injection in a stripper being built now?

M. R. Beychok - A preliminary report on SRI's work covering that subject  was given at
the API meeting in  Chicago a few weeks ago.  I think that SRI did a very good job of
quantifying that the optimum caustic injection point is at the top of the stripper with the
feed.  They also defined the amount of caustic required.  That report is available to you
to use  now. The final report may not be issued for some months, but the preliminary
report  is available and provides enough information to be used now in designing for
caustic injection.
                                         60

-------
William C. Galegar - R. S. Kerr Environmental Research Lab - I would like to turn
around the question that was asked a little earlier this morning.  If I include Mr. Beychok
in the group sitting at the table, you broadly represent the group that we have turned to
for handling the research programs that have been undertaken to provide answers.  The
question I would like to ask the group is: are the research programs related to the refining
industry wastewater treatment providing adequate information?  If not, how can they be
improved?  I am talking to your group as a whole since you represent universities, industry
and consultants.

M. R. Beychok - I'll give everyone a chance to catch their breath while I try to respond
to that as an individual private consultant.  My opinion is somewhat as stated earlier by
Bob Denbo.   We've worked the problems of refinery wastewater technology quite a bit
and I  think we now have a good data base.  My experience with some of the research
programs such as the SRI work for the API and the Carnegie-Mellon work for ERDA indi-
cates that our emphasis and attention should now be devoted to  laboratory analytical
procedures.   We haven't defined those procedures well enough yet and there are many
problems with procedures that we've always considered we 11-proven and adequate.  We
don't yet really know how to analyze for ammonia or for  cyanides. At least, that's what
SRI found out.  Dave, do you agree with that statement? """

Dave Bomberger - Yes,  I think so.

F. S. Manning -  The RSKERL has measured the analytical variability of 5 wastewater
parameters for refinery effluents.   This data can be obtained from EPA Document 600/2-
76-234 (September 1976).

Arthur J. Raymond - Sun Oil Co.  - In your talk you talked about your analysis of poly-
nuclear aromatics and chlorinated hydrocarbons.  You said the EPA used gc-ms, what
methods are you using?

Judy Thatcher - Well, the W-20 study was started before the toxic settlement agreement
and before the  EPA sampling protocol came out, so we used a variety of methods.  The
PNA method was developed by Exxon R&E and involves solvent  extraction,  clean up of
the sample over an alumina column, elution, separation of individual  PNA's by gas
chromatography,  trapping of each peak, and quantitative measurement by UV spectro-
photometer with results based on  ^C labeled internal standards.  For the organo-halides,
that involved purging of the volatile organo-halides and  passage through a gas chromato-
graph.  The detector was a microcoulomenter which only measures halogenated hydro-
carbons. The volatile organics were purged from the sample and analyzed by gc; the non-
volatiles were solvent extracted,  I think mainly by carbon tetrachloride. There was some
preliminary  gc-ms work done on the non-volatiles.  With regard to the W-22 studies, we
have  Radian  Corp.  who is running the gc-ms protocol as specified by EPA.  We've made a
couple of changes, one involves the final effluent samples from each one of the refineries.
We'll be spiking each sample with certain of the pollutants.  As far as I know EPA is not
doing any spiking of the actual water samples.  We will run a final effluent sample and
then spike a duplicate of this, and run that and try to get some  feeling for the accuracy

                                        61

-------
of the method.  And then in addition we're looking at polynuclear aromatics by the
Exxon method which can measure, I believe, up to 18 individual PNA's.  We  know al-
ready from preliminary results that the gc-ms method in the protocol does not separate all
the PNA's and so you end up measuring three PNA's under one peak.
                                      62

-------
                             SESSION III

                OPEN QUESTION AND ANSWER SESSION

INDIVIDUAL PROBLEMS IN  MANAGEMENT OF REFINERY WASTEWATER


                        Chairman

                       * Francis S. Manning
                        University of Tulsa
                        Tulsa, Oklahoma


                        Panel Members

                        Milton Beychok
                        Consulting Engineer
                        Irvine, California
                        W. Wesley Eckenfelder
                        Vanderbilt University
                        Nashville, Tennessee
                        Davis Ford
                        Engineering Science, Inc.
                        Austin, Texas
                        H.E. Knowlton
                        Chevron Research Company
                        Richmond, California
                        R.N. Simonsen
                        Standard Oil Company of Ohio
                        Cleveland,  Ohio

                       ''Biography on Page 52

                                 63

-------
                    OPEN QUESTION AND ANSWER SESSION -
        INDIVIDUAL PROBLEMS IN MANAGEMENT OF REFINERY WASTEWATER

                                  Milton R. Beychok
                        Consulting Engineer, Irvine, California

                             W. Wesley Eckenfelder, Jr.
                     Vanderbilt University,  Nashville, Tennessee

                                   Davis L. Ford
                        Engineering Science  Inc., Austin, Texas

                           , *, ,,..   H.SE.  Knowlton
                   Chevron Research Company,  Richmond, California

                                  R. N. Simonsen
                     Standard Oil  Co. of Ohio, Cleveland, Ohio

 N. Seppi - Marathon Oil Co. - I would like  to ask Mr. Eckenfelder if he would discuss
 nitrification and go into the prevention of upsets which result in loss of nitrifying bacteria
 and possibly talk about methods of recovery from the loss of nitrifying bacteria.

 W. W.  Eckenfelder - Our experience to date has tended to indicate that dependable
 nitrification plus carbonaceous BOD removal can be achieved in the treatment of domestic
 wastewater in one stage.  Highly variable success has been attained on most industrial
 wastewaters.  Nitrification is better considered  in a second stage after removal of
 inhibition in a first stage.  The second stage could be either activated sludge  or rotating
 biological contactor or an up-flow aerobic filter.  It also appears that the use of powdered
 activated carbon in the activated sludge process  tends to greatly enhance nitrification.
 The carbon tends to remove compounds and materials which would tend to be inhibitory to
 the nitrification process.  The case to which I refer while not a petroleum-refining
 effluent, indicated that straight nitrification required an average sludge age of about 45
days and that with the addition of the powdered activated  carbon, was reduced down to
about 8 days.  Again, of course, there is an economic trade off here, but apart from the
economics, having a greater dependability of operation would mitigate towards powdered
 carbon or possibly towards a two-stage process as opposed to one.

Davis Ford - I think the question was asked regarding recovery of nitrifiers, and it is a
difficult question,  but I  might communicate one or two things. The first step, and we've
had some experience in this area, is if you have  low nitrification periods which had been

                                         64

-------
preceded by good nitrification, temperature may be the problem because as you know
nitrification tends to drop off in winter climates.  When lower temperature prevails, we
try to reseed from plants which have had good nitrification.  We've done this down'in  the
South Texas area with some success.  Of course we know all the other environmental
factors which we can correct, namely the pH sensitivity, the sludge sensitivity, and I
certainly agree with Wes on the sludge age.  Take sludge age and look at some of the  old
graphs and we expect certain percent nitrification that just isn't there. And that's
particularly true when you have a high concentration of organic nitrogen, TKN, particu-
larly amines.  I think in plants wherever applicable, control of amines could be an answer
to nitrification in biological processes. And of course, the third response as far as
correction goes is to get a good lawyer. We get into very strict ammonia nitrogen con-
centration requirements on NPDES permits.  It is my opinion that many of the permits that
go into effect on July  1, 1977 are going to be noncompliant in terms of ammonia.  So  it
takes the combination of a lot of negotiation with regulatory authorities and some of the
steps  I outlined.

Eldon Rucker - API  - There was some mention this morning about the  fact that the refining
industry appears to be well on its way to solving many of the problems of  the priority
pollutants.  I wonder  if any particular members of the panel, or the entire panel, might
comment on whether some of these problems could be transferred into the  sludge, and if
this is the  case, what are the accepted methods for handling sludges  generated in
refineries?

 H. E. Knowlton - As a member of the API Solid Waste Committee, one of our concerns is
the new Resource Conservation and Recovery Act which affects handling and disposal of
oily sludges.  Many people recognize that land farming is an excellent way of treating
oily solids and biosolids.  Land farming of oily solids and biosolids done properly will  not
put contaminants into the ground water, there is not a run off problem because you
normally dike these installations so any net water is put back into the wastewater treat-
ment system.  The studies so far show that the net water from a land farming area has very
 low amounts of contaminants such as oil.  We also believe we should recover as much  oil
as economically feasible so that the amount going to land farm will be relatively small.

 Davis Ford - I might just make one comment with respect to sludges from  refineries.
Speaking now as a consulting engineer, you start off with  land farming and then back  off.
 Because of process applicability in many areas in the U.S., successful land farming can
be achieved for oily and biological sludges.  Now this normally forces digestion to satisfy
most state  regulations,  then combining with biological sludges, and then to land farming.
Another comment is that the key here is dewatering.  I think technology  has certainly
progressed a long way in dewatering of oily sludges.  I know.that EXXON and others are
having good success with belt filters,  so if you can successfully dewater through belt-
press, or other filters, or other comparable processes, and combine that with the digested
activated sludge, land farming is the place to start.  This is far better than less palatable
methods such as incineration, so I think that's our starting point.
                                         65

-------
E. A. Buckley - Lion Oil Co. - On your land farming of the sludge, how do you deter-
mine how much oil you've got say in an alum sludge where you're coming off a DAF
system?  Of course you have floating oils which you can separate, but how about
separation of the alum sludge?

H.  E.  Knowlton - My answer to you is that at Chevron we don't have any alum sludge.
And we don't have any for a very real reason.  About 3 or 4 years ago we said, let's go
to the new induced air flotation process; we don't want the problem of handling-disposing
of that sloppy mess.  And it is a mess.  So we don't have this problem and I can't give you
any solutions for a non-existent problem.

N. Sepp? - Marathon Oil Co. - What's a rule of thumb as far as the barrels of oil per
acre that you can soil farm?

H.  E. Knowlton - In a recent paper on  land farming prepared by Sun Tech and given at
the CREC meeting  in Chicago, they were land farming 600 barrels of oil per acre per year.
I  think the answer  is, where are you?  If you're in an area where it's quite warm such as
in Southern U.S.A. where on  a hot day (95 F) we measured 130 F one inch down !n our
soil of the land farm, then you can expect the oil to  biodegrade rapidly.  This assumes
that your land farm is set up properly.  In the northern areas obviously your working time
is shorter and your temperatures lower, so biodegradation potential per acre is less.

Ben B. Buchanan - Phillips Petroleum Co. - I would  like to ask the  panel if they could
comment on the revolving disc biological process as compared to activated sludge, and if
they know of differences in the types of bacteria used for both processes? Of course the
slime forming bacteria and the revolving disc process  I don't know how different they are
from the activated sludge bacteria.  Can you have nitrifyers in both systems,  or are there
limitations to what bacteria on revolving discs can do compared to what they do in
activated sludge?

R.  N. Simonsen - We have had pilot plant experience at one location having a well
operated aerated lagoon system with several days aeration time. We are interested  in
getting nitrification .and ran both activated sludge and rotating biological surface pilot
units using aeration basin effluent for feed. Although we  got effective nitrification with
activated sludge, there were frequent upsets and recovery was very  slow. The RBS unit
in the same application worked quite well. We don't know what differences  in bacteria
type there may have been.  Dave Rulison who is in the audience and did the  work might
comment further.

Dave  Rulison - Sohio - Only that in the case of activated sludge the variability of  the
feed coupled with  the difficulty in retaining nitrifyers in the system meant that small up-
sets would ruin nitrification.  With the  RBS unit, even significant changes in feed quality
didn't make much difference.
                                         66

-------
H. E. Knowlton  - One of your questions was comparison of activated sludge vs. biodiscs.
To answer that, in six Chevron refineries, one is using activated sludge, two are using
biodiscs and three are using ponds.  The criteria used to pick a biosystem for these
individual situations was which system will cost the least.  I will make one comment on
what we see with one of our biodisc systems that went on stream about December 1976.  A
month ago an operator dumped 11 tons of acid into the system.  This was not discovered
until morning, by which time the acid removed most of the biosolids from the eighteen 11
foot diameter by 25 foot long discs.  The acid cleaned those discs of that material.  The
item that really made us quite happy was that at the end of two days we were back meeting
our phenol spec.  In other words the discs had recovered sufficiently so that we were
meeting our phenol specification  in the final effluent.  We thought this was quite remark-
able, because of the severity of the upset.  We have found by other experiences that of the
contaminants removed in the biodisc process that phenol removal  is the last to recover.

Davis Ford - I would like to make a few comments on the RBS-activated sludge comparison.
Let me use RBS (rotating biological surface) because "Bio Disc" is a trade name and  I
don't want to advertise.  Let .me say that I have personally come  to full cycle about  4 or
5 years ago, inherently I was rather negative on the whole RBS concept.  I think today I
have a much different view about that.  Chevron Research was partly responsible.  Let me
make a few  specific comments, and I'll  discuss that in more detail Wednesday morning.
First of all,  from an operator's point of  view and from a cost point of view,  RBS or bio
disc  has a lot of innate advantages, mainly on power-connected horsepower as compared
to activated sludge.  It's much simpler to  operate, and I think the capital costs with
BOD's below a certain level - I'm not sure what that level is - but capital costs are
probably more cost-effective or cheaper than the activated sludge.  I don't want to make
these positive comments at  the expense  of activated sludge; that's been the "modus
operand!" in the petroleum refining industry for a long time and it's really the basis for
BPT and will probably be the basis to some extent for BAT in the remaining guidelines.  I
believe that as the  influent BOD  gets higher, let's say 750 mg/l or higher, then activated
sludge might be equally or  more attractive than RBS systems. Again, I don't have
specific numbers for this but when you get into high BOD's you'd better check the  cost-
effectiveness quite  carefully before making the process decision.   We designed both  RBS
systems and activated sludge so we have no inherent process bias. But let me make a few
comments on questions that I still  have about RBS  systems.  One,  I think you have  to be
careful when comparing the ability to withstand upsets between RBS and activated  sludge
because you're getting full  raw waste load impact on those first discs; for example, an
acid spill.  Whereas in a completely-mixed activated sludge system the acid equivalent
or potential  bio-toxic or bio-static level per bacteria is reduced, it is less than in RBS
systems.  I think the jury is still out on  the ability to withstand upsets, recognizing too
that you have the ability to operate at very high sludge inventory on any activated sludge
system, and  you  kind of ride the line  on the RBS system.  It has been proposed, primarily
by the vendors, that sludge settleability in RBS systems is better than that of activated
sludge. I think that would  be the case, although I have yet to see it proven.  So, I  still
have a question on the effluent TSS ability of an RBS system as compared to activated
sludge. With  respect to BOD less than  750 or 500 mg/l, we found that activated sludge


                                        67

-------
systems take out about 1.5 to 1.6 Ibs of BOD per horsepower hour.  If you can use that
parameter to determine the power efficiency or power utilization removing BOD, whereas
RBS systems on the paper studies that we have done preliminary to the detailed design
have been  about 2.3,  so it is more energy-effective. At low BOD  concentrations you
have an inherent incentive to consider RBS systems, particularly in  land-limited situations.
One other  comment - RBS has been proposed by the organic chemicals industry and we're
a bit reticent  to accept that system carte blanche when  you have COD problems because
your contact time between the  bacteria and the sludge in RBS systems is going to be less
than it is,  for example for extended aeration.  You can solve the BOD problem but
possibly not the COD because the hydraulic contact time is quite low.  There is no way
to accurately  calculate sludge  age, which you can of course in an activated sludge system.
One quick comment on nitrification - I believe that the RBS system offers advantages in
nitrification as compared to activated sludge, because of the ability of the latter stages
in the  RBS  system to develop a  good nitrifying bacteria.  And also,  it has been quite
successful in adjusting the pH, raising  the pH in the latter stages to the optimum level for
nitrification.

Bill Ruggles - Phillips Petroleum (Bartlesville) - With the increasing emphasis on hydro-
carbons in  the atmosphere, I am wondering if any member of the panel has experience
with possible effects on this situation that has resulted from  the land farming of sludge?

H. E.  Knowlton - There are no numbers available on this to our knowledge.  In fact, one
of the  items on the API budget  for next year is to measure hydrocarbon emissions from a
land farming area.  We expect very little as we've spread oily sludges and rototilled them
in and have not been able to smell any odor in the area.

E. A.  Buckley - Lion  Oil Co.  - One other question. What are the most feasible as well
as most economic ways of reusing treated wastewater?

M. R. Beychok - There are literally hundreds of answers to  that question and they are all
refinery or site specific.  The first answer that comes to mind is the reuse of treated waste-
water  as cool ing tower makeup. Another answer  is the  reuse of treated wastewater to
produce low pressure steam (rather than high pressure steam  which requires relatively pure
water  to avoid excessive and uneconomic blowdown).  The low pressure steam might be
used in atomizing the oil fired  in process furnaces, or as flare steam,  or as stripping steam
in crude unit sidecut reboilers and in other process steam uses.  Using treated wastewater
for low pressure steam  generation may require 10-15% blowdown to avoid fouling
problems in the steam generation units. If you attempt  to reuse treated wastewater for
high pressure steam generation, you can't afford such high blowdown rates and you will
need tertiary treatment following your  secondary  treatment to make the water suitable for
high pressure boiler feedwater.  As you may know, the API  CREC funded a pilot program
studying the reuse of treated wastewater in cooling towers to take advantage of the
evaporation in the cooling tower so as  to result in a more concentrated waste in the form
of the  cooling tower blowdown.
                                         68

-------
E. A. Buckley - Lion Oil Co.  - What parameters would you put on the IDS in evaporated
cooling water blowdown?

M. R. Beychok - It depends upon  the cycles of concentration at which you  operate the
cooling tower.  It also depends upon the economics and whether you can justify a slip
stream filter or a slip stream softener on the circulating cooling water. Some refineries
have so-called  "clean water" and "dirty water" cooling towers which  adds another
parameter of flexibility for  reusing treated wastewaters.

H. E. Knowlton -  I can give you  an example of reuse recently completed which worked
out a lot  better than we  hoped. This refinery had 800 GPM of effluent; it now reuses 400
GPM of filtered effluent 250 GPM into the cooling towers and 150 GPM into the process
water fire-water system.  We've been running only a month and a half or so but this has
worked out very well. We  see no problems so far and the effluent is reduced to 400 GPM.
We do not have a lot of  analyses as yet to tell us if anything is unusual, but we don't see
anything. Also we're surprised, we thought that maybe the calcium might double but so
far we can't see it  changing much.  Some things happen in a practical system that when
we plan we don't anticipate.   Other refiners, by the way, have used API-separator water
in their process water and fire water systems for 25 years.  Their water is very poor
quality compared to that we are reusing which has a BOD of less than  10 PPM and an oil
and grease content of less than  10 PPM.

R. N. Simonsen - We have used effluent in fire water systems, but I should relate what
happened when this was  tried at one refinery which has since been shut down.  Effluent
had been used to pressure the fire water system for several months.  Then, during a plant
fire fighting training exercise, attempts to extinguish a fire with a foam generator failed
because the powder would not make foam.  New powder from the storehouse also failed
to make foam.  This was how the refinery discovered  it had been operating several months
without fire foam protection.   Neither we nor our supplier ever learned why and, of course,
the refinery changed back to its former fire water supply.  A number of refineries use
stripped foul condensate and blowdown from oily cooling towers in desalters as another
form of recycle.

M. R. Beychok - This isn't another specific example, but rather a general caution.  When
designing any process unit, you must be very careful  about recycling.  You must avoid a
closed loop from which there is no way to bleed out any buildup of a recycled  impurity.
The same thing holds true for reuse of treated wastewater.  You don't want to recycle  in
such a manner that you create a closed loop and build up an intolerable level of dissolved
solids or other contaminants.  So you want to thing in terms of cascading systems.  If your
treated wastewater originated from steam  condensate, reuse  it to generate steam which
will  be used in other services.  Or if you reuse the water in a cooling tower, be sure not
to lock yourself into a closed loop which has no way to bleed out or blowdown impurities.

Ed Sebesta - Brown & Root, Inc. - I would  like to go back to the subject of biological
rotating discs reactors, and mainly some comments on the maintenance - good or bad,  on
some of these systems that have been in operation for some time.

                                        69

-------
R. N. Simonsen - We haven't had any maintenance problems since oOr full-sized RBS
units haven't been installed yet.

H.  E. Knowlton - We have had one in operation for about 2£ years, it's a styrofoam disc
unit.  We've had a few things happen, like the carbon steel  nuts on the ends of tie rods
corroded off, and then the styrofoam discs moved out of position.  One of the problems
with rotating disc units is that people tend to never  look at them, tt sits there and runs
with very little attention. Mainly one tries to control feed quality to it because what
you put in determines what comes out.  So the  main effort is  to keep a reasonable plant
control on your input quality.  We suggest now that every three months they go to and
actually check eqdh disc and make sure that nothing is wrong.   There should also be some
surveillance.  The operator should walk by each of the discs  once per shift ro see if there
is noise or anything indicating a problem.  A second unit we have in operation has been
operating roughly six months. We see no problems there; it is a large system of 18 shafts.
There was a recent NPRA paper on rotating discs; (AM-77-27),  Refinery Use of Rotating
Biological Surfaces in Waste Water Treating.  In this paper we  gathered the available
information on four existing commercial size installations that were! processing refinery
waste water in the U.S. in February 1977.  You can ask Herb Bruch, NPRA Technical
Director, for a copy.  There is a fair amount of detail  in  it on this subject.   Are  there any
other questions on this one ?

Ed Sebesta ••* Brown & Root, Inc. - A follow up while you're  up. When ypU did have that
problem with the one at Salt Lake City/ to redo it how did you operate while you're dding
repairs, or what did you have to do to repair it?

H.  E. Knowlton - They stop  the  rotation and bypass the water to the next shaft.  Then
they patched the styrofoam discs  that were torn. You can by-pass these things.  This
system is set up.with four shafts in a row and has two rows; if  necessary you can by-pass
a whole row, but normally it isn't necessary to do  that, you can just by-pass the one
involved and work on it.

L. D. Erchull - Union Oil of California - I'd like to address  this question to the panel as
a whole.  Oh nitrogen removal from refinery waste water streams after they  have  been
steam stripped, are there any installations that you know of where some proven technology
has been applied where it can work year round, say  in the northern climates of the country?

R. N. Simonsen - Are you talking about treating stripped foul condensate by itself?  We
hqveh't tried that,-but nitrification of refinery wastewater containing stripped foul con-
densate can be accomplished with the RBS in northern Ohio.  But temperdtute is Very
important and the colder it is the more surfdce  is needed.

Davis Ford - I would like to add one more comment to  that.  I mentioned a case history
where the nitrification has been consistently 85%.  I might mention that the  ammonia
coming in about 10-15 g/liter, that's a pretty low concentration for a raw wdste  load
going through a biological system just to put that in  perspective.  I'm hot sure that we


                                         70

-------
could get that much nitrification if we had higher concentrations coming in. Another
thing that we haven't mentioned yet on the RBS activated sludge system is that you've got
to remember that the heat loss in an  RBS system is much, much less than in activated
sludge, which is in effect a cooling  tower with mechanical aerators.  So, that would give
us an added advantage on nitrification which is so temperature-sensitive in the northern
climates, using  RBS compared to activated sludge.  With respect to pH, it has been indi-
cated that about 7.5 Ibs of alkalinity as calcium  carbonate is destroyed per  Ib of nitrogen
nitrified.  I think perhaps we would  have a little better pH control  in an RBS system  using
the approach I discussed earlier where you add alkalinity to the latter stages that have
already experienced an initial pH drop through the production of CO«.  So  even though
you have a pH problem in both activated sludge and RBS, you might have a little more
positive control on the RBS system by controlling  pH through  the various stages.

W. W. Eckenfelder -,| think one of the benefits of the rotating  contactor for nitrification
as opposed to activated sludge stems from the fact that you will  generate roughly 50  Ibs of
bio mass per 100 Ibs of BOD removed, but only about 15 Ibs of bio mass per  100 Ibs of
nitrogen oxidized. < What this means is in a completely-mixed activated sludge where both
BOD and ammonia are  to be oxidized the population of nitrifyers is going to be very very
low; in the order of probably 1  to 4%.  In a rotating contactor, admittedly in your initial
stages where you are primarily rmoving carbonaceous organics you will have a very very
low population  of nitrifyers, but once the carbonaceous organics are essentially gone,
then 40 to 50% of the bio mass will be effective nitrifying culture.  This provides a
cushion against both changes in concentration and also changes  in operating temperature.
Where  nitrification is important, and this refers to my earlier comments on one stage vs
multistage activated sludge, that you should consider the relative concentration of
nitrifyers as opposed to the concentration of other organisms.  With respect to pH, 7.15
Ibs of alkalinity are required per Ib  of nitrogen oxidized.  Normally the amount of
nitrogen to be oxidized should not probably pose  a major pH problem. But it could be a
problem if the available  alkalinity is low and your amount of nitrogen to be oxidized is
high.

H.  E.  Knowlton - I  would like to  make a comment to  the Union man. If we find we have
an ammonia problem we go back into the process  units and find the ammonia sources.
Then we can strip it better; or reduce the volume of the stream before stripping it.  We've
done both and we end up with a significant NHo  reduction.  We also find high content
NhL streams going into waste water treating  from sources not previously recognized. We
of course send them to NHk recovery.
                  !;.." i  •
M. R.  Beychok - Well, Buzz beat me to it because he didn't play tennis this morning and
he can get up faster  than I can. But I want to second  his comments.  I think that very
few refineries really have good nitrogen balance  data  across their wastewater systems.
The first thing is to find out where your,ammonia  is coming from and find out what ammonia
containing streams are  bypassing your stripper. Then.you should provide better ammonia
stripping or reduce the volumes of  those streams or both.  The end-of-the-pipe treatment
for nitrogen ought to be a last resort. ,
                                         71

-------
Davis Ford - Just this comment - I don't want to mention names, but there certainly are
plants.  One plant in particular that I'm thinking about was not designed for nitrification
but its nitrification level has been consistently at about 85% through the whole 12 months.
It is in the Southern  part of the U.S. so the temperature is favorable.  Right next door,
however,  there is a problem in nitrification. To get back to my earlier comments,  there
are so many variables - specific toxicity of certain organic compounds, etc. - so that it
is just not that predictable.  But there are case histories of good nitrification throughout
the year.   You're from Union,  I guess.  I know when you get up around the Chicago area
and you have a significant drop in temperature, that tendsQto play havoc with nitrification
systems - and I'm talking now of just around 15, 10, or 8   C.

Bob Carloni - Lion Oil  Co. - We've just recently put on a treatment plant which uses a
mixed media gravity filter  and we weren't getting very good solids removal until we
started using a polyelectrolyte. Is that normal  experience?  If it is what sort of concen-
trations of polyelectrolytes should we be considering?  We're currently using 2 ppm and
we've heard that we should be down to around  .5 ppm or lower.

Davis Ford - What type of  system are you talking about?

Bob Carloni - This is a filter which follows a clarifier which follows an RBS system.

Davis Ford - The first comment is, yes, polyelectrolyte is being used to keep people out
of jail all over the country right now.  Concerning control of effluent TSS levels without
filters, and I  can  cite you  example after example,  it is not free.   In some cases the con-
centration requirements are 2,  4 and 5 mg/l and I know the polyelectrolyte salesmen are
going to be happy.  It is about $1.00 - $1.50 per Ib, so if you have a high flow it's an
expensive  operational procedure.  It's been working quite  well, however, just for final
clarifiers where just the poly is added to the activated sludge or RBS effluent and is mixed
going to the final clarifier.  With respect to post filtration, it  is my opinion that you have
got to have at least  a poly addition capability  prior to post filtration.  Once you get into
the operation you might want to back off,  but as a capital expenditure you'd better have
the ability to add a  coagulant or coagulant aid to enhance filtration.  By the way, the
enhancement of effluent TSS quality by virtue of adding poly,  which is normally 1 - l£
ppm, but sometimes  a little higher, has often been around  30% to 50% improvement in
equivalent TSS.

H. E. Knowlton  - Chevron has only one refinery with a filter on a final effluent.  On
this one final filter we do  not use polyelectrolyte. As I recall we get roughly a 50% drop
in BOD.  We possibly could  get more removal  with polyelectrolyte but we don't use it
because we have no need. We see these results of the filter—approximately 50% oil
reduction, and 30% BOD reduction.  We run this filter on a time  cycle only because we
find that the AP control system was not practical.

R.  J. Churchill - Tretolite - We're one of the polymer companies that are getting rich
according to Davis.   Polymers are useful but only as a tool.  They are not a panacea.

                                          72

-------
y^e've been called into places where people have bulking type problems and ask us to add
a polymer to solve that.  You're treating a symptom you're not solving a problem.  We
have had success in the application of filters by going into the clarifier, that is adding to
the mixed liquor on the way to the second clarifier, which does two things. One, it
knocks down the TSS leaving the clarifier going to the filter; second, almost uniformly or
almost characteristically we have seen it knocks down the variability so that you don't
slug the filter with a high TSS concentration episodically.  That does two things and it
reduces the total solids load put onto the filter which should extend your run life.  It
should also then give you better net TSS out of the filters since you're starting with a
lower TSS in.   It also may turn out to be more economical to go to a split addition rather
than trying to do it all in one step.   I think the key to the polymer is simply to use it as
a tool  that is to take out the fluctuations and the total amount of TSS leaving the clarifier
going to the filter. To answer the point about one-half a part or 2 parts or 10 ppm,  we've
treated systems with less than a part very low concentrations at the secondary clarifier.
We've also treated refinery systems as high as ten parts at the secondary clarifier.   I think
that relates back to some of the points that  Buzz was mentioning.  You can have bulking
or nonsettling type sludges that's not a problem to be solved by a polymer.  You can oil
carry over into your biological system and into the clarifier which affects settleability.
Again, that's not a polymer problem. It's a process problem or a waste management
problem. I think it's hard to draw any hard fast rules but excluding my bias I do think it's
a tool, but it's a tool that only can  be used wisely.

Rich Sheridan - Brown and Caldwell  - I  think it was Mr. Ford that mentioned the appropri-
ateness of the bio  discs in situations where nitrification  would be required  in cold climates
and therefore the bio discs have the advantage of less heat loss.  I would  like for someone
to  comment on the requirement to nitrify in cold climates, as  to its appropriateness, cost
effectiveness,  these types of things.

W. W. Eckenfelder -  I think it's obvious to all that in most cases involving cold climates,
as  far as water quality goes, nitrification is neither justified  nor  is it required. My own
opinion  is that these cases should consider a two tiered permit in  order to avoid a large
economic penalty  for nitrification in cold weather conditions. As far as water quality
goes,  the fact  that the process of nitrification radically slows down in the  treatment
process, it equally slows down In the receiving water.

 Eldon Rucker - API - In view of the existing widespread use of lagoon systems in the
refining industry,  would the members of the panel  comment on the efficiency of these
systems compared with some of the RBS and activated sludge which have been previously
discussed ?

M. R. Beychok -  It depends upon what sort of lagoon system you're talking about.   I will
be very specific and talk about surface-aerated systems using mechanical aeration as
against those that depend simply upon photosynthesis.  I'm talking about lagoons with 3-
4 days retention,  12-15 feet deep, and  using up to 50 hp per million gallons of retention
for mixing and aeration.  About 4-5 years ago, an AlChE committee  surveyed existing


                                          73

-------
aerated lagoons In the refining,  petrochemical and the pulp and paper industries.  I don't
know if Wes Eckenfelder remembers it or not, but one of his colleagues reviewed our final
conclusions from that AlChE survey and disagreed with us.  In any event, the data showed
that 85-90% BOD removal  could be achieved in those lagoons,  within the parameters
mentioned:  12-15 feet depth, 3-4 days retention and up to 50 hp per million gallons of
retention.  That contrasts with about 90-95% BOD removal for activated sludge systems.
I'm not sure what the average percent BOD removal in an RBS system is, but I would
imagine it is somewhere between the aerated lagoon and the activated sludge systems.

W. W. Eckenfelder - I think one thing is pertinent and that is where you have an aerated
lagoon system, or a waste stabilization pond, you are going to be much more subject to
temperature effects during the colder weather than you would be from either an RBS or
activated sludge process.  The variability month by month would generally be higher from
an aerated lagoon than it would  be from those processes that tend to minimize heat loss
during the colder climates.  I would agree from all of the data that I have seen that a
properly designed and operated aerated lagoon process should be capable of doing about
the same thing as the other processes relative  to BOD removal.  One of the obvious
problems with  an aerated lagoon is the effluent suspended solids.

M. R. Beychok - To further qualify my earlier answer regarding 85-90% BOD removal in
aerated lagoons, we adjusted for the variability of temperature by using a power law
function.  We converted all of the data  to a base temperature of 77  F.  We also
recommended that aerated lagoons  be designed for the coldest temperature in order to
achieve the required BOD removal during the winter.  Then one could take advantage of
the oversized system capabilities during  the summer months to schedule maintenance and
turnarounds.

John C. Doolittle - Shell Oil Co.  - Relative to the aerated lagoons, what effect would
an occasional  oil film on the  lagoon surface have upon BOD removal?

M. R. Beychok - I suspect  it  would depend upon whether you have 15 hp or 50 hp per
million gallons of retention.  The key thing in my opinion is mixing and not aeration.  A
high degree of mixing will offset any effect that an oil film might have on oxygen
absorption.  50 hp per million gallons of retention will more than provide the  oxygen
required.

H.  E.  Knowlton - Eckenfelder said that the lagoon system is very dependent on temper-
ature.  And in  certain areas of the country the lagoon system wouldn't be effective.  This
also has been our experience. In the southern part of the U.S.  or Hawaii we  find that the
lagoon system  is excellent even though we have some TSS problems.  In fact in one
refinery we exceed comfortably  "77" permit conditions with a lagoon system.   I would
like to mention that in the  first refinery in the U.S.A. to use biodiscs, we did have a
lagoon system.  The only reason  we put in biodiscs was that in the winter time when the
effluent water was 33  F, we couldn't get enough BOD reduction.  We put  in the bio-
discs and even  though the water  into the biodisc is at 50  F, it drops only about 1  F at


                                        74

-------
an ambient temperature of 20  F.  In our second installation during the severe cold
weather of last year, we saw only a 6 to 8 drop across the bio disc system with ambient
temperatures of 0  F which is a pretty minimal drop.  We did not suffer at all from the
severe cold weather.  This is one of the pluses we see for bio discs.

Davis Ford - I  have one other comment on the aerated lagoon waste stabilization pond.
I'm not really sure that your question is on waste stabilization ponds or aerated lagoons,
but let  me add a postscript to the aerated lagoon discussion.  I  really agree, and some-
times it's difficult to convince regulatory authorities, that the aerated lagoon  system is a
BPT system in many cases, and I really believe that.  Of course, this depends  on your
raw waste load and how far under the guideline number you are on RWL.  The TSS comes
back to haunt us.  It's going to be  interesting to see how the TSS numbers come out in the
remanded BAT guidelines for the petroleum refining industry in  September or October,
because that's really the key to how applicable these systems are going to be.  In many
cases an aerated lagoon or waste stabilization pond concept can produce, from refinery
wastewater, a much lower COD than can an activated sludge system.  So there are
inherent advantages.  We just have the problem of algae proliferation and biological TSS
to contend with, and in effect the  remanded guidelines and the implementation thereof
will dictate just how applicable these systems are.  In addition of course we have some pH
problems with waste stabilization ponds; 6 to 9 being the normal pH on those permits; and
as we undergo photosynthesis using CO« the pH goes over 9.  We have been successful,
and others have too in some cases, in  having a time stipulated on the permit as to when
the pH  is to be taken, namely, early in the morning to exclude that CO« utilization by
algae.

E. G.  Kominek - Environtech Corp.  - It is apparent that there is a lot of interest on the
part  of this group  in discussing nitrification, and it has been my experience recently that
in many areas that denitrification is becoming more and more of a problem.  Is this some-
thing,  in the opinion of the panel, that the refineries have to be thinking about at this
time?

Davis Ford - I don't think so unless your effluent discharges into a drinking water source
so that  you'll have to comply with  the Safe Drinking Water Act which has, I believe,  a
10 mg/l as N maximum.  If you discharge into a body of water that is used for drinking
water it could  be a problem. None of the permits  we're involved with have nitrates on it
and I don't anticipate any.  So given that drinking water exclusion I don't think you'll
have any problem  there.

R. N. SImonsen - One other comment on aerated lagoons.  I believe  the results of a
recent API survey  on cyanides in refinery effluents shows that the lowest concentrations
are from refineries with aerated lagoons or long residence time oxidation ponds.  This
parameter is one of the 129 toxics and could become a serious problem for refineries dis-
charging to low flow streams.
                                         75

-------
F. J. Kuserk - Texaco - We have an activated sludge system that does a very good job
with most parameters, except we do have one problem with it.  It seems to increase the
one pollutant,  hexavalent chrome.  I was wondering if one of the panelists could tell us
what's happening here?  We're coming out  of our API separator at hexavalent chrome
levels of less than 20 parts per billion, we go through our waste water treatment  plant and
we're coming out in  the range of 60 to 80 parts per billion.

M. R. Beychok - I don't have an answer to that.  In fact, it amazes me because  I recall
that the API -EPA 1972 raw waste characterization surveys actually showed a decrease in
total  chromium and heavy metals across activated sludge units.  Are you in effect saying
that some total chromium is being created?

F. J. Kuserk - No we're not creating total  chrome we're just going from the trivalent
state to the hexavalent state.

Robert Carloni  - Lion Oil Co. - Are.any of the panel members familiar with any
installations using ozone to improve the treatability of refinery waste waters?

R. N. Simonsen - I don't know of any  in the U.S., but when Cities Service operated
their Trafalgar refinery in Ontario,  ozone was one of a great number of treatment steps
used.  I don't believe ozone is being used there now.

Ralph Churchill - Tretolite - Could  I get some discussion from the panel relative  to the
chrome problem and  the alternatives to meeting chrome guidelines, namely chrome
removal from cooling tower systems, or closer control of chrome in cooling towers,  or non-
chromate  inhibitor programs?

H. E. Knowlton - Ralph, I can tell you what our status is. We do not see any problems
with hexavalent or total  chrome contaminant levels in our effluents.  The hexavalent
chrome is reduced and total chrome  removed by our waste water treating system operation.
So we don't really see the need at this point of changing corrosion inhibitor or installing
chrome removal systems.  Especially if we can reduce our final effluent to a trickle, in
the future, which we feel is desirable to get out of the clutches of EPA.

 R. N. Simonsen - We have a refinery  and a chemical plant side by side at one location.
 Both have similar water supplies and cooling towers and both used chromate inhibitors.
 Chromium is reduced and seems to drop out to some extent in the refinery system  and is
really not a problem.  Chrome has been a problem at the chemical plant and it has been
necessary to switch to non-chromate treatment.

M. R. Beychok - I would like to respond for those cases where you do have a problem.
For example,  industrial plants in California are faced with a coastal water discharge
regulation limiting total  chromium to 0.01  ppm (10% of time) and 0.005 ppm (50% of
time). The treatment alternatives to consider include chemical reduction and precipitation
which creates a gelatinous  sludge which is  as difficult a problem as the original  chromium
problem.  The  other alternative is ion  exchange, of which there are two types which

                                         76

-------
might be used.  One ion exchange process is the fixed-bed type which recovers the
chromium for reuse in the cooling water.  Cities Service has such units in a petrochemical
plant in Louisiana and I assume that they are working well.  The other ion exchange
process is a fluidized bed system with the ion exchange resins regenerated in a U-bend
type system.  As for using non-chromate corrosion inhibitors,  it has been my experience
that the petroleum industry doesn't think that the phosphate or zinc inhibitors work as well
as chromate.

Arthur J. Raymond - Sun Oil Co. - Is there any data available on the correlation of BOD
vs. TOC analysis?

Davis Ford - To answer that, yes, a lot of it but it's very site specific and many things
influence that correlation.  First, just the severity of the process will  have one impact on
the ratio and then the degree of treatment through a biological system will have another
impact. So  I would say in order to be substituting TOC in terms of BOD you will have to
develop that correlation  for your own site.  You get into a lot of trouble by extrapolating
that correlation to your own site and accepting a permit on that basis.

J.  Dewell - Phillips Petroleum Co. - Does anybody have an experience with a new vendor
process called dispersed air flotation as compared to what's called induced air flotation,
and do the same chemical additives work?

H. E. Knowlton - We have a couple  of these in operation.  For four years we've been
using induced air flotation in various  refineries right after the API separator.  I think the
dispersed air flotation uses less horsepower. You might get slightly less removal too,  but
you still use the chemicals. We see it as simpler to operate and maintain than induced
air flotation.

J.  Dewell - What size of air bubbles  do you get with the induced air flotation?

H. E. Knowlton - I haven't worried too much about that detail since the system works so
well in practice.

John Byeseda - Tulsa University - I am working on induced air flotation now and in general
the bubble size depends on the surface tension between the liquid and the vapor.  If
you've got anywhere from about £% of salt the bubble  size will be in the range of 1 mm.
For dissolved air flotation you'll get somewhere around 100 microns practically independent
of the interfacial tensions.  And for dispersed air it's about the same size as induced air -
it will be in  the range of 1  mm,  if there's very little salt in the water the bubble size will
go up to about 2 or 3 or 4 millimeters.  There is  approximately an order of magnitude
difference, about 100 microns.

N. F. Seppi - Marathon Oil  - I had another question about soil farming.  I was wondering
how important is fertilization to the rate of degradation, and what effect do metals have
on the bacteria degradation of oily sludges?


                                         77

-------
H. E. Knowlton - There are some Shell reports out on this, on both topics.  To fix the
application rate you'd have to check the soil and see how much you need.  In fact in one
of our refineries we did not add any nitrogen and still got sufficient activity.  So, it's a
matter of finding out how much is there and  calculate how much you need to add  if any to
achieve a desired result.

John R. Kampfhenkel - Sun Oil Co. - I have a question for Mr. Kuserk of Texaco. Are
your samples on the chrome taken or analyzed on a composite sample?

F. J. Kuserk - The last samples we had were done by an outside analyst.

John R. Kampfhenkel - And you got the same increase in your hex chrome?  We've found
down in Corpus that we were getting the same problem but we were getting it on a
composite sample.  It seems that we were getting a reduction of our hex Cr chrome in  our
sample bottles.  Our influent water had a larger amount of oil  in it naturally than our
effluent did and we were getting reduction because of the oil in our sample containers.
On a grab sample we didn't find that to be true, we found that the hex in and  out was the
same.  I don't know whether it was just typical  of our particular facility or not but we did
find that to be  the case in the sample  container because we went back and spiked some of
our samples and found that was what was happening.

Robert Carloni  - Lion Oil Co. - My question has two parts.  First, what is the state-of-
the-art now in  automatic pH control systems?  And, second, would a system which con-
sisted of one well-mix basin with feed-back control be satisfactory? In this particular
situation we could tolerate a certain amount of swing in the pH of the effluent from the
control system.

Davis Ford - I can't really answer that question specifically, but yes,  the state-of-the-art
is feed forward, feed back in pH control systems, but I  have yet to see  one work very
well.  Every time you go to the plant, the first thing the operator cusses is the pH control
system, so I guess that is the state-of-the-art.  A two-stage control system is good when
your titration curve indicates you can fine-tune on the second stage and satisfy most of
the neutralization requirements in the first stage. That probably helps.  With respect  to
swings in the PH of biological systems, of course you're going to have a normal swing
there. As for the petroleum refining industry, I believe pH compliance has not been a
major problem except for the algae effects,  at least in the plants I've  been involved with.
We have pH swings -  you know what you normally have  in wastewaters, unless you have
an acid dump or acid spill  - and you try to control it in  an off-spec basin or you try to
control it at the source.  I  just haven't much experience with pH excursions from  the 6 to
9 permit limits.
                                         78

-------
BIOGRAPHY     M.E. Knowlron

        H.E.  (Buzz) Knowlton is a Senior
Staff Engineer of Manufacturing  Process En-
gineering, Chevron Research Company, Richmond,
California. He served in process design, process
development, technical service and process planning
at Chevron.  Previous experience includes Chief
Process Engineer at Chevron East Refinery, Perrh
Amboy>  New Jersey from 1967-1972 when he
assumed  his present position assisting corporate
refineries on environmental problems.  He holds a P
a Ph.D. degree in Chemical  Engineering from
Ohio State University.   Mr.  Knowlton is also
a member of the API committee on Refinery Control
and Chairman of the API Solid Waste Management
Committee.
 BIOGRAPHY            Robert N. Slmonsen

        Robert N. Sfmonsen is Senior Environmental Con-
 sultant with The Standard Oil Co. (Ohio) where he has
 been involved in environmental affairs since 1950.  He is
 a past chairman of API's Committee on Refinery Environ-
 mental Control,  and is a member of API's Environmental
 Affairs Water Quality Committee.  He received his B.S.
 in Chemical Engineering from Lehigh University.
 Milton Beychok Blogrdphy on Page  300
 W. Wesley Eckehfefdet Btograghy on Page  476
 Davis Ford Biography on Page
                                        79

-------
                                    BANQUET
Master of Ceremony — Dr. Allen R. Soltow
                      Dean of the Graduate School and Director of Research
                      The University of Tulsa, Tulsa, Oklahoma

Allen Soltow;         "Contrary to your program, my name is Allen Soltow.  I am pinch-
hitting for Dr. John Dowgray, who had to attend a funeral  in Kansas City. First, as
representative of the University of Tulsa, it is my privilege and pleasure to welcome you
to this obviously successful Open Forum and to this banquet.  Before I introduce the head
table, may I  tell you how much the University of Tulsa values its good relations with the
petroleum industry. We have taken two recent steps to expand and enhance these
relationships. First, we have created a National Capital in Energy-Law and  Policy
Institute (NELPI).  We are one of the few university institutes addressing the new and very
complex field of energy litigation.  Dr. Kent  Frizzell, former Secretary of the Interior in
the Nixon and Ford administrations,  is the director of NELPI.  Second, we have formed a
Petroleum and Energy Research Institute (PERI) which serves as an  umbrella organization
for extensive research activities  in petroleum  production and refining.  Frank Manning  is
our PERI director.

                      As your M.C. for this  banquet it is my pleasure to  introduce the
head  table.  Starting at my extreme right, that is your left: -

Milton Beychok,      Consulting Engineer, Irvine, California

Judy  Thatcher,        Representing the API,  one of the sponsors of this symposium

Bill Galegar,         Director, R. S. Kerr Environmental  Research Laboratory, Ada,
                      Oklahoma.  Bill, as you remember welcomed us this morning.
                      Bill is also representing the EPA

Dale  Kingsley,        Our banquet speaker

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

Frank Manning,        Professor  of Chemical  Engineering at Tulsa University.  Frank is
                      Project Director of the EPA Grant financing this  Open Forum.

May I also acknowledge two other people not at the head table.  Ridgway Hall, Associate
Geneml  Counsel for Water,  EPA, and 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.  Fred deserves the credit for engaging  the speakers at this
Open Forum."
                                        80

-------
BIOGRAPHY             Allen R. Soltow

       Dr. Soltow holds a B.A. in Mathematics and
Business Administration from Luther College, and M.S.
and Ph.D. degrees in Economics from Iowa State
University. Dr.  Soltow is currently Dean of the Graduate
School and Director of Research.  Prior to this, Dr.
Soltow has been Chairman of the Division of Economics,
Acting Dean and Acting Associate Dean of the College of
Business Administration at the University of Tulsa.  Dr.
Soltow has been most involved in  both national  and local
community activities and  also participates frequently in
both local and national economic conferences.
                                        81

-------
             SPEECH PRESENTED AT THE WASTE WATER SYMPOSIUM

                                 Dale L. Kingsley
                      Vice President  of Refining, CRA, Inc.

       Thank you,  Mr.  Chairman, and I appreciate the honor to visit with this group
tonight.  It is gratifying to see this large attendance, and it strengthens my belief that
we in business need to meet and exchange ideas, and to determine the best way to
effectively plan and act to improve the quality of our environment.

       We at CRA, and I  believe the other refiners that are represented here tonight,
build plants to supply energy at a competitive cost, but at the same time we  realize that
these facilities must operate in a manner that will not degrade the quality of air and
water in our neighborhoods.  Industry must be a good environmental neighbor because it
is not only good business,  but it is also our public responsibility. We continually meet
and talk about the environment as if  it  were a new world, but actually the earth  has been
in a changing climate since its formation.  It is only in the  last  two decades  that  it has
become obvious that man must correct his wasteful ways if we  are to have a good  lifestyle
in the future. Since man responds slowly to change, laws have  been passed that  will
affect our destiny.  One early law — The River and Harbor Act of 1886 — gently started
us on the road to improved environment.  The Federal Water Pollution Control Act of
1948 was another milestone along this road, and the 1972 Amendment to the  Act  of 1948
was the atomic blast that really got our attention.  We now respond to EPA,  and  Public
Law 92-500, and others that provide  guidelines and goals that we strive to meet.  The
requirements on  waste water quality were established to help,  aid and protect you and me.
But, is it really working out this way?  And, can we afford this protection?  I am
speaking of "we" as industry, as a country, and as individuals.  I suggest we immediately
take a serious look at the pot at the end of the rainbow as related to the 1985 water
quality regulations.  I am wondering  when we get there if there will really be a pot of
gold at the end of the rainbow, or will  it be a pot of I. O.  U.s?  The cost of zero
pollutant discharge  will not be cheap.  Since this is going to be a monumental cost, who
is paying the bill?  I have looked for the  blank check to handle our cost, and have tried
to analyze who is going to honor it.  I  conclude it makes no difference if it is paid for by
industry or by government — in the end, I am paying for it, and I am starting to  wonder
how much non-pollution I can afford.  Have you  looked at it under this light?

       Our industry sponsors a national organization known as the American  Petroleum
Institute. Through this membership, data  are collected recapping the industry's efforts,
costs, and benefits received, relative to waste water management.  The API  summary for
1975 indicates the U. S. petroleum industry spent 629 million dollars for waste water

                                        82

-------
treatment plants, and 356 million dollars for operating costs on these plants.  This is
about 1 million dollars a day for operating a system that does  not bring revenue to the
industry.  I don't know where this operating money comes from, but I feel if the industry
is to survive, this fund will come from products sold. Currently the industry refines about
16 million barrels per day of crude oil.  On this basis,  each barrel of petroleum products
sold  is increased in value by 6$ just to cover the water treatment cost, and  this is only a
starter.  CRA has spent several million dollars in the last two  years to construct waste
water treatment facilities, and I am sure all of you have had comparable expenditures.
Many of these facilities will  meet and slightly exceed the  1977 standards, but they do
not, and let me stress, do not meet the 1985 standards.  It is risky to predict what con-
struction and operating costs will be by 1985, and the degree and amount of equipment
that must be purchased if the industry meets the pollutant-free effluent standards.
Information developed by Batelle in Columbus, Ohio predicts the U. S. petroleum
industry will be spending at least 10 billion dollars a year —  possibly as much as 17
billion dollars annually by 1985 to meet environmental  regulations, and these are 1975
dollars.  If the industry is processing 20,000 barrels per day of crude in 1985,  the
consumer cost for environmental improvement can be as much  as 5%$ per gallon on
today's energy requirements, and again, this is on today's dollar.  This could be inflated
to 10 to 12$ per gallon or more by 1985. The cost exposure of municipal  water treating
systems and other heavy  industries can be equally as great, especially to the user-
consumer.  The Brook ings Institution estimates that compliance with proposed amendments
to the Clean Air and Clean Water Acts by 1985 could cost this nation at least 500 billion
dollars.  This number is so astronomical it is hard to comprehend,  but it works out on the
average to nearly $2,500 per man, woman and child in this country.

        I am not proposing that we cancel our efforts to improve the environment, but
let's look at it with common  sense.  We should continue to improve our water quality —
that is a must!  but do it in such a manner that we reserve capital funds for productive
industrial growth,  maintaining adequate employment opportunities, and preserving  our
quality of lifestyles, and still have an acceptable, healthful atmosphere in which to live.
This is possible if Congress and others who are directing our National  commitments
redirect their efforts to programs that are not wasteful,  but are directionally correct, and
lead us to where we really want to go.  We in industry are morally responsible to provide
this guidance and leadership.  We need to talk to our employees — no, not the ones in
the plants, but the ones  in Washington that we are supporting ~ the ones whom we have
elected and the ones who are working for us. I  encourage you to redouble  your efforts
along these  lines.   Our responsibilities are great and it will take lots of hard work  on our
pant to correct the situation  at hand.

       You have a busy week scheduled.  I wish you success  in your meetings.  I hope
you  will continue to think and direct your efforts toward the amount of pollution abate-
ment that we can afford and  really need.
                                         83

-------
BIOGRAPHY            DaleL.  Kingsley

       Dale L.  Kingsley, Vice President of Refining,
CRA, Inc., holds a B.S. in Chemical Engineering from
the University of Missouri at Rolla.  He started working
at CRA,  Inc., a subsidiary of Farmland Industries, in
1952.  He  currently is responsible for the operation of
CRA refineries located in Coffeyville and Phillipsburg,
Kansas, and Scottsbluff, Nebraska as well as five gas
processing  plants.
                                        84

-------
         SESSION  IV


ORIGIN AND INTERPRETATION


    Chairman

    Marion Buercklin
    Sun  Oil Company,  Inc.
    Tulsa, Oklahoma


    Speakers

    Ridgway M. Hall,  Jr.
    "Regulation of Problem Pollutants Under the  Federal
    Water Pollution Control Decree"

    W.M. Shackelford
    "Evolution of the Priority Pollutant List
    from the Consent Decree"

    R.W. Dellinger
    "Incorporation of the Priority Pollutants into Petroleum
     Refining"

     Leon Myers
     "Generating Problem  Pollutants Data for the EGD
     Document: Refining"

    A. Karim Ahmed
     "Considerations for Defining Substances Hazardous to
     the Environment"
               85

-------
BIOGRAPHY          Marion A. Buercklfn

     Marion A. Buercklin Is Southwest Region
Coordinator of Environmental Affairs for Sun
Company, Inc.  He has a B.S. degree in
Chemistry from the University of Arkansas. He
is currently chairman of the Oklahoma Refinery
Waste Control Council and is active on several
API task forces.
                                      86

-------
          "REGULATION OF PROBLEM POLLUTANTS UNDER THE FEDERAL
          WATER POLLUTION CONTROL ACT: THE 1976 CONSENT DECREE"

                           Ridgway M. Hall, Jr.
                   Associate General Counsel for Water
                   U.S. Environmental Protection Agency
                             Washington, D.C.

I.   INTRODUCTION

    The U.S.  Environmental Protection Agency is presently engaged in a major
regulatory program under the Federal Water Pollution Control Act to limit
the discharges of harmful or toxic pollutants to the Nation's waters.  The
strategy which the Agency devised to implement this program is reflected in
a consent decree entered on June 9, 1976, by the United States District
Court for the District of Columbia, NRDC v. Train. 8 E.R.C. 2120 (D.D.C.
1976), which  settled four lawsuits against the Agency relating to the regu-
lation of harmful pollutants.  By the terms of this decree, EPA has promul-
gated toxic pollutant effluent standards for six highly toxic pollutants,
and is developing effluent limitation guidelines, pretreatment standards,
and new source performance standards for 21 major industries, covering 65
harmful pollutants.

     The 65 pollutants were selected by an EPA task force,*  assisted by
outside consultants.  The selection criteria emphasized primarily the car-
cinogenic, teratogenic, mutagenic, or other toxic properties of the com-
pounds to humans or important aquatic organisms, as well as their presence
in industrial discharges.  The 21 industries were selected based upon data
gathered through the National Pollutant Discharge Elimination System (NPDES)
permit program, contractor reports, and other surveys indicating that each
of these industries discharges substantial amounts of at least some of the
pollutants on this list.

     By gathering data and developing the regulations on an industry-by-in-
dustry basis, EPA expects to provide each industry with a complete package
of regulations covering both existing plants and new sources, with discharges
either directly to the Nation's waters or to publicly-owned treatment sys-
tems.  While  most of the regulations will be primarily technology based,
consideration will also be given to toxicity data and potential human health
effects.  Where the technology-based regulations are inadequate to protect
against adverse effects to human health, drinking water supplies, or the
ecosystem, more stringent limitations will be set.  Finally, in this con-
nection the regulatory scheme, like the statute itself, provides for con-
sideration of State water quality standards.

     It will  be the function of this presentation to describe the process by
which the Agency selected the 65 problem pollutants for which it is com-
mencing the regulatory process under the Act, as well as the 21 major in-
dustries for  which these regulations are being developed.

2.   STATUTORY FRAMEWORK
                                    87

-------
     The starting point for any discussion of EPA's regulatory programs  is
the statutory framework.  EPA is required under the Federal Water Pollution
Control Act to regulate the discharge of pollutants into the Nation's waters.
The goals of this Act include the elimination of the "discharge of  toxic pol-
lutants in toxic amounts", and, by 1985, the elimination of all discharge of
pollutants iato the navigable waters.

     There are numerous sections of this complex 89 page statute which are
designed in various ways to achieve these goals.  Two of the most important
sections are 301 and 304, under which the Agency is directed to develop
technology-based effluent limitations guidelines by which all industrial
point source dischargers are to achieve limitations by July 1, 1977, which
reflect the best practicable control technology currently available or "BPT".
Similar regulations must require achievement by July 1, 1983, of effluent
limitations based upon best available technology economically achievable, or
"BAT".  For new sources, the Agency is required under Section 306 to develop
standards of performance based upon best available demonstrated control
technology.

     These regulations are in turn applied to specific plants through permits
issued under the NPDES permit program established under Section 402 of the
Act.  The permits are issued by the EPA Regional Administrator, or the State
Director of an approved State program, of which there are at this time 27.

     For "indirect dischargers", who discharge to publicly owned treatment
works, the Agency is to develop pretreatment standards under Sections 307(b)
and  (c) for existing and new sources, respectively.

     Finally, for a limited number of highly toxic and persistent compounds,
the Agency may set stringent toxic pollutant effluent standards under Sec-
tion 307(a) with no express requirement that technology or costs be con-
sidered.  These must be met by the affected industry within one year of
promulgation.

     Under these sections the Agency has established effluent limitations
guidelines and new source performance standards for 43 major industries,
covering some 418 industrial subcategories.  For many of these industries
we have also issued pretreatment standards for indirect dischargers.

     During the first few years of our implementation of the Act, the Agency
concentrated primarily on regulating the traditional sanitary parameters:
biochemical oxygen demand, suspended solids, and pH.  Although certain pol-
lutants of demonstrated toxicity, including certain organic compounds and
heavy metals, were also limited, these received secondary emphasis.  More
recently, however, we have all become increasingly aware of the need to
shift our primary emphasis to the regulation of specific toxic pollutants.
This concern has been sparked not only by widespread reports of environ-
mental damage from toxic pollutants, but also by their presence in drinking
water and fish, often at disturbingly high levels, traceable in many cases
to specific industrial discharges.
                                     88

-------
     The Agency has also made use, albeit limited, of Section 307(a).  Be-
cause that section embodies a pollutant-by-pollutant approach rather'than an
industry-by-industry approach, widespread use of this section would confront
corporate decision-makers with a moving target.  It can be disruptive and
costly to industry to invest substantial sums in treatment technology for
one toxic pollutant, only to be told a few months later that another toxic
pollutant is to be regulated, forcing the installation of other technology
which might or might not be consistent with what was installed for the first
pollutant.  The short one-year compliance time also imposes limitations on
the use of Section 307(a), given the length of time which it takes to install
sophisticated control technology.

     Not surprisingly therefore, the Agency has promulgated toxic pollutant
effluent standards under Section 307(a) for only six pollutants:  Aldrin/
Dieldrin, DDT, Endrin, Toxaphene, Benzidine, and Polychlorinated Biphenyls
(PCB's).  At the present time, it is expected that 307(a) will be held in
reserve for those situations where the toxicity or public health threat is
such that prompt and stringent control is required regardless of the avail-
ability of control technology.

     In light of this experience, the Agency concluded in October, 1975,
that if broad and effective regulation of problem pollutants is to be
achieved, it must be accomplished on an industry-by-industry basis.  This
meant that the Agency would expand its use of Sections 301, 304, and 306,
as well as the pretreatment authorities in Section 307 (b) and (c), in con-
trolling these pollutants.

     Incidentally, this approach of using technology-based regulations to
control the discharge of toxic pollutants by 1983 recently received strong
support from President Carter.  In his environmental message to Congress on
May  23, he said:

              I have instructed the Environmental Protection
              Agency to give its highest priority to develop-
              ing 1983-best-available-technology industrial
              effluent standards which will control toxic
              pollutants under the Federal Water Pollution
              Control Act, and to incorporate these standards
              into discharge permits.

3.   DEVELOPMENT OF PRIORITIZED LIST OF POLLUTANTS

     As the first step in developing a strategy for regulating these "toxic"
or "problem" pollutants, the Agency decided in the Fall of 1975, to assemble
a work group to develop a prioritized list of toxic pollutants which should
receive primary regulatory attention under the Water Act.  The members of
the  group included staff scientists from virtually all of the Agency's
offices and divisions having responsibilities or interests in this area,
with the Office of Water Planning and Standards taking the lead.  Among the
members were public health specialists with expertise in areas of toxicology
including carcinogenicity, chemists, biochemists, biologists, engineers, and
a statistician.  Two outside scientific consultants were engaged to assist


                                     89

-------
in the project, and additional review and consultation was provided from such
sources as the National Cancer Institute and the Oak Ridge National Labora-
tory.

     The objectives of the group were, first, to develop criteria  for  the
selection and prioritization of toxic pollutants, and second,  to relate  those
to the discharges of major industrial sources, and thereby derive  a list of
pollutants appropriate for regulation under the Act.

     The group began by an exhaustive literature search, including lists of
pollutants prepared by other organizations with expertise in this  area.   We
focused our attention on both the hazard of the pollutant and  the  degree of
actual or likely exposure to humans or wildlife as a result of industrial
discharges.

     The basic criteria which this task force used to evaluate the potential
hazard of a toxic pollutant were: (1) evidence that the compound,  or its
degradation products and metabolites, pose an actual or potential  health
hazard, based on laboratory or human evidence that the substance produces
carcinogenic, mutagenic, or teratogenic effects, adverse effects on repro-
duction or behavior, or adverse effects on any organ system, (2) persistence,
(3) ability to bioaccumulate in organisms, and (4) evidence of synergistic
propensities for these toxic effects.  In addition, the task force  took into
"consideration all available evidence that the toxic pollutants might cause
lethal or sublethal effects on wildlife, particularly aquatic organisms and
those species dependant upon aquatic organisms for their food supply.

     With respect to exposure, we considered the following factors, wherever
such information was available: (1) total production of the substance, (2)
use patterns, (3) estimated extent, both qualitative and quantitive, of
actual or likely point source water discharges, (4) the consequent  exposure
of man and wildlife to the substance or its breakdown products and meta-
bolites, and (5) analytical methodology capabilities.

     Among the more significant reports which we reviewed were the  follow-
ing:

              "Water Quality Criteria, A Report of the National
              Technical Advisory Committee to the Secretary of
              the Interior", U.S. Government Printing Office,
              Washington, D. C. (1968).  ("Green Book")

              "Water Quality Criteria, 1972".  National Academy
              of Sciences, and National Academy of Engineering,
              U.S.  Government Printing Office, Washington, B.C.
              (1974).   ("Blue Book")

              FWPCA Section 311 Supplement to Development Docu-
              ment  on Hazardous Substances (EPA November 1975).
              EPA 440/9-75-009.
                                     90

-------
             "A Report Assessment  of  Health Risk From
             Organics in Drinking  Water by an Ad Hoc
             Study Group to  the Hazardous Materials
             Advisory Committee,"   Science Advisory
             Board, EPA, April  30,  1975.   ("Drinking
             Water Report")

             "Identification of Organic Compounds in
             Effluents From  Industrial  Sources"  Versar,
             Inc., EPA-560/3-75-002,  April 1975.

             "Final Report of National  Science Founda-
             tion Workshop Panel to Select Organic Com-
             pounds Hazardous to the  Environment."
             (Developed in conjunction  with Stanford
             Research Institute) Dr.  Norton Nelson et al.,
             NSF, October 1975.

             "Preliminary Assessment  of Suspected Car-
             cinogens in Drinking  Water", EPA Office  of
             Toxic Substances,  June 1975.

             "Organics in Drinking Water: Listing of
             Identified Chemicals  Part  I".  Junk, et  al.,
             Ames Lab, ERDA,  Iowa  State University,
             July 1975.   ("Iowa Report")

             "Registry of Toxic Effects of Chemical Sub-
             stances" National  Institute for Occupational
             Safety and Health  (NIOSH)  (1975 ed.) and
             "Suspected Carcinogens"  A  Subfile of the NIOSH
             Toxic Substance List" (NIOSH, 1975  Ed.).

             Oak Ridge National Laboratory Environmental
             Mutagen Information Center Study dated Nov.
             7, 1975.

             "International  Agency for  Research  on Cancer
             Monographs on the  Evaluation of Carcinogenic
             Risk of Chemicals  to  Man", World Health  Or-
             ganization  (1972-1973).

             "Survey of Compounds  Which Have Been Tested
             for Carcinogenic Activity",  P.H.S.  Publica-
             tion No. 149, USHEW.

     After reviewing the work of others  in this area,  several  things became
apparent.  First, while these studies  provided much helpful information,
none provided a direct answer to the difficult question of how to select and
prioritize toxic pollutants solely  from  an industrial  point source effluent
discharge perspective.  Second,  our initial selection  criteria mentioned
above, would almost certainly undergo  an evolution, or refinement, as we


                                    91

-------
began to apply them to raw data and evaluate the results.

     The work group began its selection process by examining as large a num-
ber of compounds with potentially hazardous effects in water as would be
practical and feasible.  The National Science Foundation, for example,
initially looked at 337 organic compounds, later narrowing this list to 80
deemed to have the highest interest for further research.  This was somewhat
more helpful than the NIOSH toxic substance list, which indicates that the
number of unique substances for which toxic effects information may be avail-
able is approximately 100,000.

     As a starting point the work group assembled a list of some 232 com-
pounds believed to be actually or potentially hazardous, and present in
water.

     After having gathered as much information as possible with respect to
the initial working list of compounds, the work group determined that an
initial screening procedure was necessary in order to reduce the list to
those compounds requiring highest priority for in-depth review.  The ini-
tial screening process employed the following selection criteria:

              (1)  Evidence of actual presence in effluent;

              (2)  Evidence of carcinogenic, mutagenic, or
                   teratogenic effects in laboratory test
                   systems, or human epidemiological studies;
                   or evidence of a high degree of toxicity
                   to aquatic organisms or systems.

Application of these criteria reduced the initial list of 232 to approxi-
mately 75.  This screening procedure produced a workable group of compounds
for which there was both some evidence of point source discharge and for
which there was evidence of serious potential hazard to man or the environ-
ment.

     This screened list of compounds and underlying data, as well as the
criteria, were then subjected to further analysis and quality control
evaluation.  This led to a further refinement of both the list and the
criteria, with some compounds dropped and others added.

     In applying the criteria to our initial list of compounds little ef-
fort was actually made to incorporate human epidemiological information
dealing with possible acute toxic effects.  This was due primarily to the
fact that the main purpose of the regulations for which the list was being
developed was to protect against serious adverse effects which might be
posed by long term, or chronic exposure to these pollutants.

     With regard to exposure, in order to determine whether or not a
particular substance is present in industrial effluent discharges, the work
group investigated all data sources available to the Agency.  Only those
compounds for which there was information indicating actual presence in
point source effluent, or strong likelihood that the compound is present,


                                    92

-------
were considered candidates for priority consideration.  In determining which
compounds are present in point source discharges,  the work group undertook
the following steps.

     First,  EPA's Effluent Guidelines Division compiled a "direct discharge
emissions inventory", which in turn was followed by a "discharge parameter
matrix" in which the pollutants known or strongly  believed to be discharged
were listed  by industry, coupled with such information as was available on
quantification.  The principal sources of the raw  data were permits and per-
mit applications, monitoring data, and information gathered by EPA contrac-
tors in the  course of their developing the required data base to support
existing effluent limitations guidelines.

     Second, we prepared an "effluent load reduction study", based on a sam-
ple of 1800  national pollutant discharge elimination system permits, which
list the maximum allowable discharge for various pollutants.  An estimate
was made of  the quantity of each pollutant discharged per day for each of
the regulated 43 industrial point source categories.  By summing these
estimates for each pollutant, a rough estimate of  the total discharge for
point sources could then be made for many of the pollutants.  Even when sup-
plemented by monitoring reports, there are statistical and other limitations
on this type of data, and therefore it was used with caution.

     In addition, the work group reviewed the several studies mentioned
earlier on the presence of many of the selected compounds in surface water
as well as drinking and well water.

     With respect to food particularly fish and fish products) and humans,
additional monitoring sources were examined.  These included specifically
the FDA Market Basket Survey and the EPA Human Monitoring Survey, which were
notably helpful regarding pesticides and heavy metals.

     The prime source for data on annual production was the Stanford Research
Economic Handbook, which apparently derives a good deal of its information
from Federal Trade Commission data as well as from industry surveys. Rela-
tively little data were obtained on use patterns.  For the purposes of this
selection process, however, it was less important  to know the ultimate use
than it was  to know the source and amount of the industrial discharge of
these compounds.

     With respect to the final "exposure" criterion, "analytical methodology
capabilities", the work group ascertained that there were and are currently
available analytical methods capable of detecting  all of the 65 high priority
compounds and/or generic categories on our final lists, although the degree
of confidence in the analytical methods used to detect these pollutants
varies somewhat especially at low concentrations.

     The group completed its work in February, 1976.  Its final work product
was four lists, consisting of 29, 18, 18, and 11 substances respectively.
The 29 pollutants on the first list which has the  highest priority, meet the
following criteria:
                                     93

-------
A.  The pollutants are known to occur in point  source  effluents,  in aquatic
    environments, in fish, or in drinking water.

B.  There is substantial evidence of carcinogenicity,  mutagenicity and/or
    teratogenicity in human epidemiological studies  or in  animal  bioassay
    systems.

C.  It is likely that point source effluents contribute  substantially  to
    human hazard, at least locally.  This latter judgment  is based on  con-
    sideration of the quantities emitted, the persistence  of the  compounds
    in aquatic systems, their tendency to be stored  in organisms  used  for
    human food, and available information on effective doses in animal tests.

This first list contained the following substances:

                                  LIST 1

     1.  Acenaphthene
     2.  Aldrin/Dieldrin
     3.  Arsenic compounds
     4.  Asbestos
     5.  Benzene
     6.  Benzidine
     7.  Beryllium compounds
     8.  Cadmium compounds
     9-  Carbon tetrachloride
    10.  Chlordane (technical mixture and metabolites)
    11.  Chloroalkyl ethers (chloromethyl, chloroethyl,  and mixed ethers)
    12.  Chloroform
    13.  Chromium compounds
    14.  DDT and metabolites
    15.  Dichlorobenzenes (1, 2-, 1, 3-, and 1, 4-dichlorobenzenes)
    16.  Dichlorobenzidine
    17.  Diphenylhydrazine
    18.  Heptachlor and metabolites
    19.  Hexachlorocyclohexane (all isomers)
    20.  Lead compounds
    21.  Mercury compounds
    22.  Nickel compounds
    23.  Nitrosamines
    24.  Polychlorinated biphenyls (PCBs)
    25.  Polynuclear aromatic hydrocarbons (including  benzanthracenes,
                                            benzopyrenes,  benzofluoranthene,
                                            chrysenes, dibenzanthracenes,
                                            and indenopyrenes)
    26.  2,3,7,8 - Tetrachlorodibenzo-p-dioxin  (TCDD)
    27.  Thallium compounds
    28.  Trichloroethylene
    29.  Vinyl chloride


                                    94

-------
    The 18 pollutants on the second highest priority list satisfy criterion
A, above,  i.e.,  occurrence in point source effluents, in aquatic environ-
ments,  in fish,  or drinking water.  However, in some cases the evidence of
carcinogenicity, mutagenicity, and teratogenicity is based primarily upon
structual similarity to compounds in List 1, or upon mutagenic activity in
bacterial screening systems; or testing has shown some evidence of carcino-
genicity,  mutagenicity, or teratogenicity,  but these results presently ap-
pear to be incomplete or equivocal.  In addition, a few compounds are in-
cluded in List II on the basis of serious toxic effects other than carcino-
genicity,  mutagenicity, or teratogenicity.  These other serious toxic effects
are, for the most part, on aquatic organisms.  Finally, the possibility of
significant human exposure attributable to point source effluents was judged
to be somewhat less than that for compounds in List I.  (This judgment was
based on relatively small volume of discharges, or relatively low propen-
sity to persist in water or to accumulate in organisms.)

    The substances on List II are as follows:

                                 LIST II

      1.   Chlorinated benzenes (other than dichlorobenzenes)
      2.   Chlorinated ethanes (including 1,2-dichloroethane,  1,1,1-tri-
                               chloroethane,  and hexachloroethane)
      3.   2-chlorophenol
      4.   Dichloroethylenes (1,1-and 1,2-dichloroethylene)
      5.   2,4-dichlorophenol
      6.   2,4-dimethylphenol
      7.   Dichloropropane and dichloropropene
      8.   Endosulfan and metabolites
      9.   Endrin and metabolites
     10.   Fluoranthene
     11.   Haloethers (other than those in List I;  includes chlorophenyl
                      ethers, bromophenylphenyl ether,  bis(dischloroiso-
                      prophyl) ether, bis-(chloroethoxy) methane and poly-
                      chlorinated diphenyl ethers)
     12.   Halomethanes (other than those in List I;  includes methylene
                        chloride, methylbromide, bromoform, dichlorobro-
                        momethane, trichlorofluoromethane,  dichlorodifluoro-
                        methane, methylchloride)
     13.   Hexachlorobutadiene
     14.   Naphthalene
     15.   Pentachlorophenol
     16.   Phthalate esters
     17.   Tetrachloroethylene
     18.   Toxaphene

     With  respect to the 18 compounds on priority List  III, all satisfy
criterion  A:  occurrence in aquatic environments, in fish,  or drinking water.
In addition,  all are known to have toxic effects on human or aquatic organ-
isms at relatively low concentrations.   However, there  is no substantial
evidence  that these compounds have primary carcinogenic, mutagenic,  or tera-
togenic effects.   List III contains the following compounds:


                                    95

-------
                                 LIST III

      1.  Acrolein
      2.  Acrylonitrile
      3.  Antimony compounds
      4.  Chlorinated naphthalene
      5.  Chlorinated phenols (other than those in List II;  including
                               trichlorophenols and chlorinated  cresols)
      6.  Copper compounds
      7.  Cyanides
      8.  Dinitrotoluene
      9.  Ethylbenzene
     10.  Hexachlorocyclopentadiene
     11.  Isophorone
     12.  Nitrobenzene
     13.  Nitrophenols (including 2,4-dinitrophenol, dinitrocresol)
     14.  Phenol
     15.  Selenium compounds
     16.  Silver compounds
     17.  Toluene
     18.  Zinc compounds
     The 65 compounds on these 3 lists are receiving the highest priority in
the Agency's regulatory programs currently being implemented under the Fed-
eral Water Pollution Control Act.  They are listed together as Appendix A to
the consent decree issued in June, 1976, by the U. S. District Court for the
District of Columbia which was mentioned earlier.
     In addition to these three lists, the work group developed a fourth
list of compounds which is receiving secondary regulatory attention by the
Agency.  Compounds on this list are also known to occur in effluents or
drinking water.  On the basis of existing evidence, they are judged to pre-
sent a less substantial direct hazard than the chemicals on Lists I-III.
However, there is reason to believe that they may be converted in the en-
vironment into derivatives or breakdown products which may be regarded as
precursors of hazardous compounds rather than major hazards in themselves.
List IV originally consisted of 11 items, and chlorine has since been added.
Under the consent decree, EPA is to gather data on these 12 pollutants.  But
is not obligated to issue regulations for them.  This list is as follows:
                                     96

-------
4.   SELECTION OF 21 INDUSTRIES

    For  the 43 industrial categories for which EPA has already issued efflu-
ent limitations guidelines in connection with the first round of permits
directed towards the July 1, 1977, date, considerable data were obtained on
which industries were discharging which pollutants, and in what volume.  The
principal sources of this data include the NPDES permit applications, moni-
toring reports, "Development Documents" and other surveys and analyses pre-
pared by EPA's contractors in connection with the development of the first
group of regulations, sampling and analysis data, and other reports.  The
staff report of the National Commission on Water Quality was also considered.

     The 21 industries which were selected for the primary regulatory effort
have all been found to be discharging a substantial number of pollutants on
the list of 65, and in most instances the volumes of such discharges are
also substantial.

     There was nothing magic in the number 21.  The Agency did prepare a
rough prioritized ranking of the 43 industries, in the order of discharges
of number and volume of pollutants.  The selection of the top 21 of these
industries reflects nothing more sophisticated than hard negotiations be-
tween the Agency and the plaintiff environmental groups in the four lawsuits
mentioned above.  The environmental groups sought a high number to assure
maximum coverage.  The Agency sought a lower number to preserve maximum
flexibility.  In fact there is no question but that regulations ought to be
developed for these 21 industries.  Moreover, to the extent that resources
permit,  the Agency may well go beyond the 21 industries and develop compar-
able regulations for other industries as well.

     The order in which the industries appear on the Agency's present work-
ing list is not the same as the original priority order.  The present order
reflects to a large degree the availability of contractors and EPA project
officers, and similar administrative considerations.  This list of indus-
tries is set forth below.
                                 LIST  IV
      1.  Acetone
      2.  n-alkanes  (Cin - C
      3.  Biphenyl
      4.  Dialkyl ethers
      5.  Dibenzofuran
      6.  Diphenyl ether
      7.  Methylethyl ketone
      8.  Nitrites
      9.  Secondary  amines
     10.  Styrene
     11.  Terpenes
     12.  Chlorine
                                     97

-------
                     INDUSTRIES FOR WHICH REGULATIONS
                        WILL BE DEVELOPED UNDER THE
                      FEDERAL WATER POLLUTION CONTROL
                     ACT TO CONTROL PROBLEM POLLUTANTS
      1.  Timber Products Processing
      2.  Steam Electric Power Plants
      3.  Leather Tanning & Finishing
      4.  Iron & Steel Manufacturing
      5.  Petroleum Refining
      6.  Nonferrous Metals Manufacturing
      7.  Paving & Roofing Materials
      8.  Paint & Ink Formulation & Printing
      9.  Ore Mining & Dressing
     10.  Coal Mining
     11.  Organic Chemicals Manufacturing
     12.  Inorganic Chemicals Manufacturing
     13.  Textile Mills
     14.  Plastic & Synthetic Material Manufacturing
     15.  Pulp & Paperboard Mills
     16.  Rubber Processing
     17.  Soap & Detergent Manufacturing
     18.  Auto & Other Laundries
     19.  Miscellaneous Chemicals
     20.  Machinery & Mechanical Products
     21.  Electroplating


5.  SUBSEQUENT REFINEMENT OF LISTS

    Each of these lists is subject to refinement, and the consent decree pro-
vides the necessary flexibility to do so.  Industries or industrial subcate-
gories may be added or deleted.  Scientific knowledge never stands still.
It may be that as the program is implemented, some substances will be added
and others deleted.  While great efforts and talents were brought to bear in
selecting the original 65, no one suggests that this list is perfect.

     It should be noted that some of the items on this list are actually
families of compounds.  As of this date, in implementing the program, EPA
has broken out the families into individual substances, and has further
prioritized within these families.  This has resulted in a refined working
list of 129 substances.

     The process by which this further refinement has been-accomplished will
be discussed next by Walter M. Shackelford of EPA's Athens, Georgia, Labora-
tory.

     This concludes my presentation.  I will be glad to answer any questions
you may have.
                                    98

-------
                                REFERENCES
     The author wishes to express appreciation to the Chairman of this task
force,  John C.  Kolojeski, who was formerly with EPA's Office of General
Counsel and Office of Water and Hazardous Materials, and is now President
of Clement Associates, Inc., scientific regulatory consultants in Washington,
D.C., for his assistance in documenting various activities of the task force
as discussed in this paper.
BIOGRAPHY

   Ridgway M.  Hall,  Jr.  is Associate
General Counsel,  Water Quality Divi-
sion, of the  Environmental Protection
Agency. He  is responsible for ad-
vising and representing the Agency on
all legal  matters relating to the
quality of the nation's waters, which
includes primarily issues arising un-
der the Federal Water Pollution Con-
trol Act and the  Safe Drinking Water
Act.  He received his B.A. degree from
Yale University in 1963 and his L.L.B.
degree in  1966 from Harvard Law School.
He is a former partner in the law firm
of Cummings  and Lockwood of Stamford,
Connecticut, where he practiced law
from 1966-1975.  Mr. Hall is a member
of the American Bar Association Sec-
tions on Litigation and Natural Re-
sources Law.
                                    99

-------
DISCUSSION

Richard B. Costa:  (CEC Thompson Engineering, Inc. Houston, TX)   How many
substances on the "list of priority pollutants" are common to both  industrial
and municipal effluents?

Ridgway Hall:  We do have some information in that area but not  a lot.  We
know that a lot of them are coming out of POTW's.  The major effort in
developing these lists was on direct discharges.  We then sought to deter-
mine how much of what goes into the public owned treatment works in turn
passes through.  We have only the roughest kind of data in that  area.  We
know that a number of them do pass through some plants to some degree, but
there is still more data gathering to do in that area, and frankly  the dis-
charges from the POTW's were of secondary priority to the work group.  We
will be gathering data in that area as we go about implementing  this settle-
ment decree in connection with the development of pretreatment standards.
The principal thrust of the pretreatment standards is to require whatever
steps are necessary to prevent either upset of the POTW, or pass-through
by harmful pollutants in harmful amounts.  Where we find these problems, it
will indicate to us that those pollutants should be controlled by appro-
priate pretreatment standards.

Mr. Costa;  When would you expect that to be available?

Mr. Hall:  It will vary from industry to industry and follow to  some extent
the schedule which is outlined in the consent decree for the contracts.  As
you probably know there are five series of dates for the issuance of con-
tracts, the development of proposed regulations, and promulgation.  The
earlier industries on the administrative timetable will probably see that
data start falling out sooner.  Most of the contracts have already  been let,
and people are in the process of gathering data.  I would suspect that some
of that data will be available in a good deal less than a year and  perhaps
for some of those early industries as soon as 7 or 8 months from now.  The
thing to do is contact the project officers on those if you're interested
in that kind of data.

A. Karim Ahmed;  Will you please explain the list of twelve compounds which
you showed on your list No. 4 on the screen?  I'm a little unclear  about its
relation to the implementation of the Consent Decree.

Ridgway Hall;  That list survives now as Appendix C to the Settlement Agree-
ment and it's governed specifically under paragraph 4(b) of the  Settlement
Agreement for those of you who follow those sort of things with  a fine
tooth comb.  Under the Decree, as the Agency goes about its industry-by-
industry data-gathering and sampling - analysis efforts, wherever we find
any of those 12 compounds present in effluents we will gather data.  We
will record it so that we will know who is putting out which of  the com-
pounds on that list and in what amounts.  It will therefore give the Agency
a good data base so if we later find out that the compounds are  indeed
causing problems either through synergistic reactions, or breakdown or
metabolite reactions, we'll have the data in hand to go out and  promptly
develop regulations to deal with those problems rather than having  to come


                                     100

-------
back a second  time and bother these 21 Industries again with another massive
data gathering effort.

Peter Foley;   The President's message also had a recommendation for changes
to Section 307.  Could you give us some insight what problems the Agency
has with that  Section and what amendments they will be seeking?

Ridgway Hall;   Yes, I'd be glad to.  The principal problem that we have with
307 is the fact that it basically is a pollutant-by-pollutant approach rather
than an industry-industry approach.  Another problem is that both the stan-
dard-setting  time frame and the compliance time are very short.  But the
biggest limitation on its usefulness is that it proceeds on a pollutant-by-
pollutant basis.  It is inherently disruptive when you are trying to develop
a program which will give   major coverage to industries for a lot of pol-
lutants.  You're trying to regulate and get controls in place for a large
number of toxic pollutants.  As those of you who are corporate managers and
decision makers know, if the Agency can come to you and say, based on   data
which together we jointly assembled, "The following controls appear sensi-
ble and necessary to control everything in your effluent that looks as though
it's going to be harmful", you can then go and plan on that.  You can buy
the technology, you can install it, and you've got a reasonable installation
time.  You don't have the moving target problem of EPA coming down the road
3 or 6 months later with "Oh, by the way, we just found mercury in your
effluent, you've got to put the following technology in for that."

     I think the other troublesome features with it are as follows.  First,
the Agency was told to develop a list of pollutants for regulation under
the section within 90 days after passage of the Act.  It was then told to
develop toxic effluent standards within 180 days after that, and then run
formal rulemaking hearings, and within 6 months after proposing the stan-
dards to promulgate them.  Then industry has only 1 year to comply with
those standards.  That's a very rigid timetable for the regulatory program
in that we have to list the pollutants practically before we go out and
gather the data to find out if the pollutants are really troublesome or not.
Furthermore,  one year is an awfully short time for industry to install
meaningful control technology=

     I think therefore that the kinds of amendments which you might see are,
for one thing, language that would alter the regulatory trigger mechanism
by disconnecting the section from 90 days following passage of the Act,
which is long since past.  While I cannot speak for the Administration at
this point, in my personal view the time for listing a pollutant under
Section 307 (a) should be when information comes to the attention of the
Administrator on the basis of which he deems it appropriate to regulate it.
This way you gather the data first and then make the regulatory decision.
There is also discussion about switching the kind of rulemaking from the
formal adversary trial-type proceedings to the more informal rulemaking
such as you see under Section 307(b) for pretreatment standards.  This would
be notice and comment rulemaking with the option of a hearing, if you want
it.  This was the procedure essentially set out by Congress last year when
it passed the Toxic Substances Control Act.
                                     101

-------
     There was a bill which was passed by the House earlier this year which
would have allowed some flexibility to extend the one-year compliance time
up to three years if the Administrator determines that a further extension
beyond the one year is necessary because of technological reasons.  That
bill died in conference, but I suspect it will come up again.  The adminis-
tration might want to include a provision on that amendment that would direct
the Administrator to consider any adverse effects on human health or the en-
vironment in determining whether or not to extend the period beyond the one
year.  One can envision the situation where technology might be installable
let's say 2 years down the road but the human health hazard might be so bad
you would want to shorten the time notwithstanding the temporary unavail-
ability of technology.  Those are the kind of amendments that I would fore-
see.
                                    102

-------
                EVOLUTION OF THE PRIORITY POLLUTANT LIST
                          FROM THE CONSENT DECREE

                       W. M. Shackelford and L.  H. Keith*
                      U. S. Environmental Protection Agency
               Environmental Research Laboratory, Athens, Georgia

INTRODUCTION

       The Consent Decree of 7 June 1976 (1) required the  Environmental Protection
Agency to assess industrial wastewater with respect to 65 compounds and compound
classes. To provide chemical analysis data at a reasonable cost, however,  the compound
classes listed in the Consent Decree had to be resolved  into  representative individual
compounds.  The factors to be considered  for such a resolution procedure  include the
known frequency of occurrence of the compound in water, the number of  manufacturing
sites and quantity of material manufactured, and the availability of a reference standard.
Once the compound classes have been resolved by EPA, generation of a protocol for
screening analysis of a variety of effluent types becomes a problem that may be
quantified in terms of required analysis procedures such as the number of extractions,
instrument runs, etc.

CONSENT DECREE COMPOUNDS

       For chemical analysis, the compounds listed in the Consent Decree can be
divided into three broad groups.  The largest group is composed of those compounds and
classes of organic chemicals that are amenable to analysis by gas chromatography-mass
spectrometry.  Of the 50 listed in this group, 13 are classes of compounds containing as
few as two and as many as several  hundred known members.  The second group consists of
13 metals and their compounds.  As with the first group, each of the  13 compound
classes in this group could be construed to contain hundreds  of constituents.  The
elements of this group may be analyzed individually by plasma emission spectroscopy,
spark source mass spectrometry, or some other method of multielement analysis.  The
final grouping consists of cyanide and asbestos, two substances that require  specialized
analysis procedures not applicable to groups of materials.

       Obviously, the  challenge of chemical analysis of a sample for literally hundreds
of components is staggering — especially  when it is considered that these components
may be only trace substftuents (parts per billion) in the overall  sample matrix.  The


*Present Location:  Organic Chemistry Dept., Radian Corporation, Austin,  Texas 78760

                                       103

-------
drain on resources in governmental laboratories as well as laboratories in the private
sector in terms of manpower and money would be overwhelming if analysis were attempted
of effluent samples for the compounds and compound classes of the Consent Decree when
construed to their fullest extent. For these reasons it was deemed necessary to resolve
the list  of 65 Consent Decree compounds and classes into a list of individual  compounds
that represents faithfully each multicomponent class in the original list of 65.  The
specific compounds mentioned in the Consent Decree,  however, were automatically
passed through to the resolved list without  further question.  In addition, because
technology for identification of individual  metal compounds, particularly organometallic
compounds, is not well developed, the metals were to be analyzed without regard to
oxidation state.  This segregation of compounds  leaves the problem of defining criteria
to properly choose representatives of the remaining broad  chemical classes.

CRITERIA USED  FOR RESOLVING LIST

        To resolve the  remaining Consent Decree compounds into a list that presents  an
analytical  problem of finite limits requires  that decisions be made that address the spirit
and intent of the Consent Decree.  For instance, it seems plausible that in deciding
which members of a compound class adequately represent the whole class, those
compounds that  have little chance of finding their way into industrial effluents should be
of low priority.  Those compounds,  however,  that are manufactured in quantity, are
used in  manufacturing  processes, or are by-products of other processes would get high
priority as group representatives.

        Another aspect that may be used to prioritize the compounds that represent a
group is the previous occurrence of a given compound in an analysis of water samples.
If a candidate for group representation has  not been identified in water in some  previous
study, whereas another candidate has been previously identified, the priority for group
representation should be assigned to the latter candidate.   The fact that a compound has
been identified  in water previously is an indication that it is apt to be present in
additional  similar studies.

        A final aspect  that must be considered in prioritizing chemical group represen-
tatives  is the availability of analytical reference standards.  Identification and  quantifi-
cation of chemical compounds depends upon the use of reliable standards.  If a reliable
standard of a candidate compound is not obtainable, quantitative analysis is  impossible.

DATA USED IN  RESOLVING LIST

        The data used to resolve the Consent  Decree list of compound classes falls into
three sets that are consistent with the above rationales.  First, an EPA study  of organic
compounds identified in water was used to  provide the data on the compounds previously
detected in surveys of water (2). Second,  data  concerning the manufacture of compounds
was gathered fronrv previously conducted surveys  of the chemical industry (3,4).   Finally,
catalogues from  various chemical supply houses were consulted to determine whether
analytical  standards were  available (5-14).

                                         104

-------
       Data previously gathered through analyses of various water types throughout the
world provided the most useful basis for identifying the candidates for group represen-
tation on the basis of their known existence and known amenability to state-of-the-art
analysis techniques.  Of the original 65 compounds and classes,  5 had never been
detected in water samples.  Moreover, 2 of the 13 groups of organics had no represen-
tatives that had been previously found in water.  In the remaining 11 groups,  however,
priorities could be set according to the past occurrence of a given group member in
water.  Table 1 lists the 13 groups and their known frequency of occurrence in water.

       Of the individual members of a group that had been found in water previously,
those that represented less than 5% of the total occurrence were removed from con-
sideration in the interest of reducing analysis costs.  Table 2 shows the 13 groups and the
number of compounds that had been found in water within each group.  Also listed is the
number of compounds within each group that met the greater than 5% occurrence
criterion.

        Manufacturing data help prioritize candidate compounds as representatives of
chemical groups by indicating the amount of a given chemical that the environment may
have to absorb.  Of the 13 groups of organics that were listed in the  Consent  Decree,
only 9 contained compounds that were manufactured in quantity.  Compounds in these 9
groups that had no known manufacturers were given lower priority than those for which
production data existed.  Table 3 lists the  13 groups of compounds and the number of
compounds for which production data existed.  The production data overlapped with the
frequency data in most cases except that two chlorinated naphthalenes and one nitro-
samine are produced but have  never been found in water.  N-nitrosodiphenyl amine, the
rubber chemical, and  2-chloronaphthalene were added to the list on  the basis of
production.  The other materials had already been included because they met the
frequency criterion.

        The availability of analytical reference standards for the candidate compounds
was determined by searching chemical supplier catalogues.  As expected, standards for
most of the compounds that are manufactured or have been  found in water previously
were readily available from a  number of chemical supply houses.  Where possible, non-
commercial sources were  found for compounds mentioned specifically in the Consent
Decree. Those compounds that were only candidates for group representation, however,
were dropped from consideration.  Table 4 lists the 13  groups and the total number of
considered candidates in  each group along with the number eliminated because of lack of
a standard.

GENERATION OF AN ANALYTICAL PROTOCOL

        The reason for resolving the Consent Decree list was to make  possible  the analysis
of industrial effluents for these compounds  and classes of compounds within the constraints
of available time and  money.   EPA's Environmental Monitoring and Support Laboratory
in Cincinnati was given the responsibility of developing an analysis protrocol for
screening the effluents for the compounds on the resolved Priority Pollutant List. The

                                         105

-------
Athens Environmental Research Laboratory was asked to provide input for the screening
analysis protocol for those organic compounds not amenable to the purge and trap
(Bellar-Lichtenberg) technique.  Once the Priority Pollutant List had been evolved,
establishment of the analysis protocol for screening became a problem of applying known
methods and techniques to the specific set of compounds on the Priority Pollutant List.

       In this case, the development of an analysis protocol also involves the consider-
ation of two aspects that are many times not considered in survey analysis.  First, since
the actual  laboratory work is more than any one contractor could handle,  provisions must
be made to insure the compatibility of data from one laboratory to the next.  Second,
this protocol must be limited to looking for only the priority pollutants. The method
used (computerized GC-MS),  however, would also see—were there time to interpret the
data—some of those compounds eliminated as class representatives and others that at
some  later date may be of interest.

       Even though it is recognized that each effluent will present its own set of unique
problems, the protocol provides a measure of compatibility by giving a general approach
to follow and specifying techniques within the general approach.  For instance,  gas
chromatographic column packing  materials are specified and a common lot is to be used
by all laboratories involved in the screening analysis.

       To save data that will  allow the future determination of compounds not on the
Priority Pollutant List involves two factors.  First,  the analysis protrocol must be general
enough to allow the detection of  compounds other than those on the list.  This require-
ment  is satisfied by using extraction,  concentration and detection techniques that do not
discriminate against any compounds that can be gas chromatographed—essentially state-
of-the-art survey analysis with mass spectrometric detection.  Second, a method must be
developed for long-term storage of the raw data. This requirement is satisfied by
coupling a computer system  directly to the mass spectrometer so that raw data may be
transferred to magnetic tape.

CONCLUSIONS

       In listing classes of compounds rather than specific compounds, the Consent
Decree presented a problem that to the analytical chemist seemed without practical
limit. By using  available data concerning previous detection in water, manufactured
quantities, and availability of standards, however, the ambiguous compound classes can
be replaced with specific compounds of high priority.  Once the specific  compound list
has been resolved,  however, the  necessity for interlaboratory comparability of data and
effective ways to save all raw data becomes even more acute in the development of the
final  protocol.

REFERENCES

1.     Consent  Decree, U.S. District Court for the District of Columbia, 7 June 1976.
2.     Shackelford, W.M.  and L.M. Keith.  Frequency of Organic  Compounds

                                        106

-------
       Identified in Water, U.S. Environmental Protection Agency, Athens, GA
       Publication  No. EPA-600/4-76-062. December 1976.
 3.     1976 Directory of Chemical  Producers, USA.  Chemical Information Service,
       Stanford Research Institute,  Menlo Park, CA.
 4.     Organic Chemical Producers' Data Base Program, Volume II.  Radian Corp.,
       Austin, TX.  Prepared for U.S. Environmental Protection Agency, Athens, GA,
       under Contract No.  68-02-1319.  August 1976. (Unpublished report).
 5.     1977-1978 Catalog.  Aldrich Chemical  Co., Milwaukee, Wl.
 6.     Catalog 18.  Analabs, Inc., North Haven,  CT. June  1976.
 7.     Catalog 750.  J.T. Baker Chemical  Co., Phillipsberg, NJ.  July 1975.
 8.     Bulletin CS-100-8.  Chem-Service,  West Chester, PA.  1975.
 9.     1975 Catalog.  Chemical Procurement Laboratories, College Point,  NY.
10.     Catalog 48.  Eastman Kodak Co., Rochester,  NY.  1976.
11.     Catalog No. 10.  K & K Fine Chemical,  Plainsville,  NY.  1975.
12.     Nanogen Index.  Nanogens International, Freedom, CA.  1975.
13.     1976 Catalog.  Pfaltz and Bauer Chemical Co., Stamford, CT.
14.     Chemical Standards for Air-Water-Foods. RFR Corp.,  Hope,  Rl.  1975.

 DISCUSSION

 J.  D. Hallett, Shell Oil Co.:  I have two questions.  I have heard the words "screening
 analysis" used several times both by you and Mr.  Hall.  Can you define this term and
 would you describe your efforts on the "lab-to-lab" comparability program?

 W. M. Shackelford: Screening  analysis as intended in the analysis protocol is the quali-
 tative and semi-quantitative analysis of effluents for the 129 priority pollutants.  This
 differs from a survey analysis in  that a survey would be concerned with all components
 in a sample.

 J.  D. Hallett;  Define semi-quantitative.

 W. M. Shackelford: Semi-quantitative as we mean it here is quantification by a method
 that has  not been standardized—in this case mass spectrometry.

 J.  D. Hallett: Are you talking  about orders of magnitude?

 W. M. Shackelford: As indicated in the protocol,  concentration values are to be
 reported  in ranges—10-100 parts per billion (ppb) and greater than 100 ppb.  To answer
 your second question, the program to establish some measure of lab to lab comparability
 includes supplying contractors with  the same lots of GC column packings,  analytical
 standards, and requiring that all data be saved on 9-track magnetic tape in a specified
 format.

 J.  D. Hallett; Will the analytical  standards be available to industrial labs?

 W. M. Shackelford: As far as I  know, they will be available commercially.

                                        107

-------
Don Rosebrook,  Radian Corp.;  Radian CorporaHon is the EPA contractor for preparation
of the Consent Decree standards.  When this order is filled, the standards will be made
available to the general public.

Leo Duffy,  Standard Oil Co. of Indiana;  Many analytical chemists regard the GC-MS
to be a totally qualitative instrument.  Would you remark on how one establishes a limit
of detection with a qualitative GC-MS instrument? Doesn't this limit of detection
float and depend on instrument response?

W. M.  Shackelford;  The term  limit of detection as used in the analysis protocol is mis-
leading.  Actually, the numbers referred to as limits of detection are required concen-
trations that must be detected.  The 40  ng figure that is given was derived from the
amount of material injected after the standard extraction and concentration  procedure
was performed on a 10 ppb solution. For some compounds the  figure is 200 ng. So these
are practical  working levels—not  actually limits of detection.  I agree that the mass
spectrometer is not quantitative by present methods.  However, with the use of our
internal standard, d]Q anthracene, and calculation of response factors for each compound,
mass spectrometry appears adequate to establish  the ranges of concentrations we have
proscribed.

Leo Duffy:  In the semi-volatile compounds  you  list three key  fragment ions  along with
the GC retention time of the compound. Would you remark on the certainty of identifi-
cation of specific compounds, restricting yourself only to those three key fragment ions?

W. M. Shackelford:  If the GC retention time is considered, the confidence in identifi-
cation by three  fragment ions is increased.  A window of + 1 minute,  which corresponds
to  about  10 MS  scans on each side of the peak,  has proved to  be adequate.  Of course,
identification with absolute certainty can only occur  if a standard and unknown are run
and the complete spectra compared. For the purposes of the protocol, however, the
present method is sufficient.

F.  L. Robertaccio, E. I. DuPont de Nemours Inc.; How much of your experience has
been on applying these methods to complex  wastewaters?

W. M. Shackelford:  The analysis protocol essentially involved the use of techniques
used at the Athens laboratory for the past three to five years.  The Analytical  Chemistry
Branch  has been concerned with characterization of effluents and essentially all of our
experience has been using methods similar to those in the protocol. Although some work
has been  done on drinking water, much has  been done with actual effluents. The
analysis protocol has also been run on effluents in our lab.

N. F. Seppi, Marathon Oil Co.;  You  mentioned chlorination of waters. Would you
elaborate on whether or not this affects the  concentration of chlorinated compounds
found in water?

W. M.  Shackelford:  The only  elaboration I can make is to quote papers and reports that

                                        108

-------
indicate that after chlorination of water containing organic materials occurs, chlorinated
organics are found.  Whether or not increases in the concentration of chlorinated com-
pounds present before chlorination occurred, I  do  not know. Studies have shown that
chlorinated compounds not found before chlorination do show up after chlorination.

BIOGRAPHIES

                       Walter M.  Shackelford

       Walter M. Shackelford is a Research Chemist
in the Analytical  Chemistry Branch of EPA's Athens
Environmental Research Laboratory.  He has a  B.S.
degree in Chemistry from the  University of Mississippi
and a Ph.D. degree in Analytical Chemistry from
Georgia Tech.  Prior to his work with EPA, Dr.
Shackelford was a post-doctoral research associate
with G.G. Guilbault at the University of New
Orleans and spent two years in research for the U.S.
Army at Edgewood Arsenal, Md.
                       Lawrence H. Keith

       Lawrence H. Keith received his Ph.D. from
 the University of Georgia in 1966.  He worked at
 the Athens Environmental  Research Laboratory in
 Athens, Georgia for over  10 years under the U. S.
 EPA and its predecessors.  Dr. Keith is the editor
 of the book, Identification and Analysis of Organic
 Pollutants in Water, and author or co-author of 39
 technical  publications including 10 chapters in five
 books. At Radian Corporation he is Head  of the
 Organic Chemistry Department and is responsible
 for the direction of a technical staff of 24.  He is
 the Secretary of the ACS Division of Environmental
 Chemistry, Chairman of the Division Steering
 Committee for the North American Council on
 Organic Pollutants (NACOP), and Vice-Chairman
 of the next Gordon  Research Conference on Environ-
 mental Sciences: Water.
                                       109

-------
   TABLE 1  TOTAL FREQUENCY OF OCCURRENCE FOR 13 CHEMICAL GROUPS

Consent Decree Group                                  Number of Observations*

Chlorinated Benzenes                                          247
Chlorinated Ethanes                                           138
Chloroalkyl Ethers                                              67
Chlorinated Naphthalenes                                        0
Chlorinated Phenols                                             92
Dinitrotoluenes                                                 14
Haloethers (other than above)                                     15
Halomethanes                                                360
Nitrophenols                                                   13
Nitrosamines                                                   0
Phthalate Esters                                               183
Poly nuclear Aromatic Hydrocarbons                                49
DDT Metabolites                                                23

*5700 Total Observations for Study
                TABLE 2 FREQUENCY OF PRIOR OCCURRENCE
               OF COMPOUNDS WITHIN 13 CHEMICAL GROUPS

                                      Number of        Number of Compounds
Consent Decree Group                  Compounds        Meeting 5% Criterion

Chlorinated Benzenes                       28                    9
Chlorinated Ethanes                         19                    9
Chloroalkyl Ethers                          15                    4
Chlorinated Naphthalenes                     0
Chlorinated Phenols                         16                    5
Dinitrotoluenes                              4                    4
Haloethers (other than above)                 4                    4
Halomethanes                              17                    9
Nitrophenols                                4                    4
Nitrosamines                                0
Phthalate Esters                            19                    7
Polynuclear Aromatic Hydrocarbons            12                    9
DDT Metabolites                            2                    2
                                     110

-------
        TABLE 3 MANUFACTURING DATA FOR 13 CHEMICAL GROUPS

                                                 Number of Compounds
Consent Decree Group                              With Production Data

Chlorinated Benzenes           -                            4
Chlorinated Ethanes                                         5
Chloroalkyl Ethers                                          0
Chlorinated Naphthalenes                                   2
Chlorinated Phenols                                         3
Dinitrotoluenes                                             2
Haloethers (other than above)                                0
Halomethanes                                              4
Nitrophenols                                               4
Nitrosamines                                               1
Phthalate Esters                                            2
Pol/nuclear Aromatic Hydrocarbons                           0
DDT Metabolites                                           0
               TABLE 4 NUMBER OF COMPOUNDS ELIMINATED
                  DUE TO LACK OF ANALYTICAL STANDARDS

                                      Number of
                                      Considered           Number of
Consent Decree Group                  Compounds      Compounds Eliminated

Chlorinated Benzenes                         9                  2
Chlorinated Ethanes                           9                  2
Chloroalkyl Ethers                            8                  2
Chlorinated Naphthalenes                      2                  0
Chlorinated Phenols                           6                  0
Dinitrotoluenes                               4                  1
Halothers (other than above)                   4                  1
Halomethanes                               10                  0
Nitrophenols                                 6                  0
Nitrosamines                                 3                  0
Phthalate Esters                               7                  2
Polynuclear Aromatic Hydrocarbons             9                  2
DDT Metabolites                              2                  0
                                     111

-------
                  "INCORPORATION OF THE PRIORITY POLLUTANTS
                INTO EFFLUENT GUIDELINES DIVISION DOCUMENTS"

                             Robert W. Dellinger
               Chemical Engineer, Effluent Guidelines Division
                    U. S. Environmental Protection Agency

The purpose of this paper is to explain the role of the  Effluent  Guidelines
Division relative to the 1976 Settlement Agreement (1).  It will also provide
some  incite  into  the  overall relationship between the Effluent Guidelines
Division and ether responsible offices  regarding  future  regulations  which
will be affected by the Settlement Agreement.

BACKGROUND

The  Effluent  Guidelines Division (EGD) is a division of the Office of Water
Planning and Standards within the Office of Water  and  Hazardous  Materials.
The  EGD  was  founded  in  August  of  1972  with  the  primary  function of
contributing to the establishment of  effluent limitations and guidelines  and
standards  of  performance  for new sources pursuant to Sections 301, 304, and
306 of  the Federal Water Pollution Control Act as amended in 1972.  This  Act
is  commonly  referred  to  as Public  Law 92-500.  The division was originally
headed  by Allen Cywin and is now headed by Robert Schaffer.   Over  the  past
several years,   the  EGD   has  been  involved in the publication of some 1700
regulations covering over 250 subcategories within 43 industrial point source
categories.

As of June 7, 1976, the function of the EGD has taken a new slant.   On  June
7, 1976, the Settlement Agreement with the Natural Resources Defense Council,
et  al., came into  effect and settled several cases in the District Court for
the District of Columbia  (1).  This Settlement Agreement  requires  that  EPA
give  consideration  to  65 specific chemicals or classes of chemicals when
establishing effluent limitations reflecting the  best  available  technology
economically  achievable,   also referred to as BAT or 1983 limitations.  This
requirement for consideration of the  list of chemicals carries  over   to  the
establishment of  New  Source Performance Standards and to the establishment of
pretreatment  standards  for  new and for existing sources.  The authority by
which EPA has been  directed to perform this task derives from  Sections  301,
304   (b), 306, 307  (b), and 307  (c) of Public Law 92-500  (2).  The Settlement
Agreement also establishes  a schedule within which contracts are to  be  let,
regulations  proposed,  and regulations promulgated for a prioritized  list of
21 industrial categories.

PRELIMINARY EFFORTS

Several activities  had to be  completed  before  the  role  of   the  Effluent
Guidelines   Division   relative   to  the  Settlement  Agreement   could  be
definitized.  These activities included further work  with  the  list   of  65
chemicals  or  classes  of  chemicals,  development   of sampling and analysis
procedures, and the letting of contracts in order that work could begin on   a
timely  schedule.
                                      112

-------
In  order  to  enable  the  Agency  to  conduct  a proper  scientific study of
industrial discharges, the various pollutant parameters were  further.defined.
Without this information as a point of reference, the  analytical  portion  of
the  BAT  review  would  be unmanageable given the limited time schedules and
resources of the Agency.  This list of 65  chemicals  or classes  of  chemicals
was  defined  in  November  of  1976  to   include  123 specific  unambiguous
compounds.  This list of 123 specific compounds  appeared in  Portfolio  B  of
the  Petroleum  Refining  Industry  Survey distributed  to  the  industry in
February of 1977.  Since that time, the list has been  revised.  This occurred
in April of 1977 with the addition of 6 compounds; now the priority pollutant
list totals 129 compounds   (3).   The  six additional compounds  include  5
additional  PCB's  and  di-n-octyl  phthalate.   We  believe  that  this list
fulfills the  requirements  of  the  court approved  agreement  and  can  be
evaluated analytically.

In  addition, the Agency has established procedures  for sampling and analysis
of industrial effluents for the priority pollutants.  These  procedures  are
defined  in the "Sampling and Analysis Procedures for  Screening of Industrial
Effluents for Priority Pollutants" published in  March  of 1977 and revised  in
April of 1977 (4).

While  these  tasks  were being completed  by personnel within various program
areas of EPA, the Effluent Guidelines  Division  set  about  to  fulfill  its
requirements relative to the Settlement Agreement.   Due to a  relatively small
staff, the EGD has historically depended on outside  contractors to aid  in the
fulfillment of its tasks.  EGD technical personnel serve as project  officers.
Their  function   is  to monitor, assist, direct,  and  provide various  inputs to
the  contractor.   The project officer  works directly  with  the  contracting
firm,  usually  on   a  daily basis.  One of his  main duties is to provide the
contractor with information regarding EPA  policy.

The  Effluent Guidelines Division   should   meet   the  schedule for   executing
contracts  that was  established in the Settlement Agreement.  The EGD let ten
contracts prior to January  1, 1977  (These  included the timber products, steam
and  electric, leather tanning, iron and steel, petroleum,  paint and  ink, coal
mining,  ore mining,  nonferrous metals, and paving and   roofing  industries.).
In  addition,  contractors  have been selected  for ten  of  the  remaining  eleven
industrial categories and negotiations are currently  going  on  between  the
selected contracting firms and EPA procurement. The  miscellaneous  chemicals
request  for proposal will be readvertized  and will   involve  multiple   awards
rather   than  a   single award.  Six of the eleven contracts are scheduled for
award by June 30, 1977.  The additional five will be awarded  by  October  31,
1977.    The Scope of Work to be fulfilled  by each of the  contracting firms is
defined  in "Request  for Proposal  No.  WA  77-B074" dated February 7,  1977  (5).
The  contractor currently involved in the petroleum refining industry study is
Burns and Roe of  Paramus,  N.J,    EPA  personnel  from  the  Robert  S. Kerr
Environmental  Research Laboratory (RSKERL) are  participating in the sampling
and  analysis  phase of  the  study.   Another   contracting   firm,  Ryckman,
Edgerley,  Tomlinson &  Assoc.,   Inc.  (RETA)  of St. Louis, Missouri, is also
assisting in sampling and analysis.
                                       113

-------
DUTIES OF THE EFFLUENT GUIDELINES DIVISION

The Effluent Guidelines Division with the help of  contractors  will  perform
the  function  of  gathering  and  analyzing  data.   Its  tasks  include  the
determination of presence or absence of the priority pollutants within  waste
water  generated  by  industrial point source categories or subcategories, an
evaluation of the quantities of priority pollutants present  in   these  waste
waters,  a  determination  of  the  extent to which these priority pollutants
exist subsequent  to  treatment  by  various  control  technologies,  and  an
estimate  of  the  cost  of  implementing these various control and treatment
technologies.  This information  will  be  used  during  the  decision-making
processes established within the Agency in fulfilling the requirements of  the
Settlement Agreement.

The first step involved in completion of these tasks involves the development
of  a  profile  of  the  industrial  point source category.  This involves an
identification of the plants which comprise  the  point  source   category  in
terms  of  size,  location,  and number.  It also includes a determination of
such information as the numbers of direct  versus  indirect  dischargers,  an
identification  of the treatment systems employed at each plant,  both end-of-
pipe and in-plant,  and  an  identification  of  products  produced  or  unit
operations  employed  at  plants in the industry.  This information forms  the
foundation on which other information is based and will be  usually  obtained
by  submission  of  a  questionnaire to all or a statistically representative
portion of the industry.  Portfolio A  of  the  Petroleum  Refining  Industry
Survey  provides  the  Agency  the  means  to  profile the petroleum refining
industry.

An important part of the study is the determination of presence or absence of
the priority pollutants.  This will be accomplished  in  part  by literature
searches,  but  the major thrust of_this exercise involves the implementation
of a  sampling  program.   Presence  or  absence  will  be  determined  by  a
"screening  study"   (4).   This screening study involves the gathering of raw
water, raw waste water, and final treated effluent samples for  analyses   for
the   priority   pollutants.    Analyses   involve   gas  chromatography/mass
spectrometry (GC/MS) techniques for volatile and  semi-volatile   organics  as
well  as  analysis for metals and pesticides (4).  Upon obtaining the results
of this screening survey, decisions will be made regarding the  necessity  to
further investigate certain of the priority pollutants.  These decisions will
be  based on the guidance provided in paragraph 8 of the Settlement Agreement
(1).  Exclusion of parameters can be justified if the specific  pollutant  is
present  soley  as  a  result  of  its  presence  in intake waters, or if  the
specific pollutant is either not present in the discharge or  is  present  in
insignificant quantities and not likely to cause toxic effects.

Upon making the decision of which parameters to be concerned with relative to
a  point  source  category  or  subcategory, the second phase of  the sampling
program is implemented.  This involves verification or quantification of   the
amount  of  priority  pollutants  present.   It  also,  ideally,  involves a
determination of the  unit  operation  or  operations  which  result  in   the
discharge   of   the   specific  priority  pollutants  of  concern.   In   the
                                      114

-------
quantification phase the degree to which the priority pollutant is removed by
various existing treatment systems within  the point  source category  will  be
evaluated.

As  more  of  these  BAT  studies get underway  and data become available, the
potential exists for supplementing information  relative to one  point  source
category  with data from another point  source category.  This could involve a
transfer of technology from a point  source category  where   a  more  advanced
technology is practiced to one in which a  lesser  technology  is practiced.  It
is  very  early  in  the  program  to   predict  to what degree, if any, direct
transfer of technology could or would be applicable  or whether or not it  is,
in fact, necessary or desirable.

The EGD will also develop cost data  to  correspond to the various technologies
thought  to be potential BAT technologies. This  may include in-plant control
measures, recycling  techniques  to  reduce wastewaster  discharges,  source
control  prior to biological treatment, or end-of-pipe technologies.  Efforts
will be made to provide good capital and operating   cost  estimates  for  the
identified potential BAT technologies.

EGD Efforts t£ Date Relative ^o the  Petroleum Refining Industry

Regarding the petroleum refining industry  specifically, this overall EGD work
plan  has  been  altered  slightly.   Due   to   time  constraints and delays, a
decision was made to combine the screening and  quantification phases  into  a
single  program.  Initially, twelve  plants were to be sampled.  At six of the
plants, pilot activated carbon studies  will  be  conducted.  This  sampling
study  is  now  being  conducted  by RSKERL.   A rationale  was developed for
selection of twelve individual refineries  to be sampled  (6).  This  involved
establishing the following criteria  for selection:

1.   The selected plants must be BPT plants; i.e., they must be attaining BPT
limitations or fulfilling the requirements of   their July   1,  1977,  permit
requirements.

2.   The  selected  plants  would  have a single outfall.   This criteria was
established strictly due to economic considerations, namely  the large cost of
analysis of individual samples.

3.  The selected plants would cover  as  many of  the unit operations  found  in
the petroleum refining industry as possible, and

4.   The  selected  plants would represent various crude  sources.  Because no
completed questionnaires had been received at  the time of selection,  it  was
agreed  that  this criteria would be met through  a selection of plants with a
wide geographic distribution.

A secondary consideration was that in the  selection  of   the  six  refineries
where  pilot  carbon  studies would  be  performed, at least  four would involve
activated sludge treatment.  The other  two would  also involve  some other form
of biological treatment (i.e., aerated  lagOOM, oxidation ponds, etc.).
                                      115

-------
The EPA contractor, Burns and Roe, went to work on developing the list of  12
refineries  to  be  sampled.   They  visited  EPA  regional  offices and State
offices.  They gathered Discharge Monitoring Report data  while  speaking  to
Regional  and  State  personnel.  Forty-five refineries were identified which
met the first two criteria  (attainment of  BPT  limitations   while  having  a
single  outfall)  (6).  The Oil and Gas Journal was searched to determine the
unit operations employed at the various refineries.  The  fourth  criteria  of
geographic mix was added to the selection process.

A  matrix  was  developed  which  resulted in the compilation of a list of  12
refineries to be sampled with 2 alternates  corresponding to  each  selected
refinery.  The current petroleum refining industry sampling  study is based  on
this list of 36 plants.

Various  contacts  were  made  with  the American Petroleum  Institute and the
National Petroleum Refiners' Assocation to  inform  them  of  the  refineries
selected  for  sampling.   API representatives contacted  individual plants  to
determine the mode of operation for the duration of the sampling study, March
to July.  Based on this additional information, several first choices dropped
out due to various reasons  (such  as  scheduled  turnarounds  and  strikes).
Representatives  of alternate refineries were contacted.  A  sampling schedule
was then developed around the availability of the selected   refineries.  The
decision  was  made that due to the time constraints imposed by the Court and
due to the tremendous workload on RSKERL personnel,  this sampling  schedule
would be strictly adhered to.

With  the unavailability of certain refineries and a shift to alternates, the
final list of 12 refineries to be sampled omitted  several   unit  operations.
Additional  funding   became  available  relative  to   the petroleum refining
industry study.  The  decision was made to  supplement  the   initial  sampling
program  by  sampling additional  plants.  Ryckman, Edgerley, Tomlinson, and
Associates, Inc. is conducting the supplemental sampling  program  which,  at
the  present time, includes 5 refineries (7).  Table I gives a listing of, the
petroleum refineries  to be  sampled in the BAT review and  the  scheduled  date
of initiation of the  sampling program at each individual  refinery.

THE RELATIONSHIP OF RESPONSIBLE PROGRAM AREAS

The  total  of  the   information gathered by the Effluent Guidelines Division
forms a part of the technical information on which final  effluent limitations
will be established.  Other studies are being conducted   by   other  divisions
within the Office of Water-Planning and Standards.  Major inputs will be made
by  the  Criteria and Standards Division headed by Dr. Kenneth Mackenthun,  by
the Monitoring and Data Support Division headed by Dr. Edmund M. Notzen, and
by the Office of Analysis and Evaluation headed by Mr. Swep  Davis.

The Criteria and Standards Division  (CSD) is responsible  for studies relating
to  health  and  environmental  effects, will provide  physical and background
data relating to the  priority  pollutants,  and  will  interface  with  other
government  agencies  to obtain health and environmental  effects information.
                                      116

-------
They will also  develop specific criteria that will provide justification  for
regulation  of the priority pollutants.

The Monitoring  and  Data Support Division  (MDSD) is responsible for general
studies  of  each priority pollutant including  geographical  and  quantitative
profiles of users and manufacturers, and for a summary of losses of priority
pollutants  to the environment  through  discharges   to  water  by  industrial
producers   and   users,  municipalities,  agricultural  sources,  and  natural
sources. They  will develop a  data  base  associated  with  each  industrial
category and gather information relating to ambient water quality to provide
an overview of  the magnitude and geographical  extent  of  potential  problem
areas.   This   information will be used to provide an overall  risk assessment
based on the quantity of priority pollutants released and their  environmental
effects.

The Office  of Analysis and Evaluation   (OA&E)  is  responsible  for  economic
impact  studies  which  will  include  an  increased emphasis toward benefit
analysis.   Economic studies will include a clear description   and  evaluation
of  load  reductions,  of  health  and  water  quality risk avoidance,  and of
correlations between industry location, water quality, and  potential   health
problems.

The inputs of all responsible program areas  merge at the working group  level.
Working  groups  involve personnel representing the  responsible  program areas
but include personnel representing such groups as Research  and  Development,
Toxic   Substances,   Pesticide  Programs,   Air  Programs,  General  Counsel,
Enforcement, Planning and Evaluation, and EPA  Regional  Offices.   Decision-
making  will  begin at the Working Group level with  recommendations regarding
regulations passed along to EPA management for evaluation.

At the present   time,  working  group  activity  relative  to  the  petroleum
refining  study  has been limited to plant selection for the sampling program
and other inputs relating to the  technical   (EGD)   and  economic   (OA  &  E)
studies.   As  more  information  is collected, the  Working Group will  take  a
more active role in the decision-making process.

MAJOR MILESTONES

The following is a schedule of  the  major   milestones  which  are  important
relative to the  petroleum refining BAT  study:


EPA Draft Development Document Available for Public  Comment -  March 7,  1978.

Public Meeting to Discuss Draft Development  Document - April  7,  1978.

Proposal of Regulations in the Federal  Register -  July  15,  1978.

Public Meeting to Discuss Proposed Regulations -  September  22, 1978.

Promulgation of  Regulations in the Federal Register  - January, 22,  1979.
                                      117

-------
SUMMARY

The Agency has a considerable amount of work  to  do  relative to the Settlement
Agreement.  The EGD has a major area of responsibility,  that of gathering the
technical  information on which standards will be established.  The petroleum
refining study is in the early stages of development;  we  are  awaiting  the
results of GC/MS analysis for non-volatile  organics and  the results of metals
and pesticides analyses.  Upon receipt of this data, the Agency will make the
decisions,  at  the  working  group level,  which will determine the immediate
direction which the petroleum refining study  will take.

REFERENCES

(1) Natural Resources Defense Council,  et  al.  v..  Train,  8  E.R.C.   2120
    (D.D.C. 1976).

(2) Public Law 92-500, 92nd Congress, S. 2770, October 18,  1972.

(3) "Rationale for the Development of  BAT  Priority  Pollutant  Parameters,"
    Effluent  Guidelines  Division,  Office of Water and Hazardous Materials,
    U.S. Environmental Protection Agency, Washington,  D.C., June, 1977.

(4) "Sampling and Analysis Procedures for Screening of  Industrial  Effluents
    for   Priority   Pollutants,"   U.S.    Environmental Protection  Agency,
    Environmental Monitoring and Support Laboratory, Cincinnati,  Ohio, April,
    1977.

(5) "Request for Proposal No.  WA  77-B074,"  U.S.   Environmental  Protection
    Agency, Washington, D.C., February 7, 1977.

(6) "Selection of Refineries for RSKERL  Sampling   Program,"  Burns  and Roe
    Industrial Services Corporation, Paramus, NJ, February  22, 1977.

(7) "Selection of Refineries for B&R Supplemental   Sampling  Program,"  Burns
    and Roe Industrial Services Corporation,  Paramus,  NJ, May 4,  1977.

DISCUSSION

Paul  Mikolaj.  Lion  Oil  Co.;   When  will  the results of EPA's preliminary
screening studies be available?

R^ jj^ Dellinger; ' As of next week, 14  of   the   17   plants   which  have  been
icheduled  for  sampling  will  have  been  visited.  Results of the first 6
refineries screen-sampled by RSKERL are anticipated to  be   received  by EGD
around  mid-June.   It  was  originally  anticipated  that  this data would be
received by the end of May.  Data relating  to the other  refineries sampled by
RSKERL and RETA will be available at a much later date.   It is too  early to
predict  the  actual date of availability at  this time;  we  are pretty much at
the mercy of the analytical labs with regard  to  turnaround  time.

Paul  Mikolaj;  Will this information be publicly available?



                                     118

-------
R! Ht Dellinger;   Eventually this information will be publicly available  and
published  in  an EPA development document.  Between now and that time  I would
anticipate   that   there  will  be  an  exchange  of  data  between *EPA  and
representatives of the individual refineries where split samples were  taken
In  any event, the  EPA data will be submitted to the refineries as soon as
data become available.
BIOGRAPHY

Robert W.  Dellinger is a chemical
engineer with the Effluent Guidelines
Division of EPA.   He has B.S. and M.S.
degrees in Chemical Engineering from
Virginia Polytechnic Institute and the
University of Maryland, respectively.
He has been with the Effluent Guide-
lines Division since November of 1972.
                                 TABLE I

     PETROLEUM REFINERIES SCHEDULED FOR SAMPLING AS OF JUNE 3, 1977
                                                              Group
                                                           Conducting
                                                  Carbon?   Sampling  (1)
Scheduled
Refinery
Gulf
Exxon
Clark
Hunt
Texaco
Mobil
Getty
Phillips
Shell
Conoco
Asamera
Exxon
Exxon
Quaker State
Sun
Coastal States
Arco
Location
Philadelphia, PA
Bay town, TX
Hartford, IL
Tuscaloosa, AL
Lockport, IL
Augusta, KS
El Dorado, KA
Sweeny, TX
Anacortes, WA
Ponca City, OK
Commerce City, CO
Billings, MT
Benicia, CA
Newell, WV
Toledo, OH
Corpus Christi, TX
Philadelphia, PA
Date
3/21
3/21
3/28
3/28
4/4
4/4
4/18
5/2
5/2
5/16
5/23
5/30
5/30
6/6
6/13
6/27
6/27
                                                    No
                                                    No
                                                    No
                                                    No
                                                    No
                                                    No
                                                    No
                                                    Yes
                                                    Yes
                                                    No
                                                    No
                                                    Yes
                                                    Yes
                                                    No
                                                    No
                                                    Yes
                                                    Yes
RSKERL
.RSKERL
RSKERL
RSKERL
RSKERL
RSKERL
RETA
RSKERL
RSKERL
RETA
RETA
RSKERL
RSKERL
RETA
RETA
RSKERL
RSKERL
     (1)   All sampling teams are assisted by Burns & Roe of Paramus,
       NJ
                                      119

-------
                GENERATING DATA ON THE PRIORITY POLLUTANTS
                   FOR THE EFFLUENT GUIDELINES DIVISION

                               Leon H. Myers
                    Chief, Industrial Sources Section
                         Source Management Branch
              Robert S. Kerr Environmental Research Laboratory
                               Ada, Oklahoma

     During the past decade, the Citizens of the United States have become
vitally interested and involved with the environmental conditions  of  the
Nation's water supplies.  Our Congressional leaders recognized this concern,
and in October 1972, enacted legislation "to restore and maintain  the chemical,
physical, and biological integrity of the Nation's waters."

     Within the framework of the Federal Water Pollution Control Act  Amend-
ments of 1972, the newly created Environmental Protection Agency was  directed,
as National Policy, to prohibit the discharge of toxic pollutants  in  toxic
amounts  (Section 101.(a) (3).  Also, in Section 101 of PL 92-500,  sub-section
(a) (6) declares that "it is the National Policy that a major research and
demonstration effort be made to develop technology necessary to eliminate the
discharge of pollutants into the navigable waters of the contiguous zone and
the ocean."

     Section 307 (a) of the act requires EPA to publish a list of  toxic pol-
lutants and to promulgate effluent standards for such toxic pollutants.  The
following characteristics shall be taken into account; "the toxicity  of the
pollutant, its persistance, degradability, the usual or potential  presence of
the affected organisms, and the nature and extent of the effect of the toxic
pollutant on such organisms."

     In September 1973, EPA promulgated a list of nine toxic pollutants and in
December 1973, proposed standards for these nine toxic pollutants.

     On June 7, 1976, a settlement agreement was signed between Russell Train,
EPA Administrator, and four concerned citizen organizations to investigate
the presence of 65 parent chemical constituents reported to be present in the
Nation's water supplies.  The 65 chemicals are listed in Table 1.  This set-
tlement agreement also listed 21 point source categories by Standard  Indus-
trial Classification (SIC)including 2911 Petroleum Refining.

     After the agreement was signed, the EPA's Effluent Guidelines Division
was delegated the responsibility to satisfy the legal conditions of the
document.  After preliminary meetings regarding the sampling, analytical, and
data storage protocols related to the study, the Robert S. Kerr Environmental
Research Laboratory  (RSKERL) in Ada, Oklahoma, was contacted to provide tech-
nical assistance in sampling and analysis for the petroleum refining  industry.

     A meeting was held in Ada in November with representatives of EPA's
Effluent Guidelines Division, the Office of Research and Development, an


                                      120

-------
American Petroleum Institute Task Force, and key  RSKERL  personnel.  The meet-
ing provided  an insight regarding potential sampling  and analytical problems
associated with the general protocols presented by  Effluent  Guidelines Divis-
ion's  Quality Control Branch.

    From the results of this meeting, a preliminary  plan was proposed and
accepted.  Phase I was designed as a screening study  to  determine  the presence
or absence of the priority pollutants, and Phase  II was  a screening carbon
study.  Sampling and analytical procedures presented  two major problem areas
which  needed  immediate attention; these problem areas were:

     Sampling;  The original protocol dated November  1976, required automatic
samplers be used to collect composite samples.  Most  automatic samplers use
petroleum-derived tubing which has the ability to adsorb or  desorb organics
from the tubing walls into the water sample.  Another problem in using auto-
matic samplers is the lack of electrical safety devices  required when sampl-
ing in an  explosive atmosphere area such as the API separator.  A  variance
was requested to allow the sampling be conducted by  EPA personnel;  this was
granted due  to the aforementioned problems.

Analytical

     The list of 65 chemical compounds, presented problems to the  analysts be-
cause of the  ambiguity and colossal analytical effort in separating, identify-
ing, and quantifying most of the compounds, metabolites,  and  isomers which
appeared on the list.  A list of "Unambiguous Compounds" was prepared to
specifically  name each pollutant which is to be identified.

     This  list was prepared by scientists from EPA's  Athens, Georgia, labor-
atory and  personnel of the Effluent Guidelines Division. The list of un-
ambiguous  compounds was derived from a compendium of  information gathered by
these scientists from an earlier study on organic compounds  in potable water
supplies.   The list of compounds shown in Table 2 is  identical to  the specific
compounds  listed in the settlement agreement.  The  classes of compounds are
represented  by carefully selected individual material.   The  EPA list was dis-
cussed with the plaintiffs of the settlement agreement,  and  there  was no
dissension.

     Another  analytical problem encountered was the protocol prepared in
November by EPA scientists from the Cincinnati and  Athens Laboratories.  The
November protocol was a preliminary analytical exercise  and  was distributed
for review comments by industrial, society, and EPA scientists.  In November,
a meeting  was held at Atlanta, Georgia, with EPA  participants from Effluent
Guidelines, Athens, and Ada.  A request was made  to prepare  a specific pro-
tocol for  non-volatile organics and volatile organics analytical procedure.
Athens scientists prepared the protocol for non-volatile organics, and
Cincinnati scientists prepared the protocol for the volatile organic compounds.
The procedures were merged into one protocol and  presented in March 1977.
"These were the major problems encountered in making preparations for this
study, and they were corrected prior to the study date.
                                      121

-------
     In November, RSKERL requested a list of 12 refineries  be  selected
Effluent Guidelines for the field study.  The selected  refineries  would be
representative of the five refinery categories, various crude  oil  sources,
geographic locations, and meet Best Practicable Control Technology Currently
Available (BPC1C), or EPA regional permit conditions.   An alternate list of
six refineries meeting the aforementioned conditions was to be furnished in
the event that one or more of the primary selections become unavailable.

     Effluent Guidelines would then contact the American Petroleum Institute
who would contact the refinery to determine if there were any  planned  turn-
arounds or other problems which could interfere with the purpose of the study.
After the selected refineries had been contacted by API and the results re-
layed to Effluent Guidelines, a final selection of refineries  was  prepared.
After confirming the selected refineries, the Effluent  Guidelines'  Project
Officer notified each refinery they would be visited by the RSKERL team for
the priority pollutant study.  The 12 refineries to be  visited by  RSKERL
personnel for sampling are shown in Table 3.

     This study was divided into two phases:  Phase I is considered a  screen-
ing study to determine the presence or absence of priority  pollutants,  and
Phase II is a screening study on the effectiveness of carbon treatment  to
remove the priority pollutants.  These two phases will  be discussed separately.

Phase I

     Six refineries were selected for this phase, and those refineries were:
1.  Gulf, Philadelphia          4.  Clark, Hartford
2.  Exxon, Baytown              5.  Texaco, Lockport
3.  Hunt, Tuscaloosa            6.  Mobil, Augusta

     At each refinery, a minimum of three points were selected for sampling.
The intake water to the refinery was selected to provide information on the
background quality as related to the priority pollutants.   The second sampl-
ing point represented the API separator or dissolved air flotation effluent,
which would indicate the quality of water that would be expected without
biological treatment.  The third sampling point selected was representative
of the NPDES sampling point; this sampling point represented biologically
treated wastewater.

     Sampling times were established over a three-day study period,  and the
composite samples were prepared from aliquots obtained  at three-hour inter-
vals.  On the first sampling day, the composite sampling began at  hour 1300
and was completed at hour 1000, the following morning.  Twenty-four hour
composite sampling was started at 1200 on the second day and 1100  on the
third day and completed at 0900 and 0800, respectively.  Grab  samples were
obtained at the final sampling time for each day.  Table 4  indicates the
parameter, type of samples, and preservative used for the samples.   In
addition to the priority pollutant samples, Effluent Guidelines contracted
with Ryckman,  Edgerly, Thomlinson, and Associates Laboratory to furnish con-
tainers and analysis for "classical parameters" for each sampling  point.
Table 6 is a list of the "classical parameters," type of sample, container,
                                     122

-------
and  preservative furnished by the contract  laboratory.   Equipment used to
sample  the water was glass, stainless steel,  or aluminum to  prevent sample
con t ainina t ion.

     Figure  1 is a photograph of the screen at  the Gulf-Philadelphia Refinery,
where their  intake water is obtained from the Schuylkill River.  Figure 2 is
representative of the intake water, this  particular sample point is the dis-
charge side  of the pump at the Gulf-Philadelphia Refinery.   Figure 3 is a
photograph of the corrugated plate interceptor  used at Hunt  Oil in Tuscaloosa,
Alabama;  this sampling point is also shown  in Figure 4 which is a bypass line
to the pH meter.  Figure 5 is a photograph  of the DAF unit the Hunt Refinery
uses for  a final clarifier; and Figure  6  is the sampling point which is prior
to the holding pond.

     The  six refineries selected for "screening only" have been visited and
sampled.  At this point in time, a draft  has  been prepared to describe the
refinery's wastewater treatment system, and analyses have been concluded and
recorded  for the parameters with the exception  of volatile organics and non-
volatile  organics.

     The  American Petroleum Institute contracted with Exxon  Research to
accompany the EPA sampling team and obtain  "replicate samples" at Exxon-
Baytown.   In addition, Gulf-Philadelphia; Texaco-Lockport; Mobil-Augusta;^ and
Exxon - Baytown provided company personnel  to accompany  the  EPA team to
obtain replicate samples.  A request has  been made to each refinery which
obtained  "replicate samples" to furnish their data for inclusion- in the
report.

Activated Carbon Screening Study

     EPA's  Effluent Guidelines Division also  selected six petroleum refineries
for activated carbon pilot scale screening  studies.   The petroleum refineries
selected  are:

     COMPANY                 LOCATION              CLASS    TREATMENT SYSTEM

1.  Shell               Anacortes, Wash.             B     activated sludge
2.  Phillips            Sweeney, Tex.                C     aerated lagoon
3.  Exxon               Benecia, Cal.                B     activated sludge
4.  Exxon               Billings, Mont.              C     aerated lagoon
5.  Coastal  States      Corpus Christi, Tex.         C     activated sludge
6.  ARCO                 Philadelphia, Pa.            B     activated sludge

     The  purpose of this pilot scale screening  study is  to determine if gran-
ular activated carbon possesses the capability  to adsorb any of the priority
pollutants and the treatment effectiveness  encountered when  powdered activated
carbon is used to supplement a biological treatment system which is treating
petroleum refinery wastewaters.

     There are two general types of activated carbon treatment which are
being investigated during this screening  study.   The granular carbon system
consists  of  placing a bed of granular carbon  in an enclosed  container and


                                     123

-------
percolating filtered biotreated water through  the  bed  in the downflow mode.
The second system is a powdered activated carbon system,  where the powdered
carbon is mixed with the mixed liquor of an activated  sludge system.

A.  Granular Activated Carbon System

     During this screening, three virgin activated carbons  from three carbon
manufacturers and one activated carbon, which  had  been previously used at a
refinery's activated carbon system and regenerated for recycle, are being
investigated.

     The refinery's bio-effluent is pumped through a multi-media filter
column to remove suspended solids, insoluble oils,  and pin  point floe and
scum which may be in the final effluent.  The  multi-media thick wall  glass
column is 5 ft. in length and 6 in. in diameter.   On the bottom of the filter
column is a stainless steel screen.  Three inches  of limestone rock are
placed on the screen followed by 6 in. of washed sand  and 18 in.  of anthra-
filt media for the top layer.  Figure 7 is a diagram of the filter system
used during the carbon screening study.  When  the  pressure  in the column
exceeds 15 psig, the column is backwashed with either  potable or carbon
treated water, and an alternate multi-media filter is  used  to supply  filtered
bio-effluent to the carbon columns.

     RSKERL has been supplied with virgin granular  activated carbon from  the
following manufacturers who recommended the specific carbon to be used to
treat petroleum refinery effluent:

1.  Calgon Filtrasorb 300, 8 x 30 mesh
2.  ICI Hydrodarco 3000, 8 x 30 mesh
3.  Wesvaco WVG, 12 x 40 mesh

In  addition to these three carbon sources, the Atlantic Richfield Co.  (ARCO)
supplied a source of Calgon1s Filtrasorb 300 carbon which had been used and
regenerated at ARCO's Watson Refinery's activated  carbon treatment system.
Figure 8 is a schematic of the granular carbon pilot scale  system designed
for this study.

     Each column has a stainless steel sieve screen in the  bottom of  the  col-
umn with 3 in. of limestone serving as a support bed for the specific carbon.
Fifteen pounds of activated carbon was placed  on top of the limestone rock
for a carbon bed depth of 36 to 40 in.  Multi-media filtered effluent is
pressured into the top of each of the columns  and  the  water percolates
through the column into the discharge pipe which contains a valve limiting
the flow to 0.25 g/m for each column.  Figure  9 is a schematic of the granular
carbon pilot system.

     An alternate carbon system employing powdered activated carbon treatment
is  being evaluated at the four refineries where activated sludge treatment is
employed.  In this system, powdered activated  carbon is added directly into
the mixed liquor basin, and.the carbon residual maintained  at 4000 mg/1.
                                     124

-------
    RSKERL designed  and  fabricated a 0.25 g/m complete mix activated sludge
system from carbon steel.   A schematic of the powdered carbon pilot scale
system is shown in Figure 10.   To acclimate the pilot system, mixed liquor
from the refinery's aeration basin is pumped into the pilot plant's aeration
basin, and the refinery's return sludge is added to the clarifier.  Concen-
trations of suspended solids in the mixed liquor tank are maintained to
approximate the full-scale plant's operation.  During the initial period of
acclimation in the pilot  plant, the biological plant influent will be fed
to the pilot scale aeration basin.  Total organic carbon analyses are used to
determine if the  pilot scale effluent quality is essentially the same as the
full-scale system.

    On Day 5  of  the  study,  five pounds (2,250 grams) of the selected powdered
carbon, which was predetermined by an isotherm selection procedure, will be
added to the inlet of the bio-plant in a slurry over a 10-hour time period.

    Effluent  from the final clarifier is collected in a sump; the water from
the sump flows through glass wool to catch any powdered carbon which might be
discharged.  The  glass wool and sump are emptied back into the aeration basin,
thereby maintaining a near constant powdered carbon concentration in the
pilot system.

CARBON SCREENING  SAMPLING

Priority Pollutant Program

     Screening samples are obtained from the intake, API separator, and NPDES
sources on  three  consecutive days; the protocol, preservation, and procedures
are the same as accomplished with the first six refineries.  These same
samples will be obtained on the sixth day, ninth day, twelfth day, and
fifteenth day.  A schedule of sampling periods for the priority pollutants
is shown in Table 6.

     Beginning with Day 6, the carbon screening pilot plant will be sampled
and again on Days 9,  12,  and 15.  The sampling frequency and sampling loca-
tion program is shown in Table 7.

     The collected  samples are air-freighted to Ada in ice chests.  Upon
receipt of  the samples, phenolics and cyanides are analyzed to meet 24-hour
preservation criteria.  The three 24-hour composite non-volatile organics
samples are composited into one sample and extracted in accordance with the
March protocol.   The  extracts are forwarded to a contract laboratory for
GC/MS analysis.   Metals samples are digested and analyzed by the March proto-
col procedures and  forwarded to EPA's Region V Laboratory where they will be
analyzed by plasma  emission spectrometer for comparative purposes.  The
three 24-hour  volatile organics samples are composited and the composite
forwarded to a contract laboratory for analysis.  Split extracts for the non-
volatiles are  sealed  in glass ampules and maintained a 4°C, and will be
analyzed by an EPA laboratory to provide quality control information on the
study.
                                     125

-------
     During the carbon study, total organic carbon analysis  is  obtained every
eight hours on the influent to the mixed-media filter and  the effluents from
the mixed-media filter activated carbon columns 1, 2, 3, 4,  plus  the pilot
scale activated sludge influent and final effluent.

     In addition to the priority pollutants, Effluent Guidelines  Division
contracted laboratory analyses for the same "classic parameters"  that were
analyzed in the Phase I screening study.

     A final report will be prepared which will, in essence, report  all
analytical data obtained at each refinery, the refinery's  data, analytical
quality control, and description of the refinery, crude oil  sources,  etc.

REFERENCES

(1) Public Law 92-500, 92nd Congress, S. 2770, October 18, 1972,  Page 1.

DISCUSSION

F. L. Robertaccio;  The methodology used for the "carbon-biological"  system
being screened in this study is insufficient to be an adequate representation.
I would like your comments on this remark and your opinion on how the aspect
of your study might effect selection of models for establishment  of effluent
guidelines.

Leon H. Myers;  First, I don't know how it will affect guidelines, that is
out of my baliwick; I am in the research department.  I mentioned, and  I
mentioned on purpose, that we only had three constraints.  Time,  money, and
manpower.  We are at each refinery 15 days, about four of  those days  are
taken up in the initial screening.  That leaves about 10 days that we are
there running the carbon systems.  I can't disagree with you, we  can't get
one good sludge age, much less two or three sludge ages to look at the
powdered system at all.  In my opinion, to use any of this data,  you  would
have to use it very carefully.  You have to remember that we are  operating
at a  quarter GPM over a small time period.  Our inclination is that  we need
at least a 10 GPM pilot scale study conducted parallel with  the full-scale
pilot scale study on powdered activated carbon and about 10 GPM granular
carbon study conducted at the same plant, over about a year's period  of time
because of seasonal variations.  We have discussed this with the  effluent
guidelines personnel and with research headquarters personnel.  We haven't
found anyone to disagree with us, we also haven't found any  funding response
Does that answer your comments?
                                     126

-------
BIOGRAPHY

    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 Sources Section,
of the Source Management Branch at the Robert S.
Kerr Environmental  Research Laboratory, Ada,
Oklahoma.
                                     127

-------
           TABLE 1.  SETTLEMENT AGREEMENT LIST OF 65  POLLUTANTS
Acenaphthene
Acrolein
Acrylonitrile
Aldrin/Dieldrin
Antimony and compounds
Arsenic and compounds
Asbestos
Benzene
Benzidine
Beryllium and  compounds
Cadmium and compounds
Carbon tetrachloride
Chlordane  (technical mixture and metabolites)
Chlorinated benzenes  (other than dichlorobenzenes)
Chlorinated ethanes  (including  1, 2-dichloroethane,  1,1,1-trichloroethane,
                      hexachloroethane)
Chloroalkyl ethers  (chloramethyl, chloroethyl, and mixed  ethers)
Chlorinated naphthalene
Chlorinated phenols  (other than those listed elsewhere; includes  trichloro-
                      phenols and chlorinated cresols)
Chloroform
2-chlorophenol
Chromium and compounds
Copper and compounds
Cyanides
DDT  and metabolites
                                     128

-------
Dichlorobenzenes  (1,2-,1,3-, and 1,4-dichlorobenzenes)
Dichlorobenzidine
Dichloroethylene  (1,1-and 1,2-dichloroethylene)
2,4-dichlorophenol
Dichloropropane and dichloropropene
2,4-dimethyl phenol
Dinitrotoluene
Diphenyihydrazine
Endosulfan and metabolites
Endrin and metabolites
Ethyl benzene
Fluoranthene
Haloethers (other than those listed  elsewhere;  includes chlorophenyphenyl
            ethers, bromophenylphenyl  ether,  bis (dischloroisopropyl)ether,
            ether, bis-(chloroethoxyl)  methane  and polychlorinated diphenyl
            ethers)
Halomethanes (other than those  listed  elsewhere; includes methylene chloride,
              methyl chloride, methyl bromide,  bromoform, dichlorobromomethane,
              trichlorof1uoromethamne,  dichlorodif1uoromethane)
Heptachlor and metabolites
Hexachlorobutadi ene
Hexachlorocyclohexane  (all  isomers)
Hexachl orocycl opentadi ene
Isophorone
Lead and compounds
Mercury and compounds
Naphthalene
Nickel and compounds
Nitrobenzene
                                     129

-------
Nitrophenols (Including 2,4-dinitrophenol, dinitrocresol)
Nitrosamines
Pentachlorophenol
Phenol
Phthalate esters
Polychlorinated biphenyls (PCBs)
Polynuclear aromatic hydrocarbons (including benzanthracenes, benzopyrenes,
                                   benzofluoranthene, chrysenes, diben-
                                   zanthracenes, and indenopyrenes)
Selenium and compounds
Silver and compounds
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD)
Tetrachloroethy1ene
Thallium and compounds
Toluene
Toxaphene
Trichloroethylene
Vinyl chloride
Zinc  and compounds
                                    130

-------
                  TABLE 2.   SPECIFIC COMPOUNDS
Compound
 1.   *acenaphthene
 2.   *acrolein
 3.   *acrylonitrile
 4.   *aldrin
 5.   *dieldrin
 6.   *benzene
 7.   *benzid1ne
 8.   *carbon  tetrachloride (tetrachloromethane)
 9   *chlordane (technical mixture & metabolites)

     Chlorinated benzenes (other than dichlorobenzenes)

10.   chlorobenzene
11.   1,2,4-trichlorobenzene
12.   hexachlorobenzene

     Chlorinated ethanes (including 1,2-dichloroethane, 1,1,1-
         trichloroethane and hexachloroethane)

13.       *l,2-dichloroethane
14.       *l,l,l-trichloroethane
15.       *hexachloroethane
16.        1,1-dichloroethane
17.        1,1,2-trichloroethane
18.        1,1,2,2-tetrachloroethane
19.        chloroethane

     Oiloroalkyl  ethers (chloromethyl, chloroethyl and mixed ethers)

20.       *bis(chloromethyl) ether
21.       *bis(2-chloroethyl) ether
22.        2-chloroethyl vinyl ether

     Chlorinated naphthalene

23.       2-chloronaphthalene

     1-bromodecane Std.
     1-bromododecane Std.

     Chlorinated phenols (other than those listed elsewhere; includes
         trichlorophenols and chlorinated cresols)

24.       2,4,6-trichlorophenol
25.      p-chloro-m-cresol

26.   *chloroform (trichloromethane)

                                   131

-------
27.  *2-chlorophenol

     DDT and netabol ites

28.      *4,4'-DDT
29.       4,4'-DDE
30.       4,4'-DDD  (p.p'-TDE)

     Dichlorobenzenes  (1,2-; 1,3-;  and  1,4-dichlorobenzenes)

31.      *l,2-dichlorobenzene
32.      *l,3-dichlorobenzene
33.      *l,4-dichlorobenzene
     Di chl orobenzi di ne

34.       3,3'-dichlorobenzidine

     achloroethylenes  (1,1-dichloroethylene and 1,2-dichloroethylene)

35.       *1,1-di chloroethy1ene
36.       *1,2-trans-dichloroethylene

37.  *2,4-dichlorophenol

     Di chl oropropane and  di chl oropropene

38.       1,2-dichloropropane
39.       1,3-dichloropropylene  (i,3-dichloropropene)

40.  *2,4-dimethylphenol

     Dini trotoluene

41.       2,4-dinitrotoluene
42.       2,6-dinitrotoluene

43.  *1,2-di pheny1hydrazi ne

     Endosulfan and metabolites

44.       *orendosulfan
45.       *3i-endosulfan
46.       endosulfan sulfate

     Endrin and metabolites

47.       *endrin
48.       endrin aldehyde
49.       endrin ketone
                                   132

-------
50.   *ethylbenzene
51.   *fluoranthene

     Haloethers (other than those listed elsewhere)

52.       *4-chlorophenyl phenyl ether  (p-chlorodlphenyl ether)
53.       *4-bromophenyl phenyl ether
54.       *bis (2-chloroisopropyl) ether
55.       *bis (2-chloroethoxy) methane

     Halo?rethanes (other than those listed elsewhere)

56.       *methylene chloride  (dichloromethane)
57.       *methyl chloride  (chloromethane)
58.       *methyl bromide (bromomethane)
59.       *bromoform (tribromomethane)
60.       *dichlorobromomethane
61.       *trichlorofluoromethane
62.       *dichlorodifluoromethane
63.       chlorodibromomethane

     Heptachlor and metabolites

64.      *heptachlor
65.       heptachlor epoxide

66.  *hexachlorobutadiene

     Hsxachlorocycloharane (all  isomers)

67.      *a-BHC
68.      *|3-BHC
69.      *Y-BHC  (lindane)
70.      *<5-BHC

71.  *hexachlorocyclopentadiene
72.  *isophorone
73.  *naphthalene
74.  *n1trobenzene

     MtrophenoZs  (including  2,4-dinitrophenol  and  dinitrocresol)

75.      2-nitrophenol
76.      4-nitrophenol
77.     *2,4-dinitrophenol
78.      4,6-dinitro-o-cresol

     Mtrosani nes

79.      N-nitrosodimethylamine
80.      N-nitrosodi-n-propylamine
81.      N-nitrosodiphenylamine

                                   133

-------
   82.  *pentach1orophenol
   83.  *phenol

        Phthalate esters

   84.      bis (2-ethylhexyl) phthalate
   85.      butyl benzyl phthalate
   86.      di-n-butyl phthalate
   87.      diethyl phthalate
   88.      dimethyl phthalate

        Pol ychl orz not ed bi phenyl s (PCff s)

   89.      PCB-1242 (ArocUor 1242)
   90.      PCB-1254 (Arochlor 1254)

        Polynuclear aronntic hydrocarbons (including benzanthracenes,
            benzopyrenes, benzofluoranthene, chrysenes, dibenzanthracenes,
            and indenopyrenes)

   91.      1,2-benzanthracene
   92.      benzo[a]pyrene (3,4-benzopyrene)
   93.      3,4-benzofluoranthene
   94.      11,12-benzofluoranthene
   95.     *chrysene
   96.      acenaphthylene
   97.      anthracene
   98.      1,12-benzoperylene
   99.      fluorene
  100.      phenanthrene
  101.      l,2:5,6-dibenzanthracene
  102.      indeno  (l,2,3-C,D)pyrene
  103.      pyrene

  104.  *253,7,8-tetrachlorodibenzo-p-dioxin (TCDD)
  105.  *tetrachloroethylene
  106.  *toluene

  107.      *toxaphene

  108.  *trichloroethylene
  109.  *vinyl chloride (chloroethylene)
        1-brompdecane (possible internal standard)
        1-bromododecane (possible internal standard)


                                TOTAL METALS

1.  Arsenic       5.  Lead          9.  Antimony      13.  Zinc
2.  Beryllium     6.  Mercury      10.  Copper        14.  Asbestos
3.  Cadmium       7.  Nickel        11.  Selenium      15.  Cyanides
4.  Chromium      8.  Thallium     12.  Silver


                                      134

-------
                      Table 3




REFINERIES SELECTED FOR THE PRIORITY POLLUTANT STUDY
Refinery
Gulf
Exxon
Hunt
Clark
Texaco
Mobil
Phillips
Shell
Exxon
Exxon
Coastal States
ARCO
Location
Philadelphia, Pa.
Bay town, Tex.
Tuscaloosa
Hartford, 111.
Lockport , 111 .
Augusta, Kan.
Sweeney, Tex.
Anacortes, Wash.
Benecia, Calif.
Billings, Mont.
Corpus Christi, Tex.
Philadelphia, Pa.
Refinery
Class
C
E
A
B
B
B
C
B
B
C
C
B
Treatment System
Trickling Filter
Activated Sludge
Aerated Lagoon
_ Activated Sludge
Activated Sludge
Filtration
Activated Sludge
Oxidation
Aerated Lagoon
Activated Sludge
Activated Sludge
Aerated Lagoon
Activated Sludge
Activated Sludge
                          135

-------
                                   Table 4
                  PRIORITY POLLUTANTS SAMPLING INFORMATION
  Parameter

Non Volatile

Metals

Mercury

Cyanide

Phenolics

Asbestos

Volatile
 24-hour
Composite

    X

    X
 Grab
Sample
                  X

                  X

                  X

                  X

                  X
    Sample
   Container

1 Gallon Glass

1 Gallon Plastic

1 Quart Plastic

1 Quart Plastic

1 Quart Glass

1 Quart Plastic

40 ml Vial
Preservative

    Ice

    HN03

    HN03

    NaOH

    H3P04

    Ice

    Ice
                                   Table 5
                  CLASSICAL PARAMETERS SAMPLING INFORMATION
Parameter
BOD5
TSS
COD
TOC

NH3-N
24-hour
Composite
X
X
X
X

X
Grab Sample
Sample Container
Plastic
Plastic
Plastic
Plastic

Plastic
Preservative
Cool 4°C
Cool 4°C
H2SOA. to pH<2
H2S04 to pH<2
Plus Cool, 4°C
HoSO^ to pH<2
       Cr°                X

       Sulfide            X

       Oil & Grease
                                     Plus Cool, 4°C

                          Plastic    HN03 to pH<2

                          Plastic    2 ml Zinc Acetate

                          Glass      H2SOA. to pH<2

                                     Plus Cool, 4°C
                                      136

-------
                              Table 6

                   SCHEDULE OF SAMPLING PERIODS
Day 1 234 5678
W Th F S Su M T W
S S S S S
C C C C C
R R R R R
E E E E E
E E E E E
N N N N N
Sample locations are depicted in
9 10 11 12 13 14 15 1617
Th F S Su M T W Th F
S S S
C C C
R R R
E E E
E E E
N N N
Figure 11.
Table 7
SAMPLING FREQUENCY

Sample Site
Intake
Separator
Bio-Effluent
Mixed-Media Filter Eff.
Activated Carbon Eff.
Reaenerat.pH farhnn Fff.
AND LOCATION PROGRAM
Day
1 2 3 6 9 12 15
XXX
XXX
X X X X X X X
X X X X
X X X X
X X X X
Activated Sludge/Activated
   Carbon Influent

Activated Sludge/Activated
   Carbon Eff.
                                137

-------
i
   Figure 1.  INTAKE SCREEN HOUSE
Figure 2.  INTAKE WATER SAMPLE POINT




                 138

-------

Figure 3.  Corrugated Plate Interceptor-Hunt/Tuscaloosa
Figure 4.  Bypass Line From Corrugated Plate Interceptor
                     Hunt/Tuscaloosa
                           139

-------
    Figure 5.  DAF Unit-Hunt/Tuscaloosa
Figure 6.  Final Effluent-Hunt/Tuscaloosa
                    140

-------
Bio-Tree ted
Effluent
                                 Pressure

                             © GouQe
                               „
                       :v.«/-*•*  <""'-'
                       !'4 // ' . ' ' * '.'•••'
                                     '
                       '.-" *•  /V**'  v«
                       -' *  X      / ,  *•'
                              To Corbon
                              Columns
                                             3  Limestone
                                                Rock
                                             Stainless Steel Screen
                                             Teflon Gasket
           Figure  7.   MIXED-MEDIA FILTER DIAGRAM
                               141

-------
       Multi-Media
      Filter Effluent
                          Pressure Gauge
                                Teflon Gasket
                              -•—Glass Column
                                36  Carbon
                                 3 Limestone Rock
                                     Teflon Gasket
                           Carbon Treated
                               Water
Figure 8.    CARBON  COLUMN
                   142

-------

Bio- Effluent


















i


















/ /
/ t
n '
t
>
f
/ i
•.
t i
t
t /

.
K
,, Ul /
, ' 1

y "
/ 4
m*f J
_J
~* / .
«J X
'' */
/ Ul'
/. • 2 ' '
'i ,

4
* /
4
y x '
//

•J /
H y
5
/ /

^'x
'/"/'
,



















1


























4
r 1








^
o
a:
«•
^
•i
^
>
M>
0
z


CARBON
































2.
' 1








f
5
n»
••
^^
•
>

2>


CARBON

































| 1







t

o
fl:
>

10
o
^

CARBOI






























P






a
ui
i_

oc
UJ

^
UI
0
111
^
CARBO








V T T T
Figure 9.  PICOT  CARBON SYSTEM

-------
 Mixer.
-Air Manifold
       m
g/in Bio-
System
Effluent
                                                              Final Clarifier
                                                      Weir


                                                      Sludge Blanket




                                                        Effluent
GlassWool
Packed
   Sump
  Figure  10.  PILOT  SCALE ACTIVATED SLUDGE  SYSTEM

-------
            "CONSIDERATIONS FOR DEFINING SUBSTANCES
                 HAZARDOUS TO THE ENVIRONMENT"

                        A. Karim Ahmed
      Staff Scientist, Natural Resources Defense Council

      I  am  pleased to be here today to address this conference
about  one of the most important programs embarked upon by the
federal  government to regulate the proliferation of toxic sub-
stances  in  our environment.  This morning, Mr. Ridgeway Hall of
the Environmental Protection Agency  (EPA) gave us an excellent
presentation of the events that led to the settlement agreement
- known  to  some of us as The Consent Decree - between the Natural
Resources Defense Council  (NRDC) and other environmental groups
and the  EPA.  This out of court settlement, which was approved by
the U.S. District Court for the District of Columbia on June 9,
1976,! is a wide-ranging regulatory strategy, with a clearly de-
fined timetable, for controlling the discharge of toxic sub-
stances  into our nation's waterways, as required under the
Federal  Water Pollution Control Act  (FWPCA).2  since we have
heard a  good deal about EPA's present and proposed implementation
of the Consent Decree from Mr. Hall and other speakers, I will
attempt  to  give you a somewhat different overview of the same
subject  matter - namely the problem we face when we wish to de-
fine substances that are hazardous to the environment.  I will do
so in the context of reviewing with you some of the major federal
statutes that deal with the hazardous or toxic substances issue.

      I  would like to conceptually divide major federal statutes
that have dealt with the problems of toxic substances in the past
and do so currently into two main divisions:  (1) statutes that
primarily address toxic substances in terms of their effects on
human health, and (2) statutes that were intended to regulate the
impacts  of  toxic substances on the environment.  Now I have pur-
posely divided this into black-and-white terms and I will try to
explain  to  you why.  First, let's look at the statutes that deal
with problems of human health.  These generally come under the
purview  of  the Food and Drug Administration, under the Food, Drug
and Cosmetic Act;3 the Department of Labor, under the Occupa-
tional Safety and Health Act;4 and the Consumer Product Safety
Commission, under the Consumer Product Safety Act,5 and the
Federal  Hazardous Substances Act.6  These statutes, which I will
explain  as  we go along, primarily deal with effects on human
health,  and are regulated by the three federal agencies I have
just mentioned.

      On the other hand, the EPA, which was created in the early
70fs through the reorganization of the federal bureaucracy, was
given jurisdiction to regulate impacts on the environment.  How-
ever,  as we have now discovered, the environment also includes
man.  The EPA has a particularly difficult task in trying to de-
fine toxic  substances which have, first of all, an impact on the

                               145

-------
environment, and secondly, an impact on human health,  or both.
The various statutes that EPA has authority under  are  the Federal
Water Pollution Control Act,7 the Clean Air Act,8  the  Toxic Sub-
stances Control Act,9 the Safe Drinking Water Act,lu the Resource
Conservation and Recovery Act,11 and the Ocean  Dumping Act.1-'
EPA also regulates pesticides under the Federal Insecticide, Fun-
gicide and Rodenticide Act.1^

      The problem that we're faced with here is essentially one
of overlapping jurisdiction.  Let me give you examples of some of
the problems we have had of late.  When the issue  of fluorocar-
bon's impact on stratospheric ozone was first brought  to the at-
tention of the federal agencies, we had complete chaos-   In fact,
the first thing that the agencies did was to run to the  Depart-
ment of Justice to ask for a memorandum that would sort  out the
issue of legal jurisdiction.  The Natural Resources Defense Coun-
cil had petitioned the Consumer Product Safety  Commission (CPSC)
who, in spite of having jurisdiction over only  about five to ten
percent of the total aerosol products on the market, had strong
regulatory authority under the Consumer Product Safety Act.14
When we confronted them with our petition, the  Commission argued
at first that the problem was one of environmental concern and
only indirectly would affect human health - i.e. the increase of
ultra-violet radiation, caused by environmental loss of  ozone
would indirectly lead to an increase in skin cancer incidence.
So, the CPSC felt that it really didn't fit under  their  jurisdic-
tion.  Similarly, the FDA said that they had statutory authority
with foods, drugs, and cosmetics as they affect human  health
directly, but had not been given authority to deal with  environ-
mental issues.  Lastly, the EPA put a final ironic twist to this
regulatory comedy by claiming that under the Clean Air Act it had
jurisdiction on the lower atmosphere only.  They were  not sure
about the stratosphere; perhaps some other agency  dealt  with the
upper atmosphere.  So this classic "passing-the-buck"  game began
back in 1974.  At that time, this issue was too new and  unfamil-
iar, and somewhat controversial, for any agency to want  to han-
dle.  We had a really troublesome situation where  none of the
agencies knew where their jurisdiction lay and  how they  should go
about regulating products under statutes that defined  their au-
thority narrowly, or at best, ambiguously.

      Another example, which is just beginning  to  emerge, is the
problem of genetic engineering.  In genetic engineering, we have
a naturally occurring substance called DNA which,  as you know, is
the subject of biological manipulation.  One would surely think
that the Department of Health, Education and Welfare,  under the
U.S. Public Health Service Act,15 would have clear jurisdiction
on this issue since, under credible experimental conditions, new
life forms may be produced, such as a bacterial system that could
be potentially infectious.  But the EPA now claims that, under
the Toxic Substances Control Act  (TSCA), DNA may be defined as a
chemical substance.  Consequently, under certain provisions of

                               146

-------
TSCA,  it may have jurisdiction.  Here, they're claiming jurisdic-
tion,  rather than avoiding it.  In a sense, we have another po-
tential conflict here - one of asserting  jurisdiction.

      With respect to the fluorocarbon-ozone  issue that I men-
tioned earlier,  what subsequently happened was a  formation of an
inter-agency committee under the auspices of  the  Council on Envi-
ronmental  Qulaity (CEQ) leadina to a thorough study by the Na-
tional Academy of Sciences.ib'17  A few months ago, as you know,
all three  agencies (the CPSC, and FDA and the EPA) jointly pro-
posed regulations governing the ban of fluorocarbons as propel-
lants in aerosol products.18  For the first time, a kind of coop-
erative regulatory venture is being experimented  with by the
federal agencies on a hazardous substance that does not fit under
neatly defined jurisdictions.

      There are, on the other hand, some  hazardous chemical sub-
stances that appear to be clearly defined under different stat-
utes. Let me give you an example - asbestos  is one of the best.
If one has a problem of asbestos in water, one would go to the
EPA, since under the Federal Water Pollution  Control Act or the
Safe Drinking Water Act they have jurisdiction.19 If it's an
air related problem, then one would also  go to the EPA, for under
Section  112 of the Clean Air Act, they have promulgated regula-
tions with respect to asbestos emissions.20   if it's a problem in
consumer  products, for example, in spackling  compounds, artifi-
cial fireplace logs, or asbestos-containing ceiling tile, one
would go  to the Consumer Product Safety Commission which has ju-
risdiction either under the Consumer Product  Safety Act or the
Federal  Hazardous Substance Act.21  If it's a matter of asbestos
in talcum powder or toiletries, one would go  to the FDA who have
clear authority under the Food, Drug and  Cosmetic Act.22  It it's
a labeling question - a label to warn consumers - one would go to
one of the line agencies or directly to the Federal Trade Commis-
sion  (FTC) which could issue regulations  under the Federal Fair
Labeling  and Packaging Act.23  If it's a  workplace hazard, one
would go  to the Department of Labor's Occupational Safety and
Health Administration for setting up occupational standards.z-
And finally, if it's a universal problem, one would go back to
the EPA.   Why?  Because they now have the Toxic Substances Con-
trol Act,25 enacted into law last year, which is  supposed to take
care of  everything generically and completely.  So, the EPA has
the biggest task again.

      Let's get down to specifics in defining substances that are
hazardous.  One way of defining substances that are hazardous to
the environment is to look at it in a narrow  "legal" sense.  It
is reasonable to expect that the statutes were written at^differ-
ent times  for different reasons, since the congressional intent
for enacting such laws tend to be rather  different.  But when one
examines  the various statutes on th« books, one find%that ^orae
of the laws do not have an explicit definition of what is meant

                               147

-------
by the term "toxic."  They might,  for example,  address the issue
of toxicity by alluding to questions of safety  and unsafety in
very general terms.  This is particularly  true  with the Federal
Food, Drug and Cosmetic Act  (FDCA).  The FDCA has  lengthy sec-
tions on procedural matters, about burden  of proof,  etc., but  the
actual definition of toxic substances is never  made clear.   This
also appears to be the case under  the Federal Insecticide,  Fungi-
cide and Rodenticide Act  (FIFRA),  which has no  definition of what
is a hazardous or harmful pesticide.  However,  a clearer state-
ment about hazardous substances  can be found in the Federal Haz-
ardous Substances Act (FHSA).  This particular  Act is  interesting
because it defines "hazardous" in  very narrow terms and yet has
broad application.  Under Section  2(f) of  the FHSA,  "hazardous
substance" means:

         "Any substance or mixture of substances which
         is  (1) toxic,  (2) corrosive, (3)  ah irritant,
         (4) a strong sensitizer,  (5) flammable or com-
         bustible, and  (6) generates pressure through
         decomposition, heat or  other means, if such
         substances or mixtures  may cause  substantial
         injury or substantial illness...."

Notice here that the definition  deals with either  health or
safety of human beings.  It does not deal  with  environmental is-
sues.   (This is an important section of the Act because later  I
will go back to it when I talk about a proposed EPA regulation.)
Under Section 2(g) of the same Act, we have a definition of what
is "toxic:"

         "Any substance (other than a radioactive  sub-
         stance) which has the capacity to produce in-
         jury or illness to man  through ingestion,  inha-
         lation or absorption through any  body  surface."

Here, again, toxic is defined with respect to human beings.  In
Section 2(h), we have a definition of "highly toxic,"  which is
defined in terms of three explicit guidelines that are employed
in animal tests:  (1) for ingested  substances, an "LD^Q  value which
is equal to or less than 50 milligrams per kilogram;  (2) for in-
haled substances, an LCso figure which is  equal to or  less  than
200 parts per million, or 2 milligrams per liter;  and  (3)  dermal
absorption, an LDso figure equal to or less than 200 milligrams
per kilogram.  We have here a definition of "toxic"  or "highly
toxic" in a very conventional sense of the term -  meaning acute
animal or human toxicity.  There is no reference to chronic tox-
icity, nor does it refer to environmental  harm  of  any  kind. Ob-
viously, the intent of this Act  was to regulate products or sub-
stances that would acutely injure  human beings.

      We should now examine statutes that  simultaneously deal
with human and environmental effects.  The best place  to start,

                               148

-------
I think, is the Toxic  Substances Control Act which, even thouqh
it is the most recently  passed law, is causing a great deal of
confusion in the minds of a lot of people who have to deal with
this statute.  Let  us  examine  whether we have any definition of
a hazardous or toxic substance in this Act.  It first occurs un-
der Section 4, which is  the section that deals with testing of
chemical substances and  mixtures, and here it is basically de-
fined in terms of substance or mixture which "may present an un-
reasonable risk of  injury to health or the environment."  That
is, the definition  is  very broad.  In subsection (2) (A) of Sec-
tion 4, mention is  made  of the kinds of standards or testing that
would be required to assess what is "unreasonable" - that is,
what are the unreasonable risks to health and the environment -
and mentioned are tests  for carcinogenesis, mutagenesis, terato-
genesis, behavioral disorders, cumulative or synergistic dis-
orders, etc., in addition to requiring testing for acute toxic-
ity, subacute toxicity,  and other chronic toxicity.  The same de-
finition occurs again  under Section 6 of the Act, wnich is the
section that deals  with  the actual regulation of hazardous sub-
stances, that provides the agency with the authority to remove a
toxic substance from the marketplace:

        "If  the Administrator finds that there is a
        reasonable basis to conclude that the manufac-
        ture, processing, distribution in commerce,
        use, or disposal of a chemical substance or
        mixture, or that any combination of such ac-
        tivities ,  presents or will present an unrea-
        sonable risk  of injury to health or the envi-
        ronment , the  Administrator shall by rule apply
        one  or more of  the following requirements to
        such a substance or mixture...."

     Finally, in Section 7 of the Toxic Substances Control Act,
we have the  Imminent Hazards section which, I think, is an impor-
tant provision to recognize.  Here, the Administrator is given
certain additional  powers to commence action when substances or
mixtures possess an "imminent or unreasonable risk of serious or
widespread injury to health or the environment."  And, once
again, the theme is both health and the environment.

     We shall now  examine the Federal Water Pollution Control
Act and see how clearly  the term "hazardous substance" is de-
fined.  We find that we  really do not have an unequivocal answer.
In a sense, it is a bit  of a mess because when one examines the
FWPCA even an experienced environmental attorney has a difficult
time in figuring out what the correct definitions are.  Appar-
ently the only place in  the FWPCA where the term "toxic pollu-
tant" is defined is under Section 502, paragraph 13:

        "The term  'toxic pollutant1 means those pollu-
        tants or combination of pollutants including


                               149

-------
         disease causing agents which...on the basis
         of information available to the Administrator,
         cause death, disease, behavioral abnormalities,
         cancer, genetic mutations, physiological mal-
         functions (including malfunctions in reproduc-
         tion) or physical deformities, in such organ-
         isms or their offspring."

      It would appear to be all inclusive.  One, would not use
such a broad definition to select a list of highly toxic pollu-
tants.  However, this is the manner by which "toxic pollutant"
has been defined under Section 307(a), which deals with the de-
velopment of effluent standards of toxic pollutants in the FWPCA.
The prohibition of toxic pollutants in toxic amounts is clearly
stated in Section 101(a)(3) of the Act which, when combined with
Section 502(13), gives us the "correct" intent of Section 307(a).
This is apparently the way the concept of "toxic pollutant" is
defined under the Act.

      In Section 311 of the FWPCA, on the other hand, we see a
completely different approach since it deals with the liability
of an accidental spill.  But it  has a broad definition of "dis-
charge" in this section, for it "includes, but is not limited to
spilling, leaking, pumping, pouring, emitting, emptying, or
dumping."  One would think that Section 311 would be even more
all-inclusive in terms of point discharges than accidental
spills; moreover, it goes on to define "hazardous substances" as
those that may "present an imminent and substantial danger to the
public health or welfare including, but not limited to, fish,
shell-fish, wildlife, shoreline and beaches."  Interestingly
enough, there is also a provision in the FWPCA giving the Admin-
istrator certain emergency powers under Section 504, where a si-
milar language is repeated, i.e. "presenting an iminent and sub-
stantial endangerment to the health of persons and to the welfare
of persons...."  What we have here is a provision that, on the
one hand, appears to be quite restrictive in defining emergency
powers of the Administrator and, at the same time, defines haz-
ardous substances quite broadly.  We must therefore ask: which of
the two statutory "sets" is greater?  Is Section 307(a) a "sub-
set" of Section 311; are they two separate, disjunctive "sets;"
or do they overlap?  We don't really have clear answers to
this/ in part because we don't have a Section 311 regulation pro-
mulgated as yet.  This only compounds the present regulatory con-
fusion.

      Let us now examine the problem of defining "hazardous sub-
stance" from a more technical or scientific point of view.  In
the past, promulgated regulations, reflecting the intent of fed-
eral statutes, emphasized acute toxicity - for example, the
Federal Hazardous Substances Act which, as I mentioned earlier,
only dealt with short-term acute toxicity and defined toxicity in
'-erms of LDso's and LCso's, etc.  More recently, there has been a

                               150

-------
growing concern about long-term chronic effects  and conseauentlv
about carcinogenic and mutagenic effects.  These toxic  substances
exert their  effects over a long time period when ingested in
small amounts  by man or animal and may cause cancer with latency
periods of 20  to 40 years.  As Mr. Hall mentioned this  morning
that was  one of the chief criteria used in selecting the priority
chemical  substances under the settlement agreement.  We are also
concerned with certain key environmental factors when defining
hazardous substances.  These are generally questions of:  (1) che-
mical persistence - biological and chemical degradibility,  (2)
movement  in  the environment, i.e., how does it get transported -
by air, by water, or by soil,  (3) bioconcentration in the envi-
ronment,  including biomagnification in the food  chain,  and  (4)
synergistic  and cumulative effects of the chemical substance.
These are all  now becoming part and parcel of defining  hazardous
substances from a scientific point of view.  In  selecting sub-
stances to be  regulated under the Consent Decree, all the above
factors were given appropriate weight, and we hope led  to a bet-
ter selection  process than had been used by the  agency  in the
past.

     Lastly,  I will briefly mention our views on the Consent
Decree.   As  you may recall, under Section 307(a) of the FWPCA,
the EPA had  ninety days, to publish a list of toxic substances for
which effluent standards were to be established. We interpreted
this section of the Act to apply to those highly toxic  or persis-
 tent substances for which there would be neither a technology-
based nor an economically-based effluent standard.  After initial
litigation by  NRDC, EPA in July 1973 proposed a  list of nine sub-
stances to be  regulated under Section 307(a) of  the FWPCA.  Most
of the substances were pesticides, a few metal ions, and PCB.  At
administrative hearings held by the EPA, the agency was literally
swamped with technical information from the affected industries
that claimed that the agency did not have an adequate scientific
basis  for setting proposed new effluent standards.  Consequently,
the EPA abandoned their proposal and did not promulgate effluent
standards under Section 307(a).  NRDC filed additional  suits
against EPA, claiming that there were a large number of chemical
substances  (certainly greater than nine) that could be  termed
toxic or  hazardous and ought to-be regulated under Section
307(a).

     To  make  a long story short, we finally ended in an out-of-
court  settlement which is now known as the Consent Decree.  In
arriving  at  the Consent Decree, it is very interesting  to note
some of the  issues that were agreed to by both sides.   There was
a recognition  that there were different provisions in the FWPCA
that could regulate toxic substances, and that Section  307(a) was
not necessarily the only handle that should be used.  These addi-
tional provisions include Sections 301, 304 and  306.  The basis
for developing a priority list under this concept was that  the
agency was given more flexibility in dealing with toxic

                               151

-------
substances, some of which have only recently been  shown  to  have
potential long term problems of carcinogenicity and mutagenicity.
Thus, instead of using Section 307(a) as the only  basis  for regu-
lating toxic substances, with its  limited timetable schedules  and
larger burden of proof on the agency, EPA would use a combination
of different provisions under the FWPCA, using a technology-based
and New Source Performance Standard approach as provided for in
Section 306.  What we have now is a list of some 129 chemical
compounds that have been identified as priority substances.  Ef-
fluent standards for six substances have already been promulgated
under 307(a) provisions (as required under the Consent Decree)
and the standards for the remaining substances will be issued  un-
der BAT (best available technology) provisions of  the Act.

      The big puzzle that now remains is the development of regu-
lations under Section 311 of the FWPCA.  The agency has  proposed
a list of substances to be regulated under Section 311,  which, I
think, is totally inadequate since most of the information  on
hazardous substances in the proposed regulations is based upon
aquatic toxicity, and the agency has chosen to define Section  311
in those terms.  And to make matters even more puzzling,  EPA used
the same criteria of defining toxicity as the Hazardous  Sub-
stances Control Act which, as I mentioned earlier, only  defined
acute toxicity.  The agency has essentially used the same LD$Q,
LCso figures in arriving at their conception of what is  acutely
toxic to human beings.  They have explicitly ignored questions of
chronic effects as is clear from the preamble to the proposed  re-
gulations. 26  jn fact, EPA has made a point of not being con-
cerned with issues of chronic effects, for reasons which are to-
tally unclear.  Consequently, these proposed regulations are
still pending and we do not know when they will be promulgated.

      It is clear that whatever regulations will now be  adopted
under the FWPCA, we will have to be concerned both with  the im-
plementation of the Consent Decree and with the implementation of
the Toxic Substances Control Act.  For example, a  major  issue
right now is the inventory reporting provision under TSCA.   This
provision requires EPA to obtain from the affected industries  a
complete knowledge of what chemical substances are being used, in
what amounts, and for what purpose.  Under the Consent Decree,
EPA has been gathering similar information on substances on the
priority list.  These same questions have not been addressed by
the agency under the Toxic Substances Control Act, where they  are
given authority to seek such information.  EPA has been  subjected
to a lot of criticism by NRDC  and others about the way  they've
gone about collecting this critical bit of information.   In the
final analysis, if we are to have meaningful control of  the dis-
charge of toxic substances into our environment, we will have  to
see a dovetailing of the effort between the Water  Program staff
of the EPA, who are implementing the Consent Decree program, and
the staff of the Office of Toxic Substances, who are trying to
implement the Toxic Substances Control Act.  Only  then can  we


                               152

-------
have confidence  in  the ultimate success of EPA's program to con-
trol the ever  growing impact of hazardous substances on the envi-
ronment.  Thank  you very much.
REFERENCES
 2.
 3.
 4.
 5.

 6.
 7.
 8.
 9.
10.
11
12
 13
 14
 15
 16
 17
 18.
 19.
 20.
 21.
 22.
 23.
 24.
 25.
 26
Natural Resources Defense Council,  et al.  v.  Train   8  E  R c
2120 (D.D.C. 1976).                ~         	     J* "  '
Public Law 92-500; 33 U.S.C.  §§1251 et seq.
21 U.S.C. §§301 et seq.
                             1590.
                                              1207;  90  Stat.
Public Law 91-596; 84 Stat.
Public Laws 92-573, 94-273,  94-284;  86  Stat,
503-510, 514; also 15 U.S.C.  §§2052  et  seq.
15 U.S.C. §§1261 et seq.
Supra, Reference 2.
Public Law 91-604; 84 Stat.  1676.
Public Law 94-469; 15 U.S.C.
Public Law 93-523; 88 Stat.
Public Law 94-580; 42 U.S.C.  §§6901  et  seq.
Marine Protection, Research  and  Sanctuaries  Act  of  1972;
Public Laws 92-532, 93-254,  93-62, 94-326;  86  Stat.  1052;  88
Stat. 50; 88 Stat. 1430;  89  Stat.  303;  90  Stat.  725.
Public Laws 92-516, 94-51, 94-109, 94-140;  7 U.S.C.  §§136  et
se
                              §§2601 et seq.
                             1660.
Petition of the Natural  Resources Defense Council to the
Consumer Product Safety  Commission,  November 19,  1974.
42 U.S.C. §264; also  petition of the Environmental Defense
Fund and the Natural  Resources Defense Council to the De-
partment of Health, Education and Welfare, October 28,  1976.
Fluorocarbons and  the Environment, report of the  Federal
Task Force on Inadvertent Modification of the Stratosphere,
Council on Environmental Quality, 1975.
Halocarbons; Effects  on  Stratospheric Ozone, Panel on Atmos-
pheric Chemistry,  National Academy of Sciences, Washington,
D.C., 1976.
42 Fed. Reg. 24536 et seq, May 13, 1977.
Supra, References  2 and  10.
Supra, Reference 18.
Supra, References  5 and  6.
Supra, Reference 3.
Public Law 93-608; 18 U.S.C.  §1457.
Supra, Reference 4.
Supra, Reference 4.
^0 Fed. Reg. 59959-60017, December 30, 1975.
                                153

-------
BIOGRAPHY

   A. Karim Ahmed is Staff Scientist
with Natural Resources Defense Coun-
cil, Inc. (NRDC), New York, New York,
and Adjunct Assistant Professor at
State University of New York, College
at Purchase.  He holds a B.S. degree
in Physics from University of Karachi,
a M.S. degree in Chemistry and a Ph.D.
degree in Biochemistry from University
of Minnesota.  Before assuming his
position with NRDC, Karim served as
Research Director for Minnesota Public
Interest Research Group, and Executive
Assistant to Director at Consumers
Union.  He is a member of Air Pollu-
tion Control Association, American
Chemical Society, American Public
Health Association, New York Academy
of Sciences, and Scientist's Institute
for Public Information.
                               154

-------
    SESSION V

     PROBLEMS


Chairman

George J. Putnicki

Visiting Professor, Environmental Sciences
University of Texas at Dallas, Texas


Speakers

Dwight G. Ballinger
"EPA's Analytical Development Program for
Problem Pollutants"

Fred T. Weiss
"Fates, Effects and Transport Mechanism of
Pollutants in the Aquatic Environment"

Donald I. Mount
"Measuring Aquatic Impact of Toxic Contaminants"

Davis L. Ford
"An Overview of Advanced Treatment Systems"
        155

-------
BIOGRAPHY           George J. Putnicki

       George J. Putnicki is currently a Visiting
Professor in the graduate programs in Environmental
Sciences at the University of Texas at Dallas.  He
holds a B.S. in Civil Engineering from Marquette
University and a M.S.  from Oregon State University.
Prior, Mr. Putnicki has had extensive experience
in the EPA Region Six and was Deputy Regional^
Administrator and Director of the Hazard Materials
Control  and  Surveillance Control Divisions.

        Mr.  Putnicki has received many awards and
commendations from the EPA and is a registered P.E.
in Oklahoma, Texas and Louisiana.
                                        156

-------
     ERA'S ANALYTICAL  DEVELOPMENT PROGRAM FOR PRIORITY POLLUTANTS


                     Dwight G.  Ballinger, Director
     Environmental Monitoring and Support Laboratory - Cincinnati


INTRODUCTION

     Previous  speakers  in this Open Forum have discussed the details
of the  1976 Consent Decree for priority pollutants and the development
of a list of elements and compounds for which standards are to be set.
I will  focus my  remarks on the analytical methods requirements involved
in the  implementation of the Decree and describe briefly what EPA is
doing to meet  these requirements.

     Although  the  standards will  be established by 1983, this will be
only the first step in  reducing the volume of these hazardous materials
discharged to  surface waters of the United States.  The standards will
be incorporated  in permits administered by the states and EPA and the
permit  conditions  will  require the monitoring of waste discharges and
the reporting  of the volume and concentration of these pollutants in
each discharge.  Section 304(g)  of the Federal Water Pollution Control
Act requires the Administrator of EPA to promulgate test procedures for
use in  determining compliance with permit conditions.  Under this section,
an analytical  method must be selected for each parameter listed in the
permits and these  procedures or an acceptable alternative, are to be
used in all waste  monitoring.   Test procedures were first published
in 1973 and were revised and expanded in the Federal Register of
December 1, 1976.  In practice,  the list of analytical methods are first
published as proposed,  public comments are received and considered, and
then the test  procedures are published as final regulations.  The
methods in the original  listing and later revisions were chosen in close
cooperation with the state agencies, other federal agencies, and method
standardization  groups  such as the committee for Standard Methods for
the Examination  of Water and Wastewater and ASTM Committee D-19.  The
petroleum industry is ably represented on Committee D-19 and the Committee
has provided significant input to the selection of analytical methods.

     Basically,  the analytical methods requirements for the priority
pollutants are the same as those for other pollutants	a proven,
^resented at the Second  Open Forum on Management of Petroleum Refinery
Wastewater, Tulsa, Oklahoma, June 7, 1977
                                    157

-------
defensible test procedure for each of the elements and compounds  for
which standards are set and monitoring is required.

PROBLEM DEFINITION

     Such a simple statement, however, is deceptive.  A better perspective
on the problem discloses a number of critical factors.  The original
list of 65 toxic substances contained a number of groups of organic com-
pounds such as chlorinated benzenes, chlorinated phenols, haloethers, and
polynuclear aromatic hydrocarbons.  When each of the individual compounds
in these groups are defined the list of specific compounds comes  to about
120.  If standards are set for each compound, a test procedure must be
available for that compound.  To meet the requirements for monitoring, test
procedures for these compounds must be available by 1983.  The majority of
these procedures will require analysis for complex organic structures.

     These constituents originate in a wide variety of industrial wastes
including plastics manufacturing, the production of synthetic organic
chemicals, and, of course, petroleum refining.  The priority pollutants
often occur as by-products rather than the principal product and  are
therefore in relatively small amounts.  While most of them have been
identified in waste streams, they have not often been measured in environ-
mental samples with adequate precision and accuracy.  Preliminary liter-
ature searchs have failed to produce the methodology required and a sig-
nificant amount of new development will be necessary.

     Among the list of compounds for which test procedures will be
required are a number of interest to the petroleum industry.  In  the
group of polynuclear aromatics are 16 individual compounds; there are
11 phenols and three nitrobenzenes on the list.

     The anticipated maximum permissible concentrations in the permits
have yet to be established since they will be dependent on the best
available treatment, the relative toxicity, and other factors.  These
concentrations, however, will be minimal and the analytical "target" for
methods development is 10 yg/1.  Based upon previous work, even lower
working ranges may be necessary to detect and measure the specific
pollutants.

     An appropriate test procedure must measure these concentrations of
the pollutant with positive qualitative identification, in wastewaters
containing many interferents of similar chemical structure and present
in amounts 100 to 1000 times the measured constituent.

     In addition, the test procedure should not be tailored to a  particular
wastewater, but should be applicable in all wastes where the pollutants
can occur.  This variety of substrates will require extensive separations
and cleanup to achieve specificity in the test method.

     Since the test procedure will be used by both the discharger and
the regulatory agencies, it must be within the technical and economic
capabilities of these laboratories.  The method must be as simple as


                                    158

-------
possible to minimize  the  workload, the instrumentation used must be
readily available on  the  open market at reasonable cost, and there must
be a  sufficient number  of analysts having the necessary skills and train-
ing to perform these  analyses.   These constraints are common to all test
procedures promulgated  for environmental  monitoring and must be considered
in the development and  final  selection of an analytical method to be used
nationwide.  While a  test procedure developed in an academic laboratory may
produce satisfactory  results  in the hands of the research scientist who
developed it, it may  not  meet the needs for routine monitoring.

     The problem for  EPA  and  the task assigned to the Environmental Moni-
toring and Support Laboratory in Cincinnati is to have all these analytical
methods available by  June 30, 1983.  By that time, under ideal conditions,
the methods will have been thoroughly tested on a large number of actual
wastes samples, subjected to  round-robin testing by a group of environ-
mental laboratories,  have been widely disseminated for familiarization,
and perhaps adopted  by  industry groups and the method standardization
organizations as standard procedures.

     We fully recognize the magnitude of the task and EPA is committed
to applying its resources to  see that methods are available to carry
out the monitoring requirements inherent in the Consent Decree.

CURRENT STATUS OF METHODOLOGY

     A number of elements and compounds identified as priority pollutants
are now incorporated  in discharge permits.  Test procedures for these have
been published in accordance  with Section 304(g) of PL 92-500.  The most
recent amendments, published  in December 1976, include approved methods
for 16 of  the 65 pollutants in the Consent Decree.  Thus test procedures
for all of the heavy  metals and a few of the organics are already avail-
able and published.   Methods  development research at the Cincinnati labora-
tory has produced procedures  for many of the common pesticides, chlorinated
compounds, and volatile halogenated compounds.  These methods are widely
used and are being standardized and adopted by EPA, ASTM, and Standard
Methods for the Examination of Water and Wastewater.  They will be incor-
porated in the monitoring program as permit conditions require.  At the
time of the signing  of  the Consent Decree, methods had been developed
for 34 of  the 65 elements and compounds listed, although not all of the
methods had been thoroughly tested in a variety of wastes and few of them
had been subjected to inter!aboratory study considered necessary for full
documentation of the  procedures.

METHODS DEVELOPMENT  CONTRACTS

     The methods research described has been conducted with limited re-
sources.   It is apparent that a crash program will be necessary to meet the
needs of priority pollutant standards.  To meet this challenge, EPA is
developing a series  of  research contracts leading to analytical methods for
these pollutants.  A  Request  for Proposal has been issued calling for as
many as 12 contracts  covering 114 organic compounds.  Each of the contracts
                                     159

-------
is for test procedures based upon similarity in chemical structure and/or
predicted instrumental response for a group of compounds.

     The contract effort is divided into two phases.  Phase I requires
the development and evaluation of an interim procedure in the laboratory,
based on response to pure standards of the material in a matrix of simi-
lar compounds.  After the successful completion of this task, Phase II
will require testing the method in a minimum of five industrial waste-
waters, covering at least five of the SIC codes specified in the Consent
Decree.  In addition, the contractor will determine the stability of these
samples over a seven-day period of holding and prescribe adequate methods
for the preservation of the samples.  Where possible gas chromatography is
to be the primary approach and specific columns and detectors are to be
used in the development of the interim methods.  At least two dissimilar
chromatographic columns must be used.

     The contract period is to include 12 months of laboratory and field
work, with three additional months for report preparation.   The final
report must provide complete method descriptions in a standard format,
as well as statements of precision and accuracy obtained on the actual
wastewater samples examined.  The minimum detection limit of the method
in the waste samples must also be determined and reported.   Contracts
are to be awarded by October, 1977, and the final contract product is
to be a series of test procedures suitable for routine monitoring of
the priority pollutants.  The total contract costs are projected to be
between $1 million and $2 million.

     Experience indicates that the task of developing these methods is
formidable.  Difficulties with separation of the measured constituent
from the interferring matrix can be expected.   Instrumental conditions
will need to be modified to provide quantitative results.  The detection
limits and working range desired may be difficult to obtain and the
precision and accuracy may be initially unsatisfactory.  Additional refine-
ment of the methods will be needed as working experience identifies
problems.  Since the final proof of success in methods development is  the
widespread use of the procedure, efforts will  be made to provide the
methods to interested groups for evaluation prior to any promulgation  in
the regulations.  The method descriptions and supporting data will be
forwarded to ASTM and Standard Methods for their consideration as stand-
ardized procedures.

     In keeping with the policies of EPA, a parallel effort in quality
assurance will be carried out.  The Quality Assurance Branch of the
Environmental Monitoring and Technical Support Laboratory routinely
provides reference samples for many of the contaminants included in
water supply and wastewater regulations.  Standard reference samples
for the priority pollutants will be developed to the extent that avail-
able resources permit.  A series of quality control samples, of known,
stabilized concentration, will be prepared and made available to labora-
tories performing analyses on wastewaters containing the pollutants.
A parallel set of performance samples, in the working range of the methods
but in concentrations unknown to the analyst, will be developed to evaluate


                                    160

-------
performance  capabilities of laboratories conducting analyses on priority
pollutants.   These samples will be  developed through inhouse and contract
research  and should be available  for a significant number of the compounds
by 1983.   Quality control and performance samples for most of the heavy
metals listed in the Consent Decree are already available from the Cincinnati
laboratory.

SUMMARY

     The  setting of standards for priority pollutants requires a major
effort in establishing adequate test procedures for the measurement of
these substances in environmental samples.  The method requirements in-
clude approximately 120 elements  and compounds many of which are complex
organic structures.  The hazardous  substances occur in a variety of wastes
from different industries representing significantly different substrates.
The test methods must be specific for the material in the microgram per
liter range and must be within the  technical and economic capabilities
of industrial and governmental laboratories.

     EPA is approaching the challenge by means of a series of research
contracts for 12 groups of similar  compounds, which will provide test
procedures for 109 organic contaminants.  The total contract effort is
expected to cost more than $1.5 million.  The preliminary methods should
be available for field testing by 1980 and will be promulgated as regula-
tions by 1983.
 BIOGRAPHY

       Mr. Ballinger Is Director of the EPA Environmental
 Monitoring and Support Laboratories in Cincinnati. Mr.
 Bal linger holds a B.S. degree in Chemistry from the
 University of Cincinnati,  is the author of many publications
 in the analytical chemistry of water and wastes.  He is a
 member of the American Chemical Society Water Pollution
 Control Federation, American Water Works Association,
 and Research Society of America.
                                      161

-------
DISCUSSION

Leonard Crame, Texaco, Inc.: When the 1983 EPA toxic tests are developed, what
guarantee will we have that the final EPA accepted toxic pollutant test procedures will
be consistent with those tests used to develop guidelines?

Bellinger;  Those that have been used to define the presence or concentration?

Crame: What assurance do we have that the numbers that will be  calculated for guide-
line purposes will be consistent with what the  new test procedures will be?

Bellinger:  You'll have absolute assurance that that is the case.  In the first place, the
contracts now defining BAT are based upon the qualitative aspect. We don't have the
problem of identification that we will have  in the final.  They are also contracts that
define the  presence of the material and relative concentration.  There is no assurance
that the specific test procedures will be the same but given the technical  constraints of
both there  are only so  many approaches to the determination of an organic compound and
they very likely will be consistent, but not  exactly.  I  think, for  example, that our
final test procedure for monitoring will be perhaps at lower concentrations but require
cleaner, better separations and more specific.  I  think there will be a difference; I hope
it is not a significant difference.  I don't anticipate any changes in  it.

Crame:  If it does change anything, will there be a mechanism for changing guidelines?

Ballinger:  I  don't see  how there would be a difference. If for example the standard is
based  on the concentration in effluents, that standard was developed irrespective of how
it is measured.  So a measurement technique would not  change that if both are specific
for the pollutant. It may well be that the analytical method is more sensitive than
required.  If the standard is not restrictive,  that's great; I  hope that the standard is not
below this sensitivity,  then we are in trouble.  I  don't see the problem if these are  not
just exactly the same procedures.

Crame;  If you use one analytical method now and come up with another in 2 or 3 years
from now,  will this new method actually be checked  against the old method?

Bal linger.-  They will be checked against the same wastes.

Crame; Right, from our own experience with  certain refinery wastewaters even the
slightest change in an  analytical  test will give you significantly higher numbers.

Bal linger;  That is true in some measurements, say oil.  That's because some measure-
ments  are what we call our empirical that the answer you get depends on  the method.
That's not the case in general with organic structures based on GC or GC/MS.  I think
in one case we are dealing with what amounts to a definition of the result depending on
how you  do it.  Certainly that's true with what we call oil and grease.  I don't think
that same case applies when looking at GC/MS.

                                       162

-------
Ben Buchanan, Phillips Petroleum Company:  Will EPA issue or compile the methods
after they are developed?

Ballinger:  If you mean by an independent method description, yes, EPA  intends to
publish these methods in our format.

Buchanan: They haven't always been the same method as put out by ASTM and APHA in
the past. Are they going to be the same?

Ballinger;  We hope so. We will make every effort to get ASTM D19 to adopt the
adopt the methods that we have been gathering supporting data on, but they are totally
independent groups having a number of independent opinions and we have no control
over what D19 does. So it's possible that they will come out with a different method.

Buchanan: The official will be the EPA method?

Ballinger: The official will be the EPA method.

Buchanan:  I just wanted to comment that the level of skill required when you  get down
to parts per billion hasn't been required in the past and that laboratories  that are not
used to operating down in this area may find it more difficult to get close answers and
maybe it might even come to the fact that people with greater skill would have to be
employed to do analysis down in this region.

Ballinger: I don't think there's any doubt that when you lower a detection limit or a
working range you do require a greater efficiency on the part of the operator, better
control over instrument conditions and so on.

Buchanan:  In chromatography it's very difficult dealing with interferences as you
probably know, and higher molecular weight compounds are more difficult to separate
so there may be some real  problems with these.

Ballinger: There will be some real problems.

Buchanan;  Thank you, sir.

Leo Duffy,  Standard Oil Company of Indiana;  I  took by your comments that perhaps you
intend not to  use GC/MS in the forthcoming analytical contracts.  Is this right?

Ballinger; That is correct. We do not  intend to  use GC/MS in these contracts if we can
get away with it and just use GC, because of the technical and economic factors.  A
good GC/MS as you know will put you in the $150,000-$200,000 bracked  We con-
sider that not a good approach to routine analyses.  So our contracts specifically call
bor GC and specifically for certain column and detectors.

F. L. Robertaccio, DuPont;  This question does not relate to your topic,  but I  wanted

                                      163

-------
to ask it anyway.  What is your personal opinion of the possibility of laboratory certifi-
cation?

Bellinger:  There are basically no requirements in 92-500 for certification of
laboratories.  EPA cannot require that data come only from certified laboratories
because  it is not a part of the legislation. We do feel that the agency has a responsi-
bility for the data that it uses and in decision making therefore anticipate that we will
evaluate the laboratories, not certify; judge their data accordingly.  Certifying means
that we go in and inspect and say the results coming from this laboratory are certified
and correct. We do intend  to evaluate the laboratories.  It is a fine distinction but it
is a legal one.

John Hallett, Shell Oil Co: What is being done to evaluate the performance  of
analytical  contractors being used in the refining industry priority pollutant screening
and validation surveys?

Bellinger;  Right now we do not have the performance samples completely developed for
those contracts, but we are  working on it right now.

Hallett:  Thank you.

Bal linger:  However, that is simply an evaluation of EPA's contracts.
                                       164

-------
                      METHODS DEVELOPMENT CONTRACTS

  LISTED POLLUTANT                            NUMBER  OF COMPOUNDS IN GROUP

Phthalate Esters                                               6
Haloethers                                                     7
Chlorinated  Hydrocarbons                                      9
Nitrobenzenes  and Isophorone                                  4
Nitrososamines                                                3
Dioxin                                                        1
Benzidine                                                      3
Phenols                                                      11
Polynuclear  Aromatics                                        16
Pesticides and PCBs                                          25
Purgeables  (Volatiles)                                       26
Acrolein,  Acrylonitrile, Dichlorofluoromethane               3
                                     165

-------
                   "FATES, EFFECTS & TRANSPORT MECHANISMS
                  OF POLLUTANTS IN THE AQUATIC ENVIRONMENT"
                                Fred T. Weiss
                          Shell Development Company
                               Houston, Texas
ABSTRACT
     Evaluation of environmental effects from exposure to low concentrations
of inorganic or organic compounds in the aquatic environment must  include a
concern for the reactivities and fate of these compounds under natural
conditions.  At the present time it is not possible to model the detailed
fate of many compounds due to lack of accurate information on chemical
reactivity and physical transport.  However, during the past decade much
information has been obtained outlining the general chemical and physical
behavior of certain pollutants.  Consequently it is possible to make genera-
lized predictions of the fate and extent of persistence of a number of
compounds and to point out areas of research to obtain more complete chemical
and physical data.

INTRODUCTION

     The intent of this report will be to review, in summary form, the fate,
transport mechanisms and effects of the important classifications  of pollu-
tants of concern for petroleum refinery wastewater.  These can be  listed
broadly as Hydrocarbons, Organics and Metals as indicated in Figure I.  For
this discussion organics are considered as other than Hydrocarbons.  Since
Dr. Mount will discuss toxicity testing in the paper immediately following I
do not plan to elaborate on the topic of toxicity.  For each of the principal
classifications a great deal of environmental information has been obtained
during the last decade and reported in many diverse publications.  However, in
only a limited number of these studies are any quantitative results available.
Consequently, the information obtained generally provides suggested pathways
and recommends research to establish more completely the mechanisms and rate
of the processes which exist for the ultimate removal or tranformation to
other species which may represent the end product.

HYDROCARBONS

     Since the Forum has been called to review wastewaters from petroleum
refineries it is appropriate to deal first with the hydrocarbons present as
pollutants in aquatic systems.  Because of the overall significance much work
has been done and reported on hydrocarbon properties and behavior. One report
to which reference will be made is from the National Academy of  Sciences
entitled "Petroleum in the Marine Environment".21  Another is  from the American
Institute of Biological Sciences and is the Proceedings of the 1976 Symposium
on "Sources, Effects and Sinks of Hydrocarbons in the Aquatic  Environment".2
                                      166

-------
     A study of high degree of significance  to  our  discussion  is  that funded
by the National Science Foundation - Research Applied  to  National Needs  (RANN)
Program on the "Petroleum Industry in  the Delaware  Estuary" which is currently
active.24   This project is staffed with  scientists  from Rutgers University and
the Philadelphia Academy of Natural Sciences and  is engaged in investigating
the relative effects of the petroleum  industry  and  other  sources  in the
Delaware Estuary.

     The general picture of fates of hydrocarbons in the  aquatic  system was
schematically illustrated by W. D. Garrett1  in  his  1972 publication on the
impact of  surface films on the properties of the  air-sea  interface.  Figure II
taken from this publication is an illustration  of the  forces which modify
hydrocarbon oil slicks on water.  Figure II  shows,  in  simplified  form, that
the forces which operate on oil in water include  evaporation,  solution,
emulsion formation, sedimentation, oxidation, biological  assimilation and
others.  It is important to note that  natural,  biogenic hydrocarbons, with
structures similar to petroleum hydrocarbons are  often present in natural
systems.21  The analytical differentiation between  biogenic and petroleum
hydrocarbons takes considerable effort in many  cases.

     Volatilization of the lower molecular weight hydrocarbons takes place
quite rapidly.  This is important because of the  toxicity of some of the
volatile fractions which are lost soon.  Simulated  weathering  has shown  that a
crude oil on water loses essentially all components boiling below Cj2 within
24 hours in the laboratory.3  It has been observed  that the toxicity of  a
crude oil mixed with water was greatly reduced  even by short weathering.1*

     Solution.  Evaporation and solution both act most strongly on  the low-
molecular weight hydrocarbons.  The true solubility of hydrocarbons drops
exponentially as a function of their molecular  volume.5  Consequently, although
toluene is soluble in water to 515 ppm,  anthracene  is  soluble  to  less than
0.1 ppm.  The higher polynuclear aromatics are  considerably less  soluble.

     Oxidation.  This is a selective process and  can be subdivided  into  chem-
ical oxidation and photooxidation.  Absorption  of photons by the  heavier
aromatics with suitable spectral properties  will  initiate the  reaction chain
leading to conversion of these materials.  Products of the reaction will
contain oxygen and-could consist of phenols, carboxylic acids, alcohols  and
the like.  Since these are more water  soluble than  the hydrocarbons, they will
be more completely lost to the water column  leading to greater dispersion.

     Microbial degradation.  This subject has been  studied over many years by
many competent investigators.  They conclude7'8»9»10 that microorganisms which
oxidize various hydrocarbons are widely distributed in soil and water, espe-
cially in estuaries and shorelines.  Although normal paraffins are most
susceptible, virtually all hydrocarbons are  degraded including aromatic
hydrocarbons of condensed structures.   These bacteria, which use  hydrocarbons
for an energy source, are effective in converting hydrocarbons to carbon
dioxide and water.  Probably one of the major ways  in which hydrocarbons are
removed in estuaries and coastal areas is by microbiological degradation.  For
example, Zobell7 reported that in coastal areas bacteria  can oxidize from 0.02
to 2 grams of hydrocarbon per square meter per  day  depending on several  factors


                                     167

-------
including rate of oxygen diffusion.   In shallow waters this approaches conver-
sion rates on the order of a part  per million of hydrocarbon per day.  It is
important to realize  that bacteria are active even in very cold arctic
water.1 1 >12

     Biochemical uptake metabolisms and discharge of hydrocarbons.  It has
been shown that marine organisms,  maintained in water containing oil, take up
hydrocarbons in their tissues.   Following a spill of No.  2 fuel oil near West
Falmouth, Massachusetts in September,  1969,  Blumer13'llf»15 analyzed oysters,
scallops, and other marine organisms and found that they  had taken up oil
fractions.  He kept three oysters  in flowing sea water in his laboratory. One
oyster was analyzed after it had been kept in flowing sea water for 72 days
and the other two after 180 days.   His publication in 1970ltf stated that none
of these three oysters had purged  themselves of the oil they contained prior
to the beginning of the experiment.   He concluded "Thus,  once contaminated,
shellfish cannot cleanse themselves of oil pollution".14

     A number of scientists have reinvestigated this matter and have shown
that marine animals do indeed  take up hydrocarbons but actually do cleanse
themselves.  Experiments have  been done with, literally,  hundreds of aquatic
animals.  Data now in hand clearly show that aquatic animals do take up hydro-
carbons but that, when the animals are placed in clean water, the hydrocarbons
are purged or metabolized.  For example, Lee and his co-workers16'17 described
the uptake, metabolism and discharge of radio-labeled aromatic hydrocarbons by
mussels and by fish.  They found that these compounds did indeed find their
way from sea water into the aquatic animal tissues, but that when the animals
were placed in clean  water, the hydrocarbons were lost.  The mussels purged
the hydrocarbons unchanged whereas the fish metabolized these products.  In
another study, Anderson18 carried  out exposure tests on oysters and clams,
illustrated in Figure III taken from his work.  In these  tests, oysters were
initially exposed to  a highly  aromatic fuel oil.  The animals were then placed
in clean sea water.   Specific  analyses for individual hydrocarbons were made
and purging was found to the level of analytical sensitivity.  Very similar
data have now been found in studies at other locations in the United States
including Battelle-Northwest in the State of Washington19 and more recent data
from Woods Hole20 using, in each case, local animals.  In every case examined,
purging was found to  occur.

     The National Academy of Science Report21 states: "organisms such as
mussels and oysters have been  shown to eliminate most absorbed petroleum
hydrocarbons when placed in clean  water".

     Accumulation in  the food  web.   The additional question of accumulation of
petroleum hydrocarbons in the  food chain should be considered.  Lee's
results16*17 show a rapid metabolism of certain aromatic  hydrocarbons in
fish.  Data from Anderson and  others show relatively rapid purging of hydro-
carbons from animals  placed in a clean environment.  Once the animal is no
longer subjected to water contaminated with oil, the affected organism cleanses
itself quickly of whatever oil contamination that it may have incurred.
Therefore, it is not  likely that such contamination would become concentrated
by transfer from one  trophic level to the next through the food chain.  In
fact, the National Academy Report21 states "There is no evidence for food-web
magnification of petroleum hydrocarbons in marine organisms."

                                       168

-------
    Significance of  Field Studies.  Field studies have been conducted princi-
pally in regions where oil and gas production has taken place.^5,^6 xhese have
all shown that neither oil spills nor continued production have detrimental
effects on the environment.   Grossling of the U. S. Geological Survey reported4**
that, even in areas of extremely massive oil spills of some years ago, there
are "no Death Valleys".   Natural forces which degrade oil have taken over and
converted the spilled oil into the end products of nature, carbon dioxide and
water.  Reviews of current producing areas have shown no detrimental
effects.1*5'46  Detailed chemical and biological surveys of producing area and
a control area in  the Gulf of Mexico shows that production of  petroleum had no
discernable  effect on the environment.1*5  The underwater surveys1*6 of Platforms
"Hilda" and  "Hazel"  in the Santa Barbara Channel have demonstrated much
increased marine biota in and around the platforms.  Studies of the petroleum
industry in  the Delaware Estuary24'1*7 have shown that the city of Philadelphia
and its environs contribute much more "oil and grease" than does the refining
industry in  the area.  Contributions of "oil and grease" from  the metropolitan
area are larger from drains and storm-water runoff than from the refining
sources.

POLYNUCLEAR  AROMATICS HYDROCARBONS  (PAH)

     Because of  the  concern for the presence of some carcinogenic polynuclear
aromatic hydrocarbons (PAH) in the aquatic environmental this  subject is
worthy of  separate discussion.  It must be pointed out that only a limited
number of  PAHs are carcinogenic.  Those that are carcinogenic  will contain 4,
5 or 6 rings.  Only  certain isomers are carcinogenic; for instance benzo(a)-
pyrene is  carcinogenic but not its isomers.1*2  Considerable data are now
available  from a  number of sources21'23 which show that polynuclear aromatics
are widely distributed in soils at very low concentrations and may have
occurred on  the earth's surface during geologic time.  Presumably a source of
PAHs has been from combustion such as forest fires or in more  recent times,
from burning of  coal.

     Microbial degradation.  Biological degradation of polynuclear hydrocarbons
has been discussed by Professor Gibson at the Symposium on Sources, Effects &
Sinks of Hydrocarbons in the Aquatic Environment.2  He pointed out that, since
aromatic hydrocarbons have been ubiquitously distributed throughout the
environment  over  geologic time, living organisms have evolved  enzyme systems
which oxidize these  compounds. Table I, taken from his publication25 shows the
types of aromatic  hydrocarbons known to be susceptible to microbial oxidation.
One common feature of the mechanisms used by bacteria to degrade these types
of compounds is  the introduction of hydroxyl groups.  Generally, two hydroxyl
groups are introduced and the diol so produced is then subjected to further
degradation.  A metabolic pathway  for the complete degradation of naphthalene
is indicated in  Figure IV. The compounds shown in brackets have been completely
identified.   The  end products of this mechanism are  the non-toxic products,
carbon dioxide and water.

     Less  is known about the detailed chemistry of microbial degradation of
PAHs that  contain  three or more rings.  However, evidence  is being obtained
that bacteria are  capable of oxidizing aromatic hydrocarbons that range  in
size from  benzene  to benzo(a)pyrene.  A mutant strain of bacteria,  faexj&Unefex.a

                                      169

-------
B836 has been shown to oxidize biphenyl,26 phenanthrene,27 and anthracene28
to dihydroxydiols.  Benzo(a)pyrene was oxidized  by  this  bacteria to dihydroxy
product as shown in Figure V, as well as  to  other products resulting from
further degradation.29

     Biochemical uptake and discharge.  Because  of  the  specific concern for
the PAHs, investigators have repeated uptake and purging studies with indi-
vidual PAHs and find results identical to that found with  other hydrocarbons.
That is, aquatic animals will take up PAHs in their gut  and their tissues, and
that these are purged or metabolized when the animals are  placed in clean
water.  Recent work of this nature has been  conducted by Neff39 and Neff  and
Anderson.40  The conclusions of these studies were  that  the pattern of uptake
and depuration of benzo(a)pyrene in the clam is  much the same  as observed
earlier from the naphthalenes.  Benzo(a)pyrene appears  to  be accumulated  more
rapidly, however, and released more slowly than  the naphthalenes.   In their
experiments purging was a consistent process, declining  to a level of only 1.4
percent of the accumulated material in 20 days.  Benzo(a)pyrene could not be
detected in clams maintained in isotope-free sea water  for 58  days (limit of
detection 0.01 ppm).  It is interesting that the viscera contained most of the
activity at all sampling periods.

     Photochemical oxidation.  Since the  polynuclear aromatic  hydrocarbons
have high absorptivities in the ultraviolet,  it  is  likely  that photochemical
oxidation plays an appreciable role in their degradation.   In  fact a review
states that photooxidation is probably one of the most  important processes in
removal of polycyclic hydrocarbons from the  atmosphere. **2

ORGANICS

     Many of the same factors which operate  to dispose or  degrade hydrocarbons
are also effective on many organics, depending of course on properties of the
individual compounds.  Such factors as volatilization,  solution,  sedimentation
and oxidation are likely candidates for removal  or  transformation and should
be studied.  It is well known that ethers are easily oxidized  and it is likely
that this is the mechanism of their loss  in  a natural system.  Oxidation is
also possibly the route for the loss of other relatively reactive organic
compounds.  Because of the reactivity of  organic functional groups other
reactions such as hydrolysis, reduction and  dechlorination may be important.48

     Photochemical.  Photochemical energy reaches us from  the  sun in the  form
of photons with wave lengths covering the spectra from  infrared to the far
ultraviolet and including the visible.  In fact, all the solar energy we
receive is in the cumulative energy of the photons  which fall  upon the earth.
Photosynthesis is our most important photochemical  process, without which
there could be no life on earth since plant  growth  depends upon the photo-
chemical conversion of atmospheric carbon dioxide into organic substances.
The significance of photochemical processes  is   becoming more  and more apparent
in the environmental chemistry of organic materials.  The  photochemical
oxidation and dechlorination of a number  of  the  highly  chlorinated pesticides
and similar compounds has been well established.  A particularly significant
recent paper appeared in "Science" last March indicating that  the toxic,
complex substance TCDD (2,3,7,8 tetrachlorodibenzo-p-dioxin) is photochemically


                                     170

-------
degraded by sunlight6  in the natural environment.  Presumably the degradation
reactions involve  dechlorination and perhaps oxidation.  Photo sensitived
oxidations may also  play a significant role in  the observed degradation of DDT
in the environment30 as well as with other pesticides.l+8

     Microbial degradation.  The microbiological metabolism of organics has
been the subject of  detailed study for a number of years and has been exten-
sively reported.31'32   Investigations similar to those outlined above for
polynuclear aromatics  have been reported on phenols  (Dagley, University of
Minnesota),32 on surface-active agents (Huddleston and Allred, Continental Oil
Co.),32 on insecticides (Matsumura and Boush, University of Wisconsin)32 and
on other classes of  commercial organic compounds that  could be present in
waste waters.  A number of bacteria widely distributed in nature, have the
ability to degrade phenolic compounds.  The ability  to degrade phenolics is
not confined  to bacteria.  These reactions take place  also with fungi and in
species of (UpQAg^Wu, pWlicJJUIMm and OO&poHa,  The  end products of these
conversions are simple, essentially non-toxic organic  structures.

     Organic  halogen compounds have been shown  to undergo a number of micro-
biological reactions which result in loss of halogen and its replacement
either  by hydroxyl or hydrogen.  Microbiological hydrolytic dehalogenation is
the most common reaction.  These reactions seem quite  general in natural
soils.   Some  examples which have been documented119 include lindane which
undergoes microbial decomposition under anaerobic conditions to release
chloride  ion  and  the brominated derivatives of  ethane, propane and butane
which produce either the simple hydrocarbons or alcohols upon loss of halogen.
Aliphatic  halides  such as allyl chloride and 1,3-dichloropropene hydrolyze
readily in  the soil.

     Edwards50 points out that insecticides in  soils are largely broken down
by microorganisms.  Some of molds indicated above to be able to utilize
phenolics  are also effective in converting chlordane and heptachlor  into
hydrophilic  degradation products.  Several soil bacterial species have been
found able to dechlorinate lindane.

     Biochemical  uptake and metabolism.  It is  well  known that aquatic species
concentrate  organochlorine insecticides33 but is is  perhaps not as well
recognized  that there is also an appreciable amount  of degradation by the
living species.   Johnson, Saunders, Sanders and Campbell33 used radiochemicall}
labeled aldrin and DDT in studies with freshwater invertibrates and  found, in
3-day exposures,  that some degree of degradation of  aldrin and DDT occurred
with all  organisms examined in the limited time of the tests.

     It is  interesting that natural products containing bound halogen are
produced  by  aquatic biota.  Dr. Faulkner of Scripps  presented a paper at the
Symposium on  Sources,  Effects & Sinks of Hydrocarbons  in the Aquatic
Environment2  which described a group of halogen-containing terpenes  biogeni-
cally synthesized  by aquatic organisms.

     Significance  of field observations.  There are  many natural reactions
Which degrade organic compounds, including organic pesticides, which have now
been documented  from field and laboratory observations. As a result of these


                                      171

-------
reactions it has been stated that  "there  are no  permanent organic pesticides".1*8
Degradation apparently continues onto  the non-toxic inorganic products with
rates depending upon many  factors,  such as climate, temperatures, soil type
and so on.

METALS

     To put a perspective  on the effects  of metal  ions in the aquatic environ-
ment it is useful  to turn  to an important concept  introduced by Dr.  Ketchum of
Woods Hole.  This  is the Relative  Critical Index as described in Table II,
This Table lists toxic elements considered to be of critical importance to
aquatic pollution.  The "toxicities"  (concentrations in pg/1) assigned to the
metals in the table are in a decreasing order of toxicity,  and each "toxicity"
concentration quoted is considerably smaller than  the concentration acutely
toxic to man or the aquatic environment.   These  toxicity values are obtained
from "Water Quality Criteria".41

     The concept of Relative Critical  Index in Table II is  derived by Ketchum3"*
by dividing the annual amount  of a substance mobilized by human or natural
activities by the  assumed  "toxicity".  This index  helps to  identify elements
as pollutants and  assists  in establishing the relative contributions of trace
elements from nature and from  man-made sources.  For instance, it can be seen
that the greatest  input of most trace  metals listed is from natural sources.
Where the natural  contribution is  a large portion  of the potential supply,
control of individual sources  whose concentration  levels are low would have
little effect on water quality.

     Chelation.  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 most natural waters much of the free
metals ions would  probably be  bound to organic substances naturally present in
the water.  There  is 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.35'36  This could  be due either to the fact that the organo-
metallic complex is to bulky to enter  a biological system or it could be due
to the lack of availability of the metal  ion for reaction with enzymes within
the cells.

     Reduction.  Methylation of mercury in the aquatic environment can be
caused by the reductive action of  anaerobic bacteria at the bottom-water
interface.51  These bacteria produce the  highly  toxic dimethylmercury (CHs^Hg
and methylmercury  CH3Hg+.  It  was  this type of mercury pollution which was  the
cause of the notorious "Minamata disease" in Japan in 1953.52

     Sedimentation.  Detailed  metal analysis of  sediments and benthic marine
populations have been conducted near sewage outfalls in the Pacific Ocean off
the Southern California Coast.37   No significant differences were observed  for
sole caught in polluted and unpolluted areas for the following trace elements,
mercury, cadmium,  copper,  zinc, iron and  cobalt  in the sediments.  In the
reported discussion which  followed the above presentation37 it was stated that
similar results were found in  England.  The suggestion was made that sediments
may be a useful final sink for many metals and that the adsorption of the


                                      172

-------
metal  ions  to the surface of  sediments precludes their uptake by biological
systems.  I would suggest that  an important research  study would be  that of
evaluating  the biological availability or nonavailability of sediment-associated
metals in the aquatic environment.

     Essential nutrients.  Almost any element or substance can be  toxic to
plants when present in abnormally high concentrations.  Yet many toxic elements
have been shown to be essential nutrients at low concentrations, at  least to
certain plant species.38'43   Drake of the University  of Massachusetts lists1*3
as essential micronutrients  the following: iron, manganese, copper,  zinc,
boron, molybdenum and chlorine. It has been recognized especially  by orchard
growers that essential elements must be introduced when soils are  deficient.
Consequently "zinc nails" are sometimes driven into trees when soils are
deficient.   Enzyme system in plants seem to require manganese, iron, and
molybdenum  as well as magnesium in trace amounts.  There is apparently some
evidence that selenium may be a necessary trace nutrient to some plants.38
     The assistance of Dr.  Paul Porter of the Shell Biological  Sciences
 Research Center, Modesto,  California is gratefully acknowledged in the
 preparation of several sections of this paper.
 BIOGRAPHY

        Fred T. Weiss is Senior Staff Research
 Chemist, Analytical  Department, Shell Develop-
 ment Company, Houston, Texas.  He holds
 degrees in Chemistry from the 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 applica-
 tion of methods to trace analysis.  In recent
 years much of his work has been devoted to en-
 vironmental measurements.
                                       173

-------
                                 REFERENCES
1.   Garrett,  W.  D.  (1972),  "Impact of Natural and Man-made  Surface  Films on
     Properties of the Air-Sea Surface" in "The Changing Chemistry of  the
     Oceans,"  (D.  Dryssen and D. Jagner, Eds.)» Wiley Interscience,  pp.  75-91,
     New York.

2.   American  Institute of Biological Sciences (1976), "Sources,  Effects &
     Sinks of  Hydrocarbons in the Aqutic Environment", 1401  Wilson Boulevard,
     Arlington, Virginia 22209.

3.   Kreider,  R.  E.  (1971),  "Identification of Oil Leaks and Spills,"
     Proceedings of Joint Conference on Prevention and Control  of Oil  Spills,"
     pp. 119-124, Washington, D. C.

4.   Templeton, W. L. (1974), "Summary Report on Effects of  Oil Discharges,
     Domestic  and Industrial Wastewaters on the Fisheries of Lake Maracaibo,
     Venezuela," Battelle-Northwest Laboratory, Richland, Washington.

5.   McAuliffe, C. D. (1966), "Solubility in Water of Paraffin, Cycloparaffin,
     Olefin, Acetylene, Cyclo-Olefin and Aromatic Hydrocarbons,"  J.  Phys.
     Chem., 70, 1267.

 6.   Crosby, D. G. and Wong, A. S., (1977) "Environmental Degradation  of 2, 3,
     7, 8 Tetrachlorodibenzo-p-dioxin (TCDD)", Science, 195,  1337.

 7.   Zobell, C. E. (1969), "Microbial Modification of Crude  Oil in the Sea,"
     Proceedings of Joint Conference on Prevention and Control  of Oil  Spills
     pp. 317-326, New York.

 8.   Colwell,  R. R., Walker, J. D., and Nelson, J. D., Jr.,  (1973) "Microbial
     Ecology and the Problem of Petroleum Degradation in Chesapeake  Bay," in
     Microbial Degradation of Oil Pollutants, Louisiana State University,
     Publication. No. LSU-SF-73-01.

 9.   Miget, R. J., Oppenheimer, C. H., Kator, H. I., and LaRock,  P.  A. (1969),
     "Microbial Degradation of Normal Paraffin Hydrocarbons  in  Crude Oil,"
     Proceedings of Joint Conference on Prevention and Control  of Oil  Spills,
     pp. 327-331, New York.

10.   Walker, J. B., Colwell, R. R., and Petrakis, L.  (1975)  "A  Study of  the
     Biodegradation of Louisiana Crude Oil Employing  Computerized Mass Spectro-
     metry," Proceedings of Joint Conference on Prevention and  Control of Oil
     Pollution, San Francisco, California, pp. 601-605.

11.   Kinney, P. J., Button,  D. K., and Schell, D. M.  (1969), "Kinetics of
     Dissipation and Biodegradation of Crude Oil in Alaska's Cook Inlet,
     "Proceedings of Joint Conference on Prevention and Control of Oil Spills",
     pp. 333-340, New York.
                                     174

-------
12.   Robertson,  B.,  Arhelger, S., Kinney,  P.  J.,  and Button  D  K  (1973)
     "Hydrocarbon Biodegradation in Alaska Waters," In microbial Degradation
     of  Oil  Pollutions, Louisiana State University, NO.  LSU-SG-73-01, pp.
     171—184.

13.   Blumer, M., Souza, G., Sass, J.  (1970),  "Hydrocarbon Pollution of Edible-
     Shellfish by an Oil Spill," Marine Biology.  _5, pp.  195-202.

14.   Blumer, M., Sass, J., Souza, G.,  Sanders,  H. L., Grassle,  J.  F. and
     Hampson,  G. R.,  (1970), "The West Falmouth Oil Spill," Woods Hole
     Oceanographic Institution, Woods  Hole, Massachusetts.

15.   Blumer, M., and Sass, J. (1972),  "West Falmouth Oil Spill,  Data Available
     in November 1971, II, Chemistry," WHOI 72-19,  Woods Hole Oceanographic
     Institution, Woods Hole, Massachusetts.

16.   Lee,  R.  F., Sauerheber, R., and Benson,  A.  A.  (1972), "Petroleum Hydro-
     carbons:   Uptake and Discharge by Marine Mussel Mytilus edulis," Science.
     177.  pp.  344-346.

17.   Lee,  R.  F., Sauerheber, R., and Dobbs, G.  H. (1972), "Uptake, Metabolism
     and Discharge of Polycyclic Aromatic  Hydrocarbons by Marine Fish," Marine
     Biology,  177. pp. 344-346.

18.   Anderson, J. W., Neff, J.  M., Cox, B. A.,  Tatem, H. E., and Hightower,
     G. M.,  (1974),  "The Effects of Oil on Estuarine Animals:  Toxicity,
     Uptake and Depurations, Respiration," in symposium:  Pollution and the
     Physiological Ecology and  Estuarine and  Coastal Water Organisms, sponsored
     by the Environmental Protection Agency and the Belle W. Baruch Coastal
     Research Institute of the  University  of  South Carolina, November 14-17,
     1973.

19.   Vaughan,  B. E.  (1973), "Effects of Oil and Chemically Dispersed Oil on
     Selected Marine Biota - Laboratory Study," Battelle-Northwest Laboratories,
     API Publication, No. 4191.

20.   Teal, J.  M., and Stegeman, J. J.  (1973), "Accumulation, Release and
     Retention of Petroleum Hydrocarbons by the Oyster,  CtlO&AOA&Lea (AtAg.6u.ca,"
     Marine Biology, 22, pp. 37-44.

21.   National Academy of Sciences  (1975),  Ocean Affairs Board Workshop,
     "Petroleum in the Marine Environment" Washington, D. C.

22.   Pancirov, R. J. and Brown, R. A.  (1975), "Analytical Methods for Poly-
     nuclear Aromatic Hydrocarbons in  Crude Oils, Heating Oils and Marine
     Tissues," Proceedings of Joint Conference  on Prevention and Control of
     Oil Pollution,  San Francisco, pp. 103-113.

23.   Blumer, M.  and  Youngblood, W. W.  (1975), "Polycyclic Aromatic Hydrocarbons
     in Soils  and Recent Sediments," Science, pp. 53-55.  See also Blumer,
     Science 134, 474 (1961).


                                     175

-------
24.  "Petroleum Industry  in  the Delaware Estuary" (1977) National  Science
     Foundation - RANN Program Grant Env.  74-14810 A03.

25.  Gibson, D. T.  (1976), "Microbial Degradation of Carcinogenic  Hydrocarbons
     and Related Compounds," presented at "Symposium on Sources, Effects and
     Sinks of Hydrocarbons in the Aquatic Environment", American University
     Washington, D. C., August 9-11, pages 224-238 (See Reference  2).

26.  Gibson, D. T., Roberts,  R. L.,  Wells, M. C. and Kobal, V. M.  (1973),
     "Oxidation of  Biphenyl  by a B/ce/eAoicfe/ta Species."  Biochem.  Biophys.
     Res. Commun. 50, page 211.

27.  Selander, H.,  Yagi,  H.,  Jerina, D.  M., Wells, M.C., Davey, J. F.,
     Mahadevan, V., and Gibson, D. T. (1976), "Dihydrodiols from Anthracene
     and Phenanthrene," J. Am. Chem. Soc.

28.  Akhtar, M. N., Boyd, D.  R., Thompson, N. J., Gibson, D. T., Mahadevan, V.,
     and Jerina, D. M.  (1975), "Absolute Sterochemistry of the Dihydroanthracene-
     cci- and ;t/iaxi6-l,2-diols Produced from Anthracene by Mammals  and Bacteria,"
     J. Chem. Soc., 2506.

29.  Gibson, D. T., Mahadevan, V., Jerina, D. M., Yagi, H., and Yeh, H. J. C.
      (1975), "Oxidation of the Carcinogens Benzo(a)pyrene and Benzo(a)anthracene
     to Dihydrodiols  by a Bacterium," Science, 189, page 295.

30.  Ivie,  G. W., Casida, J.  E. (1971),  "Photosensitivers of the Accelerated
     Degradation of Chlorinated Cyclodienes and other Insecticide  Chemicals
     Exposed to Sunlight  on  Bean Leaves",  J. Agr. Food Chem., 19,  410-416.

31.  Goring, C. A.  I. and Hamaker, J. (Editors) (1972) "Organic Chemicals in
     the Soil" Volumes  I  & II, Marcer Dekker, New York.

32.  McLaren, A. D. and Skujins, J., (Editors) (1971) "Soil Biochemistry"
     Marcer Dekker, New York.

33.  Johnson, B. T.,  Saunders, C. R., Saunders, H. 0., and Campbell, R. S.
      (1971) "Biological Magnification and Degradation of DDT and Aldrin by
     Freshwater Invertebrates," J. Fish, Res. Bd. Can. 28: 705-709.

34.  Ketchum, B. H.  (1973),  "Symposium on Ocean Pollution," Statement made
     before Senate  Commerce  Committee, Subcommittee on Oceans and  Atmosphere,
     June 12.

35.  Lerman, A. and Childs,  C. W. (1973),  "Metal-Organic Complexes in Natural
     Waters:  Control of  Distribution by Thermodynamic, Kinetic and Physical
     Factors,"  In  "Trace Metals and Metal-Organic Interactions in Natural
     Waters."  Edited by  Singer, P.  C.,  Ann Arbor Publ., Inc., Ann Arbor,
     Michigan.
                                     176

-------
36.  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.

37.  DeGoeij, J. J. M., Guinn, V. P., Young,  D.  R. and Mearns,  A.  J.  (1974)
     "Neutron Activation Analysis Trace Element  Studies  of  Diver  Sole Liver
    and  Marine Sediments," IAEA-SM-175/15, pages 189-200  (International
   • Atomic Energy Agency, Vienna).

38.   "Soil-The Yearbook of Agriculture"  (1957) - U.S. Department  of Agricul-
     ture, pages 165-171.

39.   Neff, J. M. (1975), "Accumulation and  Release of Petroleum - Derived
     Aromatic Hydrocarbons by Marine Animals," in Symposium on  Chemistry,
     Occurrence and Measurement of  Polynuclear Aromatic  Hydrocarbons, presented
     at Division of Petroleum Chemistry, American Chemical  Society, Chicago,
     August 24-29.

40.  Neff, J. M. and Anderson, J. W.  (1975),  "Accumulation, Release and
     Distribution of Benzo(a)pyrene C~llt in The  Clam Rang/to. Cunea&l," Joint
     Conference on Prevention and Control of  Oil Pollution, San Francisco,
     pages 469-471.

41.  National Academy of Sciences  (1974), "Water Quality Criteria," U.S.
     Government Printing Office, Washington,  D.  C.

42.  National Academy of Sciences  (1972), "Particulate  Polycyclic Organic
     Matter," Washington, D. C.

43.  Drake, M.  (1968) "Soil  Chemistry and Plant  Nutrition"  in "Chemistry of
     the  Soil"  Second Edition, Editor F. E. Bear, Reinhold  Publishing Company,
     New  York,  pages 398-400.

44.  Reference  2, page  36.

45.  El-Sayed,  S. Z.  (1974), "Effects of Oil  Production on The  Ecology of
     Phytoplankton off  the Louisiana  Coast,"  Project OV-66-JHM, Gulf
     Universities Research Consortium -  Offshore Ecology Investigations.

46.  Bascom, W., Mearns, A.  J. and  Moore, M.  D.  (1976)  "A Biological  Survey
     of Oil Platforms in the Santa  Barbara  Channel".  J. Petroleum Technology,
     November 1976, pages 1280-1284.

47.  Hunter, J. V., Yu, S. L. and Whipple,  W.,  Jr. (1976),  "Measurement of
     Urban Runoff Petroleum  in "Urbanization  and Water  Quality  Control",
     Published  by American Water Resources  Associates,  Minneapolis, Minnesota.

48.  Crosby, D. G.  (1973) "The Fate of Pesticides in the Environment" in
     Annual Reviews of  Plant Physiology  24, 467-492.
                                     177

-------
49.  Meikle, R. W.  (1972) "Decomposition:  Qualitative Reactions", in Reference
     31, pages 203-210.

50.  Edwards, C. A.  (1972) "Insecticides",  in Reference 31, pages 531-537.

51.  Wood, J. M., Kennedy, F. S. and Rosen,  C.  G.  (1968),  "Synthesis of Methyl-
     mercury Compounds on Extracts of a Methanogenic Bacterium", Nature, 220,
     pages 173-174.

52.  Dugan, P. R.,  (1972) "Biochemical Ecology of  Water Pollution",  Plenum
     Press, New York.
                                    178

-------
                                         Figure I

                SCHEMATIC REVIEW OF PRINCIPAL FATES, EFFECTS AND TRANSPORT
                    MECHANISMS OF POLLUTANTS IN THE AQUATIC ENVIRONMENT
                Hydrocarbons
                                    Organics
                                        Metals
Fates
Volatilization
Solution
Sedimentation
Oxidation
Volatilization
Solution
Sedimentation
Oxidation
Photochemical Dechlorination
Chelation
Oxidation/Reduction
Sedimentation
Transport
Mechanisms
Air-Water Interface
Sediment Transport
Air-Water Interface
Sediment Transport
Sediment
  Transport
Effects
Oxygen Consumption
Biological
  Assimilation
Toxicity
Oxygen Consumption
Biological
  Assimilation
Toxicity
Biological
  Assimilation
Essential Trace
  Elements
Toxicity

-------
      BURSTING
      BUBBLES
                                                 ATMOSPHERIC
                                                   OXIDATION
                                       RAIN  AND
                                        FALLOUT
                   SEA   SURFACE
                     OIL  SLICK
          V/ATER-1 N -01L    EMULSION
          1 I  .ss
      OIL-IN-WATER  //
        EMULSION  If

               DISSOLUTION
    \\
BUBBLED
TRANSPORT
                                          Iff
                                                         SPREADING
          CONVECTION
             AND
          UPWELLING
NONBUOYANT
OIL RESIDUES
                  MIXING AND
                   SINKING
                    CIRCULATIONS
                   ADSORPTION
                  ON NONBUOYANT
                  PARTICLES
         BUOYANT
         MATERIAL
BIOLOGICAL
ASSIMILATION
                  BOTTOM    SEDIMENT
Fig. II - Natural Force* Which Diapers* and Modify Oil Slick* on Watar Aft«r
Garrett,1                     180

-------
THE ACCUMULATION AND DISCHAPCTrv*
        NO.  2 FUEL OIL COMPONENTS
           BY OYSTERS AND CLAMS
                10       15      20
               TIME OF EXPERIMENT - DAYS
                        Figure III
                              C02+H20
                                 I
                                 t
                                  COOH
                               rr°
                               CH2 COOH
                                  COOH
                                 CHO
                       COOH
     Figure IV Metabolic Pathway for the Degradation of Naphthalene
           by Certain ?*e.udomon0U> Species. From Gibson
                     181

-------
9- Hydro* ybenjo [o] pyrene

         NONENZYMATIC
                                     T
                                                   02
                                            Beijerinckio wild lype
                                 ACID
                                 PRODUCTS
                           C/j-9,10- Dirtydro*y-9.IO-
                            dihydrobenzo [o]pyrene
Figure V   Oxidation of Benzo (a)pyrene by B&ij &tu.nc.ksia B836.
           From Gibson25
                                Table I

     AROMATIC HYDROCARBONS KNOWN TO BE OXIDIZED BY MICROORGANISMS
                                     a)
            NONOCYCLIC

    Benzene
    Toluene
    Xylenes
    Tri and tetramethylbenzenes
    Alkylbenzenes
    Cycloalkylbenzenes
             DICYCLIC

    Naphthalene
    Methylnaphthalenes  (mono and  di)
    Ethylnaphthalenes
                     POLYCYCLIC

                 Pyrene
                 Benzo(a)pyrene
                 Benzo(a)anthracene
                 Dibenzo(a)anthracene
                 Benzperylene
                 Perylene
                     TRICYCLIC

                 Phenanthrene
                 Anthracene
 a)From Gibson25
                                  182

-------
                                                   Table II
                                  METAL  TOXICITY AND RELATIONSHIP TO INPUT
                                                                           34



ELEMENT
MERCURY
CADMIUM
SILVER
NICKEL
SELENIUM
LEAD
COPPER
CHROMIUM
ARSENIC
ZINC
MANGANESE


TOXICITY
yg/i
0.1
0.2
1
2
5
10
10
10
10
20
20



RATE OF MOBILIZATION (109 G/YR)
A (MAN)

1.6
0.350
0.07
3.7
0.45
3.6
2.1
1.5
0.7
7
7.0
B (NATURAL)

2.5
2.65
11
160
7.2
110
250
200
72
720
250
C
TOTAL
4.1
3.0
11.1
164
7.7
113.6
252.1
201.5
72.7
727
257
RELATIVE CRITICAL INDEX,*
1012 1/YR
MAN

16,000
1,750
70
1,350
90
360
210
150
70
330
350
NATURE

25,000
13,250
11,000
80,000
1,440
11,000
25,000
20,000
7,200
36,000
12,500
00
           *RELATIVE CRITICAL INDEX =
 INPUT
TOXICITY

-------
                          WEISS PAPER DISCUSSION
Paul Mikolaj, Lion Oil Co^;  Do you have any  Idea of  how the information you
have presented will be used in the setting of the standards for the priority
of pollutants?

Weiss;  I would hope that the answer is yes.   If  anybody cares to answer that
from the floor I would like to hear an answer.

Ridgeway Hall, EPA:  I think without a doubt  if information of that type is
timely submitted to EPA in the course of the  rule-making proceedings,  we
certainly will consider it.

Ridgeway Hall, EPA;  On your chart of Table II which  you last had up there,
could you tell us what the significance of the right-hand column was and how
those numbers were derived and what they mean?

Weiss;  That is actually the relative critical index  itself.   These columns
come from taking the toxicity and dividing the toxicity into the rate of
mobilization to develop a ratio.  What it means is that the ratio indicates
the relative significance of the element to toxicity.

D. I. Mount, EPA, Duluth;  In regard to Table II  that  you have on the board
I think it is a useful table but I think it is also important to point out
that it makes no account of the form in which those metals are being trans-
ported or converted by man and by nature.  For example, in the case of mercury
most of the natural mercury is transported in the mineral form or in sulfide
forms that are insoluble.  In the case of much of the  man-transported mercury,
it is in vapor form in power plant stacks so  that the  biological significance
in a given amount of that metal will make it  quite different, even the total
quantities show a different picture.

George J. Putnicki, UTD;  Are there any taste and odor studies conducted
concurrently with the concentrations of the number of  two fuel oil?

Weiss;  We did not conduct any taste or odor  studies.   There are independent
studies which have been done on taste and they show depuration.  The oyster
farmers in Louisiana find it is about the same order  of time, three or four
weeks.  When they observe contaminated oysters, they  put them into a clean
bed, so the timing is the same but they were  not  done concurrently.
                                     184

-------
              MEASURING AQUATIC IMPACT OF TOXIC CONTAMINANTS

                         Donald I. Mount, Ph.D.
                    U.  S.  Environmental Protection Agency
                 Environmental Research Laboratory-Duluth
     The  impact  of  toxic pollutants on aquatic systems is often a
principal concern about the manufacture and use of chemicals, that may
be discharged  or otherwise released into the environment.  Our system of
streams and  lakes seems to function in much the same way as the lymphatic
system in our  bodies—collecting what "seeps through" as a result of our
industrialized society that liberally employs synthetic chemicals in its
day-to-day operations. Whether toxic chemicals are put in landfills,
discharged in  liquid effluents, incineratedj or lost to the air through
vaporization,  we should not be surprised to find them in our rivers and
lakes. If toxic and persistent enough, we can expect them to cause
problems, either from direct effects on the environment or through
residues  in  organisms.  We have also learned, that pollutants may be
drastically  altered in their chemical form or biological behavior once
released  into  the complex environment of streams and lakes.

     The  local area of release is not the only one of concern.  We may
well find effects of pollutants occurring far downstream or even in our
coastal waters without any discernible effect in the immediate area of
discharge.

     The  responsibility of these chemicals after discharge, no matter
how geographically  remote or how long after the release has been made,
must become  a  way of life and a part of doing business.  The alternative
under which  we have been living in the past places the burden of remedial
measures  on  those who did not cause the problem.  It is encouraging to
see that  many  companies are as concerned about the consequences of the
chemicals in their  wastes as are the regulatory agencies, and I am sure
that this concern will lead to a lessening of the crises that we have
faced in  the past few years.  Undoubtedly, with modern analytical methods
and our considerably improved knowledge of aquatic systems, the apparent
increase  in  problem chemicals may well be one of better identification
rather than  an increase in problems.

     For  all of  these reasons, and perhaps for many others, there is a
mounting  concern and effort to develop more rapid, reliable and cheaper
methods for  predicting the impact of toxic chemicals on the aquatic
environment  and  before they become problems.  The passage of the Toxic
Substances Control  Act undoubtedly was hastened by these same problems


                                   185

-------
and concerns.  Since the emphasis in that act  is  one  of  properly testing
chemicals before they are used and before they become a  problem, the
need for rapid, predictive tests has been even further strengthened.  If
indeed that is the thrust of the regulations which will  be developed,
then it is obvious that most of the predictive toxicology work will have
to be done in the laboratory and field studies will be relegated by and
large to an assessment of the accuracy of decisions made during the pre-
market testing period.

     The testing of single chemicals under  the Toxic  Substances Control
Act and the assessment of toxicity from petroleum refinery wastewater
have one thing in common.  The variety of pollutants  and mixtures in
both is so large that what can be done on any  one chemical or any one
waste will have to be quick and relatively  inexpensive.

      Petroleum refinery wastewater has posed  difficult  problems to
those trying to develop acceptable waste treatment systems to adequately
remove the toxic chemicals contained in it. These wastes contain chemicals
that are water insoluble but fat soluble, and  these are  the very ones
which most often cause residue problems in  organisms  and are often of
the highest toxicity.  Unfortunately, one cannot  find "reagent grade"
petroleum refinery wastewater on which to do his  experiments and make
predictions, so the problems associated with these wastes are compounded
as compared to the ones associated with pure,  single  chemicals.

      At this point, it would be well to discuss  the  significance of
some specific toxic effects which have received much  attention in the
past few years.  I refer particularly to carcinogenicity,  as well as
teratogenicity and mutagenicity.  There are two problem  areas in which
these effects should be considered in assessing aquatic  impact.   One
area is the induction of any one of these toxic effects  in aquatic
populations with resultant population effects.  In such  cases,  we must
recognize that these effects are no more significant  than many other
effects such as reduction in growth rate, mortality,  or  reduced reproductive
rates, to the populations of concern.  It makes no difference to society
whether aquatic organisms are killed by malignant tumors or by avoidance
to a particular material in a water body.   In  either  case, it is a
decimating factor on the population, but no particular importance is
attached to the effect because it is due to cancer.   On  the contrary
more emotional importance is attached to human suffering from cancer
than to suffering from an automobile accident.  Therefore, carcinogenic,
mutagenic, and teratogenic properties have  no  special significance for
aquatic organisms.  If, however, aquatic organisms accumulate chemicals
with these properties and thereby increase  the exposure  of human populations
to such chemicals, then we must have special concern.

     No useful purpose would be served by listing all of the tests that
are now available for assessing impact on aquatic systems in a paper
such as this one.   Great progress has been made  in the  last fifteen
years in developing more refined and sophisticated tests to measure the
toxicity of chemicals to aquatic systems.   Indeed, the state of the art
as it is practiced in aquatic laboratories  probably approaches the


                                    186

-------
quality  of  that used in mammalian toxicology  laboratories  and  standard
protocols for conducting aquatic toxicity  tests  are easy to  find and
comparable  to those for mammals.  The problems in mammalian  toxicology
and aquatic toxicology are all generically the same and revolve around
appropriate test organisms for prediction  of  effects,  quality  of test
animals, length of exposure, effects measured, and laboratory  quality
control. One who needs to test aquatic  impact now has available to him
a selection of test organisms  (U.S.EPA 1975)  which have been successfully
employed in laboratory testing.  There are at least fifteen  or twenty
aquatic  organisms ranging from fish to protozoans and  algae  that have
been successfully cultured in laboratories and are adaptable to test
conditions  in aquatic testing systems.   While some of  these  species may
not be particularly important in themselves in aquatic systems, such as
for example, Daphnia magna, the data base  concerning the sensitivity of
such organisms to a variety of chemicals strongly suggests that they are
not overly  sensitive and that they can be  used as the  "white rat"  for
predicting  effects on other organisms.

     The data base on toxicity of chemicals to^ aquatic organisms is now
reaching proportions large enough so that  some reasonable  judgments can
be made in the selection of the itf&st appropriate test  organisms.   Especially
with single chemicals produced for particular purposes, often  the  objective
in producing the chemical leads one to the selection of the  proper
organism.  Obviously, if the chemical is produced for  use  as a herbicide,
it only makes good sense to test its effects  on  plants such  as algae or
macrophytes.  Likewise, chemicals designed to kill insects should  be
tested on aquatic arthropods and preferably aquatic insects.  This is
not to suggest that a variety of organisms should not  be tested where
possible, but since time and funds are nearly always limiting, the bulk
of the effort available can be expended  on what  are likely to  be the
more sensitive organisms.

     For a variety of aquatic  species, there  are now acute and chronic
test methods available which enable one  to measure effects ranging from
short term, LC50 measurements  to very sophisticated and sensitive  life
cycle tests  (U.S. EPA, 1973; U.S. EPA, 1976;  Woelke, 1972; APHA, 1975).
In  chronic tests, one can examine the effects of a toxicant on all life
stages of the organism and measure effects on reproduction and progeny
growth as well as effects such as malformations.  There are  now approximately
100 chronic tests with several fish species on perhaps 50  to 75 different
toxic materials. An examination of this  data  base reveals  that one can
be reasonably accurate for most chemicals  by  looking at segments of the
life history in toxicity tests and avoid the  expensive chronic tests for
full life cycles  (McKim, 1977 and Macek, 1977).   These authors point out
that if one measures the toxic effects on  the eggs and larvae  of fish,
he will find as the no-effect  concentration one  that is not  greatly
different from the one that would be found if the animals  were exposed
throughout their life history beginning  with  eggs and  ending with  growth
data on the F, progeny.  Other studies have revealed that  the  gill-
cleaning reflex commonly called the "cough response" is a  highly reliable
indicator of the concentrations which will or will not have  chronic
effects in a life history exposure for fish (Spoor, et al.,  1971;  Drummond,


                                    187

-------
et_ al., 1973 and Drummond, 1974).  Through the use  of  both models and in
a different approach by using resin columns  (Neely,  et^ al.,  1974; Chiou,
e_t al., 1977 and Veith, 1976) bioconcentration properties  of chemicals
can be rather accurately predicted from either very short  tests with
organisms or through a strictly chemical, analytical approach.  Indeed
sufficient correlations exist now to enable  us to be reasonably certain
that the single chemical property of the partition  coefficient  (between
octanol and water) is a reliable predictor of whether  or not one can
expect the chemical to bioaccumulate.

     Species of fish are now available that  make possible  chronic life
history tests in a matter of two months or less  (Smith, 1973) and tests
with Daphnia and some of the aquatic insects are now fully developed and
can be expected to produce reliable results  in an acceptable period  of
time.  The cost of doing tests such as described above is  certainly
acceptably competitive with other analyses that are now required on
chemicals and the insertion of biological tests into requirements for
evaluating chemicals is becoming a matter of routine.  Physiological  and
biochemical tests on fish in particular and  other tests on invertebrates,
have yet to find a promim^^^»lace in assessing the effects  of  toxicants
on aquatic organisms.  Perhaps this is^^»esult of  the unregulated body
chemistry which is so typical of most of the aquatic poikilotherms.

      It seems quite fair to say that the ability to measure  aquatic
toxicity and expected toxic effects on aquatic populations is substantially
more  advanced than is our ability to predict the behavior  of chemicals
in the environment.  The metabolic pathways, the transport of these
chemicals from area to area, and perhaps most important of all,  the
permanency of the apparent sinks such as the sediment  in lakes,  seem the
most  difficult judgments to make when assessing environmental impact.
Estimating what concentrations will occur as a result  of expected production
and usage rates and how the chemicals will get there,  where  they will go
and how long they will stay before they are  degraded into  something
else,  is not routine.

      This leads me to express a word of caution about  the  confusion
which  can exist regarding the distinction between toxicity and  hazard.
In our rush to assure that highly toxic chemicals are  not  released into
the environment, there is a danger that we will reject a chemical because
it has a high toxicity and a low hazard in favor of another  chemical
which has a much lower toxicity but a higher hazard.   Our  systems, both
aquatic and terrestrial, have lived, evolved and thrived in the presence
of some v«ry highly toxic substances.  Ozone rates  very high on the  list
of toxic materials, and yet its persistence  in the  environment  is so
short  that it is not a problem.  Too often,  I'm afraid, demonstrated
high  toxicity casts the dye before an assessment has been  made  as to
whether or not the chemical is going to be an environmental hazard.
Experience has taught us that the chemicals  to be most concerned about
are those which are persistent and are nearly water insoluble but highly
fat soluble.  These are the ones which last  for a long time in  the
environment and accumulate in the bodies of  aquatic organisms.  They  too
are the ones that are likely to form unacceptable residues in those


                                    188

-------
organisms used for  human consumption.  During the  last  five years, this
category includes most of the emergencies that have  arisen on the'Great
Lakes, the James River, and elsewhere in the country.

    With the widespread use of rapid tests, a danger exists that we
will stop developing better methods that are both  quicker and more
accurate.  An even  greater danger is that we will  discontinue the develop-
ment of chronic toxicity data and other information  which is so vital to
understanding how chemicals affect aquatic  systems.  It has only been
through the  development of an extensive chronic  toxicity data base that
the validity of many of these short-cut methods, now available, could be
evaluated.   Unless  we continue to advance the basic  science of aquatic
toxicology,  the return on future efforts to develop  better methods and
make better  predictions will be lessened.   We have an urgent need to see
that a proper balance occurs between funds  expended  for the development
of laundry lists  of numbers for making regulatory  decisions and, on the
other  hand,  the advancement of the basic science of  aquatic toxicology.
We need better understanding of modes of action  and  further work on
selection of the  most appropriate organisms on which to perform our
tests. Neither can we forget that our ultimate  goal is to protect a
system which is not a random collection of  individuals, but rather a
relatively  intricate grouping of plant and  animal  populations which are
interdependent  on each other and which will all  be affected by a change
in any one.   Sometimes one gets the impression that  those working on
aquatic  ecosystems  think that there is a universal aquatic ecosystem
which, if understood, could explain all other systems.  Certainly such
is not the  case,  but indeed most workers do expect that the general
functions that  occur in aquatic systems are similar  enough that once the
fundamental  ones  are understood, data from  laboratory experiments such
as the toxicity  tests described above can be more  intelligently and
efficiently  applied to the problems of the  real  world.

     The increasing scarcity of natural resources  and the attendant rise
in cost  for  these materials will probably force  us to use a property of
ecosystems which  we have tried not to use in the past decade.  I refer,
to the assimilative capacity of waters which can be  so  useful to us, but
which has been  so abused during the first three  quarters of this century.
Because  we  so foolishly abused that valuable resource of aquatic systems
in the past  does  not in any way preclude the intelligent use of it in
the future.   As man becomes smarter about the total  ecological effect of
his activities which satisfy his seemingly  infinite  desire for contraptions,
we may recognize  that the mountains of sludge that we produce in our
chemical waste  treatment plants and the attendant  environmental damage
and resource drain  that accompanies the mining,  production, transportation
and application of  these chemicals, may well create  a far more serious
ecological  effect than the intelligent utilization  of  the assimilative
capacity.   If  one  takes an old fashioned and true ecology course, one
of the first principals he learns is that the environment affects all
organisms and all organisms affect the environment.  Man is an organism.
He has,  does now, and always will affect the environment in which he
lives, and  our  goal must be to affect the environment in the least
adverse way. I am  convinced that it would  be easy for  us in our efforts


                                    189

-------
to avoid aquatic impact "at any price" to produce  a  far  greater ecological
impact on the total system, that is more undesirable from man's point of
view—all because we made a mistake in the past by expecting the impossible
from our aquatic systems.

REFERENCES

American Public Health Association, American Water Works Association,
and Water Pollution Control Federation.  1975.  Standard Methods for the
Examination of Water and Wastewater.  14th ed.  Washington,  B.C. 1193 p.

Chiou, C.T., V.H. Freed, D.W. Schmedding, and R.L. Hohnert.   1977.
Partition coefficient and bioaccumulation of selected organic chemicals.
Environ. Sci. & Tech., 11: 475-478.

Drummond, R.A., G.F. Olson, and A.R. Batterman.  1974.   Cough response
and uptake of mercury by brook trout, Salvelinus fontinalis,  exposed to
mercuric compounds at different hydrogen-ion concentrations.   Trans.
Amer. Fish Soc. 101: 244-249.

Drummond, R.A., W.A. Spoor, and G.F. Olson.  1973.   Some short-term
indicators of sub-lethal effects of copper on brook  trout,  Salvelinus
fontinalis.  J. Fish. Res. Board Can. 30: 698-701.

Macek, K.J. and B.H. Sleight, III.  1977.  The utilitiy  of toxicity
tests with embryos and fry of fish in evaluating hazards associated with
the chronic toxicity of chemicals to fishes.  In:  Aquatic Toxicology
and Hazard Evaluation.   (F.L. Mayer and J.M. Hamelink, editors).  American
Society for Testing and Materials.  STP 634.  In press.

McKim, J.M. 1977.  Evaluation of tests with early  life stages of fish
for predicting long-term toxicity.  J. Fish. Res.  Board  Can.  In Press.

Neely. W.B., D. R. Branson, and G.E. Blau.  1974.  Partition coefficient
to measure bioconcentration potential of organic chemicals in fish.
Environ. Sci. & Tech. 8: 113-1115.

Smith, W.E.  A cyprinodontid fish.  Jordanella floridae,  as a laboratory
animal for rapid chronic bioassays.  1973.  Jour.  Fish.  Res.  Board  Can.
30: 329-330.

Spoor, W.A., T.W. Neiheisel, and R.A. Drummond.  1971.   An electrode
chamber for recording respiratory and other movements of free-swimming
animals.  Trans. Amer. Fish. Soc. 100: 22-28.

The Committee on Methods for Toxicity Tests with Aquatic Organisms.
1975.  Methods for acute toxicity tests with fish, macroinvertebrates,
and amphibians.  U.S. Environmental Protection Agency, Duluth, Minn.
Ecological Research Series EPA-660/3-75-009.

U.S. Environmental Protection Agency.  1976.  Bioassay procedures for
the ocean disposal permit program.  Environmental  Research Laboratory,
Gulf Breeze, Fla.  Ecological Research Series EPA-600/9-76-010.

                                    190

-------
U.S.  Environmental Protection Agency.   1973.   Biological field  and
laboratory methods for measuring the quality  of  surface waters  and
effluents.   National Environmental Research Center,  Cincinnati,  Ohio.
Ecological Research Series EPA-670/4-73-001.

Veith,  G.D.  and N.M. Austin.  1976.  Detection and isolation of  bio-
accumuable chemicals in complex effluents.  P. 297-302.  In; Identi-
fication & Analysis of Organic Pollutants  in  Water.   (Lawrence  H. Keith,
editor). Ann Arbor Science, Ann Arbor,  Mich.

Woelke, C. E.  1972.  Development of a  receiving water quality  bioassay
criterion based on the 48-hour pacific  oyster (Crassostrea gigas)
embryo. State of Washington, Dept. of Fisheries, Spokane, Wash.  Technical
Rpt. No. 9.

DISCUSSION

L. Duffy, Standard Oil of Indiana:  With the  demands and problems of the
analysis of  the problem organics, do you view bio-assay as a better
analytical  tool?

Mount;  Yes.  The only valid way to measure toxicity is with an organism.
You can't do it with, an analytical instrument and yet we seem to rely  on
the analytical approach with no regard  for biological response  even
though that  really is the goal toward which much of this work is aimed.
I think that the toxicity test or the bio-assay  should be considered as
an analytical tool and that it can do much to reduce the costs  of doing
complex analytical work.  Also, the high pressure liquid chromatographic
column, for  example, might separate away 90%  of  the compounds in a  waste
or a mixture of materials that we are not  all that concerned about  and
help us zero in on those which are going to be problems.

Paul Mikolaj, Lion Oil Company:  What is the  state of the art in the
future of continuous-flow bio-assays?

Mount:  The  chronic test that I mentioned  must be done in a continuous
flow system and as I said, I think it must remain the foundation of our
aquatic toxicity work, but it doesn't have to be a routine workhorse.  It
is through chronic tests, analagous to  the two-year rat study which is
so common in other toxicology* work, that you  find the mode of toxicity.
Such information is necessary for rapid test  development and predictive
toxicology.  So I see the chronic test as being the workhorse in the
research laboratory, so to speak, where one develops the fundamental
toxicology to evaluate the suitability  of  much faster methods.   In  the
prepared paper I have pleaded for a proper balance between the  effort
that goes on fundamental research and developing numbers in a production
laboratory.   The chronic test is an extremely essential tool as a  fundamental
test and of  course, depending on what problems you are trying to mimic
in the real  world, it may be very useful for  other purposes. If one  is
concerned about a relatively confiftnuous discharge, then obviously  the
flow-through system is the way to go.   If  one is concerned about a
Pesticide application, then the exposure period  is likely to be short,


                                    191

-------
  and then perhaps it is not  the right test.   I  think we should be careful
  that we don't confuse continuous flow with  continuous exposure,  because
  they are different tests.
BIOGRAPHY           Donald I. Mount

         Donald I. Mount holds a B.S. degree in
Wildlife Conservation from Ohio State University.
He also earned a M.S. and a Ph.D. from Ohio
State University, concentrating on fish toxicology
and physiology.  He is currently Director of the
Environmental Protection Agency Research Labora-
tory-Duluth,Minnesota.   Dr. Mount has served on
various national and federal committees in the
interest of water quality control.  He  has also
published over 40 articles.
                                       192

-------
                AN OVERVIEW OF ADVANCED TREATMENT SYSTEMS

                        Davis L, Ford, Ph.D., P.E.
                           Senior Vice President
                         Engineering-Science, Inc.


     The  term  "advanced treatment systems" can have many connotations,
ranging from what  is  generally considered a 1977 level of technology
(Best Practicable  Control Technology Currently Available) to Best Avail-
able Treatment Economically Available required by 1983, or some level in
between.  Translating these levels of technology into unit process
requirements,  the  systems discussed in this overview include biological
treatment,  bio logically-treated effluent polishing, and physical-chemical
treatment,  primarily  related to petroleum refining, petrochemical, and
organic chemical wastewaters.  Recent developments relative to process
optimization and limitations will be included, as well as documentation
of process  performance.

BIOLOGICAL  TREATMENT

     Although  biological treatment systems per se are normally not considered
"advanced," they do serve as the most important component of most treat-
ment facilities which either now or in the future must produce effluents
with a  quality consistent with advanced wastewater technology.  The
trend during the past decade has been toward the use of high rate biological
processes for  the  treatment of organic juidustrial wastewaters.  The
systems have generally included either ^fuspended-growth (activated
sludge) or  fixed-growth (rotating biological surface) processes or some
modifications  thereof.  The completely mixed activated sludge process is the
most widely applied biological system in treating industrial wastewaters
with relatively high  organic concentrations.  The problem most common
with activated sludge systems treating industrial wastes is accomplishing
effective solids-liquids separation in the gravity clarifier which
follows the aeration  basin.  Many industrial wastewaters will tend to
generate  a  significant fraction of dispersed biomass which do not adequately
separate  in the ,^^rifier.  For example, the average effluent suspended
solids  (TSS) wil^range from 25 to 75 mg/1 from an activated sludge
process.   If the effluent limitations are more restrictive than the
indicated range, then effective effluent polishing systems must be
included  in the process design, such as the ability to add organic
polymer flocculants and/or granular media filtration.

     More recently, fix^-growth systems are becoming quite popular in the
petroleum refining and petrochemical industrial categories (Ref. 1).  One
of the more popular fixed-growth systems is known as the rotating biological
surface  (RBS).  In this system the biological mass grows on the surface of
large-diameter discs  whicjfe are placed side by side on a rotat^jig shaft.
The bottom  portion of the rotating discs are emersed in a basin through
which the wastewater  flows.  It is a facultative system, with the oxygen
transferred by direct contact between the slime and atmosphere as well as
air entrainment in the turbulence associated with the rotation.  The system
overcomes some of  the disadvantages of the stationary trickling filter
                                   193

-------
approach which includes the continuous shearing of  the  fixed  biomass as  the
discs pass through the water thus preventing an accumulation  of  surface
growth, and an effective penetration of oxygen through  the outer layers  of
the biomass.  It has lower energy requirements than the activated sludge
system, particularly when mixing rather than oxygen controls  design.  Two
other advantages which are proposed but not yet demonstrated  in  the refining
and petrochemical industry is the fact that suspended solids  in  RBS effluents
are lower than those from activated sludge systems  and  nitrification can be
better accomplished by concentrating the nitrifying microorganisms in the
latter stages of discs on the shaft, and adjusting  the  pH  in  the nitrifying
stages to maximize the biochemical oxidation of ammonia.   If  the organic
concentration of the industrial wastewater is high  (the BOD5  exceeds
1,000 mg/1 for example) the capital costs and energy requirements of the
RBS system over activated sludge may be reduced.  Moreover, if many of the
organic compounds are refractory and require long contact  periods for
adequate degradation, the effective biological growth-substrate  contact  may
be insufficient to reduce these refractory compounds to the required level.
Therefore, one should be certain that the RBS system can reduce  the COD
level adequately before making the final process selection.

Process Flexibility for Biological Systems

     One of the primary limitations in applying the biological method of
treating industrial wastewaters has been the failure to incorporate proper
pretreatment and process flexibility facets into the basic design.   As the
biochemical oxidative mechanisms are complex, particularly for industrial
wastewaters discharged from the refining, petrochemical and organic chemical
industries, every effort must be made to accommodate the biological popula-
tion to the maximum extent.  Some of the approaches which  can be used in
insuring process flexibility are discussed in the following paragraph.

     Equalization can be one of the most critical single processes  in the
overall biological treatment facility.  The deleterious effect of transient
loadings on biological systems, both hydraulic and  organic, is well docu-
mented  (Ref. 2 and 3).  There are several rational  methods which can be
utilized from raw waste load variations in size and equalization basins  in
order to dampen influent variations (Ref. 4 and 5).  It is also  prudent  to
include auxiliary "off-specification" basins in the biological process
design in order to temporarily receive and store waters of inordinately
high organic concentrations or those with potential toxicity.  This water
can then be pumped from inventory back to the biological system  at  a con-
trolled hydraulic rate.  Diversion of the wastewater stream can  be accom-
plished automatically using an on-line analyzer.  Such  basins, along with
equalization, reduce the hydraulic and organic variations  to  the biological
systems and normally result in significantly higher overall removal
efficiency.

     Specific pretreatment steps of industrial,waste are often effective
in enhancing the overall performance.  For example,  it  may be prudent to
dilute the concentration of highly degradable organic constituents  to a
concentration level which will allow more effective biochemical  oxidation
of organic compounds.  This is true when biochemical inhibition  can occur


                                    194

-------
which is attributed  strictly to influent organic concentration rather than
to constituent  complexity or resistance to biodegradation.  Predilution in
this case  is a  legitimate and appropriate pretreatment step to improve the
overall performance  of the biological system, particularly when recognizing
that kinetics of  removal are more concentration-sensitive than mass-sensitive.
Another pretreatment approach which can be considered is the steam or sol-
vent stripping  of selected waste streams, with proper air emission control
measures as applicable.  This approach can reduce high organic loads,
sequester  organic load variations, and remove potentially toxic or inhibitory
contaminants,  improving the amenability of the stream to biological treat-
ment.  Probably the  most common example of this pretreatment is sour water
stripping  in  petroleum refineries, but there are numerous other instances
in the chemical processing industry where this is an effective pretreatment
step.  Recent practice has indicated that hydrolyzing selected organic
wastewater streams by adding caustic and exercising pH control can enhance
the biodegradability of the hydrolyzed stream  (Ref. 6).  This practice has
been applied  as a pretreatment step in the biological treatment of pesticide
and herbicide waste streams with positive results.

     One operational technique which can provide additional resistance to
biological upset is  increasing the inventory of biological solids in the
aeration basin of a fluidized activated sludge process.  This can be
accomplished  by increasing the sludge recycle ratio and/or reducing sludge
wastage.   The increased inventory simply implies that the quantity of bio-
toxic or biostatic constituents per bacteria is reduced.  The design MLSS
levels in activated sludge systems typically range from 2500 to 3000 mg/1
while in some cases for industrial applications, the MLSS level are main-
tained from 8000 to 10,000 mg/1 (Ref. 7).

Process Optimization of Biological Systems

     When considering advanced treatment systems, one assumes the biological
portion of the facility is designed and operated for maximum performance.
This may not  be the case and some discussion is therefore merited.

Sludge Ape  Sludge age, or the average contact time between the microorganism
and the substrate, is becoming increasingly popular as a process
control parameter (Ref. 6, 7).  Sludge age can be defined mathematically
using several approaches, the most common for activated sludge being:
          e
           c
 where:
          9   =  sludge age, days
          X -V  =  average aeration basin MLVSS,  mass (V = basin volume)
           Si
          AX  =  sludge wastage, mass per day

     In a controlled reactor, the sludge  age is  similarly defined  as:
                                    195

-------
where:                                                  -.,-,

          Q  =  total system  flow

          Q  =  solids wastage  flow
          X  =  solids wastage  concentration

          X  =  effluent  solids concentration
           e
     The sludge age and the more commonly used Food-to-Microorganism (F/M)
ratio can be related in the following manner:

          I   _   a(AF/AT)  -b
          9       X -V                                                  (3)
           c       a

where:
          a  =  sludge yield  coefficient

          AF/AT   = the mass (Ibs)  of food (COD or BOD)  per day removed
          X  *V    = the mass (Ibs)  of microorganisms in the aeration basin
           a
          b       = the endogenous rate  coefficient

     The hydraulic retention  time,  t, and its interrelationship to sludge
age has not been  adequately defined in  terms of process kinetics,  although
such a relationship would be  particularly meaningful in developing design
equations.  There are many indications  that contact time between the
biomass and the waste constituents, as  measured by hydraulic retention  time,
can also be an important  process parameter for treatment of complex organic
wastewaters.

     Recent investigations have suggested that sludge age is the best con-
trol parameter, and, contrary to some theories, an extended sludge age  of
forty days or more maximizes  performance in terms of sludge settleability,
process control,  and organic  removal efficiency (Ref. 8).   Other studies
have shown that the critical  sludge age (defined as the minimum 0  necessary
to achieve maximum organic removal) is  a function of substrate anS tempera-
ture, but does not exceed six to seven  days even for a complex chemical
waste at temperatures of  less than 10°C (Ref. 9).  It can only be concluded
from these investigations that  the optimum sludge age for an activated
sludge system treating industrial wastewaters is dependent on the nature
of the influent,  namely,  its  concentration and complexity, and the operating
temperature of the aeration basin.  In  other words, sludge age alone does
not adequately define the ability of a  suspended-growth biological
system to provide maximum removal of organics from a specific wastewatef.
The Food-to-Microorganism ratio (F/M) and hydraulic retention time are  also
important control criteria in some cases.  Treatability studies are thereby
justified to establish the design 0 and other control parameters specific
to the industrial waste and the mosi severe operating condition.

Temperature

     It is important to pursue  the temperature effects on biological systems,
as the Streeter-Phelp's empirical modification of Arrhenius' law has not
                                    196

-------
always  held  in predicting reaction rate coefficients at defined temperatures.
This equation is stated as follows:
where:
          KT  =  the reaction rate at temperature, T

          K_,  =  the reaction rate at temperature Tn
           Ll                                      1

          9   =  the temperature activity  coefficient which is a
                 constant for a given wastewater

     It has been proposed for several years  that colder  temperatures in the
aeration basins have a more pronounced effect  in terms of reduced process
efficiency for wastewaters of higher molecular complexity and solubility
(Ref. 8).  This has recently been confirmed  by determining the critical
sludge age for wastewaters of varying complexity undergoing aeration at
several temperatures as shown in Figures 1,  2,  and 3  (Ref. 9).  Based on
these studies, the critical sludge age for each temperature and wastewater
can be approximated as follows:
                                   Temperature. °C     Critical Sludge Age.
                                                            Days

     Domestic Wastewater                30°                   2
                                        10°                   3.5

     Chemical Wastewater                30°                   2.5
                                        10°                   5.5

     Petrochemical Wastewater           30°                   3.5
                                        10°                   8

     It is of paramount importance, therefore,  that designers provide sludge
ages which are adequate for maximum performance predicated on wastewater
complexity and swings in operating temperatures.  This is particularly
important for systems with long hydraulic  retention times since aeration
basin temperatures will approach ambient air temperature even if the waste-
water is quite warm before aeration.

Bulking Sludge

     The solids-liquid separation phase of biological treatment has always
been one of the more important elements in successfully  treating wastewater
using this system.  Bulking sludge is one  of the main, precursors to high
effluent TSS levels and consequently has received much attention in the
attempt to optimize biological treatment facilities.  Sludge bulking is
particularly prominent in the food processing  industries, primarily based
on the fact that an easily available carbon  source tends to promote fila-
mentous microorganisms.  However, sludge bulking also occurs in many other
industrial categories, and consequently occupies a role  of primary importance
in evaluating methods of process control.  Historically, bulking is extremely


                                    197

-------
difficult to correct once it has occurred and  thus,  a preventive approach
is generally most successful.  As filamentous  microorganisms tend to
thrive at lower pH, oxygen tension, and nutrient  levels than do the floc-
culating microflora, it is important to insure adequate pH control,
sufficient aeration, and an adequate supply  of nitrogen and phosphorus  to
the system.  The use of dissolved air flotation with polymer addition as
the final clarification step is receiving an increasing amount of attention.
If the wastewater contains an easily available carbon and  filamentous
organisms tend to historically persist, dissolved air flotation with chemi-
cal addition facilities should be given careful consideration.  It must
be recognized, however, that effluent TSS concentrations from biological
systems using DAF as the first cell separation step  will range,  as for
conventional activated sludge systems, from  25 to 75 mg/1  which still may
be inadequate to meet criteria, necessitating  effluent polishing.

     Preliminary results in evaluating contact stabilization against com-
pletely mixed activated sludge for treating  one industrial waste indicates
that the activated  sludge system is less prone to produce  a filamentous
population and thus less susceptible to sludge bulking. This is possibly
attributed to the fact that in the contact stabilization approach, a higher
food-to-microorganism ratio in the contact tank promotes more filaments
which persist through the reaeration phase of  the process.  For this reason,
conversion of contact stabilization to completely mixed activated sludge
may enhance overall sludge settleability and process performance in systems
with bulking problems.

     The addition of chemicals such as hydrogen peroxide to the aeration
basin or recycle sludge to minimize sludge bulking has had mixed success,
although operating  costs are a significant factor if the procedure must be
implemented continuously to prevent recurrence of the problem.  The concept
of biodynamic control using a controlled seeding  of  sludge microflora to
a biological waste  system treating food and  dairy wastewaters has recently
been reported as an effective method of minimizing sludge  bulking (Ref. 10).
Although this is a  theoretically sound concept, the  practicality of con-
trolled seeding in  large biological systems  should be verified in terms
of process and cost effectiveness before such  an  approach  is given serious
consideration.                                             r

Optimization of Biological Nitrification

     Stringent effluent ammonia concentration  levels required by many per-
mits has necessitated the use of nitrification in many industrial biological
treatment systems.  As nitrifying microorganisms  are extremely sensitive
to pH and temperature, and since many process  variables and trace chemical
constituents affect their performance, it is capricious to predict nitri-
fication strictly on sludge age or hydraulic retention time.  Consequently,
designers of biological systems are having to  use more sophisticated con-
cepts in order to insure the biological removal of ammonia nitrogen from
industrial wastes.  The concept of two-stage activated sludge has been
proposed, utilizing the advantages of isolating the  nitrifying microorganisms
in a second state while minimizing nitrifying  inhibition factors in the first
stage of aeration (Ref. 11).  Such a concept,  although considerably more

                                    198

-------
expensive than  a single-stage system, does provide more process control for
enhancing nitrification.  For example, the nitrifying reduction attributed
to winter temperatures can be partially offset by increasing the recycle
ratio of the nitrify ing-rich sludge in the second stage, enhancing the
overall nitrification during the critical seasons of the year.

     A recent study indicated that an aerobic submerged filter may be a
feasible approach for the economical nitrification of low-strength wastes
(Ref. 12).   This system provides an upward flow of liquid  through plastic
or natural  media and through efficient solids capture and  control of
hydraulic detention time, stable nitrification has been reported.  Surface
area effects on nitrification are not well documented, but undoubtedly are
significant when comparing suspended and fixed-growth systems relative to
nitrification.

     The rotating biological surface  (RBS) has also been quite successful
in biologically treating ammonia nitrogen.  A properly-designed RBS system
offers inherently the same advantages as a two-stage activated sludge
system, namely, allowing an enriched nitrifying population to develop in
the latter  stages of the RBS process.  As the pH tends to  drop through
an RBS system via the production of carbon dioxide, it may be necessary
to adjust pH in the nitrification stages by the addition of caustic to
raise the pH level to the nitrification optimum of 7.5 to  8.3.  It is
important to recognize that 7.1 of alkalinity  (as calcium  carbonate) can be
destroyed per unit of ammonium ion (as nitrogen) nitrified, underscoring
the need for good pH control, particularly for wastewaters with low
alkalinity.

     It should be noted that there have been several process problems in
terms of nitrification when high amine concentrations are  present in the
wastewater, probably through biochemical cleavage reactions of the amine
functional group, actually creating ammonia biochemically  in the biological
system.

     A particular problem in applying nitrification to industrial waste-
waters is the sensitivity of the nitrifying bacteria to a  wide variety
of identified and unidentified organic and inorganic chemicals.  This is
especially a problem with complex wastewaters such as those of the
chemical processing industries.  All other conditions being proper,
nitrification may still not be obtained for a given wastewater unless
the inhibitory components are found and removed.  This problem makes it
mandatory that bench and/or pilot-scale treatability studies on the
actual wastewater to be treated be conducted prior to design of nitrification
into a biological treatment system.  Often, the treatability studies can
identify certain waste streams and/or components which can be removed or
pretreated  to promote nitrification of the total waste stream.

EFFLUENT POLISHING

     Polishing the effluent from biological  treatment  systems  clearly con-
stitutes "advanced waste treatment" and  is  the model  for  BPCTCA  and/or
BATEA levels of technology for several industrial  categories.  As previously

                                   199

-------
mentioned, there are several  approaches  for improving biologically treated
effluent quality.  The methods  discussed here include chemical addition,
powdered activated carbon addition to activated sludge systems, post
filtration, and fixed-bed carbon polishing units.

Chemical Addition

     The addition of coagulants or coagulant aids  between the aeration basin
and the final  clarifier  often can enhance quality  by reducing TSS and
colloidal materials in the  final effluent.  The organic oxygen demanding
substances associated with  these suspended and colloidal materials are
correspondingly removed.

     There have been mixed  results as to the efficacy of this approach.  Nor-
mally, polymer additions in the range of 1 to 5 mg/1 result in a 10 to 30
percent reduction in effluent TSS from an activated sludge system, although
this may vary. One of the  more favorable experiences of adding polyelectro-
lytes  to an activated sludge  system treating petroleum refinery waste-
waters is shown in Figure 4.  As shown in this case, better than fifty per-
cent of the TSS were removed  upon the additon of approximately 5 mg/1 of the
polyelectrolyte.  There  are many variables which influence the applicability
of  adding coagulants or  coagulant aids to enhance  the clarification of
biologically  treated effluents, however, so test conformation studies should
be  performed.  Moreover, polyelectrolytes are very expensive and high con-
centration demands would result in excessive operating costs.  The instal-
lation of a chemical feed system and mixing basin  also would require a
capital expenditure.

Powdered Activated Carbon Treatment

     The addition of powdered activated  carbon to  activated sludge systems
to  enhance settleability and  remove residual organic materials has been
proposed for  several years  and  has been  implemented on several occasions.
The addition  of 450 mg/1 of powdered activated carbon to an activated
sludge aeration basin treating  refinery  wastewaters resulted in a relatively
substantial increase in  process performance as shown in Figures 5 and 6
 (Ref.  13).  This approach was not considered as a  long-term corrective
measure, however, based  on  the  difficulties in handling the powdered carbon
around the aeration basin.

     The most  significant project currently in operation which utilizes
powdered^ activated carbon and a biological mass and aeration system is
the 40 MGD DuPont-Chambers  Works Facility.  The initial results indicate
good performance in terms of  effluent quality, although the efficacy of
dewatering, incineration, and regeneration of the  mixed biological-
carbon sludge  has not been  proven and could represent the critical path
in  the overall applicability  of this approach (Ref. 14).  Required
carbon dosages and the ability  to reuse  the adsorbent material obviously
will dictate  the cost-effectiveness of this approach as compared to
other  alternatives.

     As with  the chemical addition option described above, only actual
testing of the method and careful cost analyses can determine its
                                    200

-------
applicability to a particular wastewater.  Both bench and pilot treatability
studies as well as the experimental use of powdered  carbon in the full-
scale system, if possible, should be used in  the  evaluation process.

Post Filtration

     Post filtration of biologically treated  effluents has been applied in
several industrial wastewater treatment facilities,  and  in fact, used as
the BPCTCA model for developing the petroleum refining guidelines.  Post
filtration systems, based on field performance, do a reasonably good job
of reducing the effluent TSS concentration to 10  to  15 mg/1, although
chemical addition to the filter influent is often required to strengthen
the floe and make it more filterable.  Additionally, there is a practical
limit of 80 to 110 mg/1 of TSS which can be charged  to the filter, an
excess of which causes inordinately short run times  and  reduces the practi-
cality of the filter process.

     The operating biological treatment-post  filtration  systems treating
refinery and petrochemical wastewaters are producing effluents with long-
term TSS averages of 10-15 mg/1.  The efficiency  of  a post filtration
system depends to some degree on  the influent TSS concentration as shown
in Figures 7 and 8.  The data presented in Figure 7  was  developed from
studies using pilot-scale filters receiving a biologically treated effluent.
It is noted that the addition of  polymers to  both the deep bed and shallow
bed downflow filters made the overall filter  performance less dependent
on influent TSS, an inherent advantage of the polymer addition.  The data
in Figure 8 were developed in pilot-scale downflow filtration studies,
indicating the same trend and TSS residual concentrations.  The reduction
of oil and grease  (O&G) compounds attributed  to the  addition of post-filters
is not dramatic, as indicated in  Figures 9 and 10 (Ref.  15).  The probability
distribution of O&G in biologically treated effluent from case histories
in the petroleum refining industry indicate a range  of 3 to 15 mg/1 (median
values) as shown in Figure 9 while the two systems in the refining industry
which have filters shown in Figure 10 produce a median value O&G concentra-
tion of 7 to 8 mg/1.

Activated Carbon Polishing

     There are presently no full-scale operating  biological fixed-bed
carbon polishing treatment facilities treating refinery, petrochemical, or
organic chemical wastewaters for  which data are available, although some
are reportedly close to beginning operations.  For  this reason,
pilot-scale studies must be used  as the data  base.   The  COD removal in
carbon columns polishing activated sludge effluent as determined in various
pilot-plant studies for petrochemical and refinery plants is tabulated in
Table 1 (Ref. 16).  A 59 to 83 percent removal is noted, indicating that
residual COD can be further reduced in such an application.  This is
true because there is an inherent process compatibility  between biological
and carbon treatment as many compounds resistant  to  biochemical degradation
are amenable to carbon adsorption (Ref. 17).   It  should  be recognized,
however, that the cost-effectiveness of carbon polishing expressed in
Ibs of BOD or COD removed per cost unit is poor,  based on the high cost of


                                    201

-------
removing a relatively low mass of residual  compounds.   For this reason,
every effort should be made to select  less  costly modes of effluent polish-
ing if regulatory constraints so allow.

     Activated carbon, applied as a  process in a physical-chemical system
treating these industrial wastewaters  is  less applicable,  as discussed in
the following section.

PHYSICAL-CHEMICAL SYSTEMS

     Most of the previous studies relative  to the physical-chemical treat-
ment of industrial wastewaters has centered around  activated carbon treat-
ment, although steam and solvent stripping,  chemical oxidation, chemical
coagulation and precipitation, and other  forms of non-biological treatment
are possibly applicable.

     It should first be understood that activated carbon treatment of organic
industrial wastewaters should be carefully  investigated prior to making
process commitments.  Several studies, for  example,  have underscored the
limitations of activated carbon as total  physical-chemical treatment process
as compared to carbon polishing of biologically treated effluent (Ref. 16,
17, 18, 19).  The estimated effluent quality for the activated sludge,
carbon, and combined treatment of refinery  wastewaters are tabulated in
Table 2  (Ref. 18).  A more recent study comparing activated carbon as a
physical-chemical or polishing process was  conducted by the Environmental
Protection Agency  (Ref. 20).  Both API Separator effluent  and biologically-
treated effluent from a petroleum refinery  were charged to pilot-scale
columns in order to obtain a comparative  evaluation.  These quality data
indicated that the carbon system was significantly  more effective when
operated  in conjunction with the biological process than when applied
singularly, both in terms of BOD and COD.   This is  consistent with the
results observed in pilot studies conducted by the  author.  The limitations
of physical-chemical systems designed  around the activated carbon adsorp-
tion process therefore are a function  of  the organic compounds in the
wastewater which are not amenable to adsorption, reducing  overall efficiency.
Even though physical-chemical systems  have  more of  an "advanced waste
treatment" connotation, they could,  in fact, produce a lower effluent
quality than biological treatment processes.

SUMMARY

     In summary, it is the author's  opinion that "advanced waste treatment"
for the petroleum refinery and organic chemical industries centers
around some form of biological treatment, at least  for the next decade.
There are no direct alternatives which are  presently as cost-effective in
terms of  chemicals or energy, and there are few likely process candidates
which are likely to be more attractive, at  least through the 1983 date
for implementation of Best Available Treatment Economically Achievable.
Physical-chemical treatment is and will continue to serve as an important
adjunct, primarily as pretreatment of  specialty streams or as polishing
units in series with biological processes.   Although sole physical-chemical
processes are possibly applicable in certain cases,  careful conceptual

                                   202

-------
planning and process confirmation through treatability studies should precede
final selection.

     The applicability of the activated carbon process in treating indus-
trial wastewaters,  for example,  is contingent on many factors, including
the amenability of  the dissolved constituents to sorption, the presence
of other substances which enhance or impede the sorption process, the sound-
ness of engineering, the degree of pretreatment and proper operation and
maintenance of  the system.   As activated carbon was one of the primary
processes factored into the development of  the 1983 Best Available Tech-
nology  (BAT)  guidelines for many industrial categories, some of which have
been remanded by the Courts, it is important to fully understand the process
and its limitations.

     Finally,  it is important to recognize  that "advanced waste treatment"
in effect will  have an EPA  definition when  the BPT, BAT, and new source
performance standards for the Organic Chemicals and Plastics and Synthetics
categories, and the BAT standards for the Petroleum Refining category
are repromulgated later this year.
DISCUSSION
Garr M. Jones/ Brown and Caldwell: We have noticed that in many industries and their
municipal wastewater treatment systems the advantages of what we call a coupled system,
a fixed-growth system followed immediately by activated sludge.  I noticed that you
passed over this particular combination and I would like to ask you to comment on the
advantages that we see, first of all lower operating costs, smaller clarifiers because of
improved solid settling of characteristics, and  a far more stable process.

Davis Ford;  Yes, I would say that would have the same basic process concept as the
trickling filters did in the past, and  I think from the process point of view it makes a lot
of sense.  You say there are lower operating costs; I would only caution there that if you
take the capital cost of that system and amortize that and include the operating costs and
it still is cheaper, then that certainly can be justified on an economical basis.  From the
process point of view I would say the addition  of a fixed-growth reactor before or even
after activated  sludge has some merit certainly to be investigated.  I certainly agree
with that concept.
                                     203

-------
                                REFERENCES
 (1)       Ford, D.L., and Tischler, L.F.,  "Recent Developments in
               Biological Treatment of  Industrial Wastes," submitted
               to Chemical Engineering.  (May,  1977) .

 (2)       Ford, D.L., and Eckenfelder,  W.W.,  "The Effect of Process
               Variables on  Sludge Floe Formation and Sludge Settling
               Characteristics," Water  Pollution Control, Federation
               Meeting, Kansas  City, Missouri  (Sept.  1966).

 (3)       Ford, D.L., "Factors  Affecting Variability  from Wastewater
               Treatment Plants," Prog.  Water  Technology,  Vol. 8,
               No.  1, pp. 91-111, Pergamon Press, London (1976).

 (4)       LaGrega,  M.D., and Keenan, John  D.,  "Effects of Equalizing
               Wastewater Flows," Journal  WPCF,  Vol.  46, No. 1,
                (January 1974).

 (5)       Speece, R.E., and  LaGrega, M.D., "Flow Equalization by  Use of
               Aeration Tank Volume," Journal  WPCF, Vol. 48, No.  11,
                (November 1976).

 (6)       Shell Chemical Company, unpublished  internal report (1975).

 (7)       Ford, D.L., "Water Pollution  Control in the Petroleum Industry,"
               Industrial Wastewater Management  Handbook,  pp. 8-1 through
               8-75,  edited  by  H. Azad,  McGraw-Hill,  New York (1977).

 (8)       Grutsch,  J.F., "A  New Perspective on the Role of the Activated
               Sludge Process and Ancillary Facilities," Proceedings of
               the  Open Forum on Management of Petroleum Refinery
               Wastewaters,  sponsored by Environmental Proection  Agency,
               American Petroleum Institute,  the National Petroleum Refiners
               Association,  and the University of Tulsa, Tulsa, Oklahoma,
                (January 1976)..

 (9)       Sayigh, B.A., "Temperature Effects  on the Activated Sludge
               Process," Doctoral Dissertation,  the University of Texas
               at Austin, (May  1977).

(10)       Chambers, J.V., "Bioengineering  an Activated Sludge Microflora
               to Improve Waste Removal Performance," Proceedings of the
               Fifth Annual  Industrial  Pollution Conference, WWEMA,
               Atlanta, Georgia,  (April 1977).

(11)       Adams, C.E., and Eckenfelder,  W.W.,  "Nitrification Design
               Approach For  High Strength  Ammonia Wastewater," Journal
               WPCF,  (March  1977).
                                    204

-------
                          REFERENCES (CONTINUED')


(12)      McCarty,  P.L., and Haug, Roger T., "Nitrogen Removal From Waste-
               waters by Biological Nitrification and Denitrification,"
               presented at Society For Applied Bacteriology. Liverpool
               England, (197lT                                        '

(13)      Rizzo,  J.A., "Case History:  Use of Powdered Activated Carbon
               in an Activated Sludge System," Proceedings of the Open
               Forum on Management of Petroleum Refinery Wastewaterst
               sponsored by Environmental Protection Agency, American
               Petroleum Institute, the National Petroleum Refiners
               Association, and the University of Tulsa, Tulsa, Oklahoma,
               (January 1976).

(14)      Davis,  J.C., "Activated Carbon:  Prime Choice to Boost Secondary
               Treatment," News Features, Chemical Engineering. (April 11,
               1977) .

(15)      Engineering-Science, Inc., Report to the National Commission on
             .  Water Quality, Petroleum Refinery Industry - Technology and
               Cost of Wastewater Control, (June 1975).

(16)      ' Ford, D.L., "Putting Activated Carbon In Perspective to 1983
               Guidelines," presented at the 1977 National Conference  on
               Treatment and Disposal of Industrial Wastewaters and
               Residues, Houston, Texas, (April 26-28, 1977).

(17)      Ford, D.L., "Advanced Wastewater Treatment of Industrial Waste-
               waters Using Carbon Adsorption," Proceedings of the Fifth
               Annual Industrial Pollution Conference, Water and Wastewater
               Equipment Manufacturer's Association, Inc., Atlanta, Georgia,
               (April 19-21, 1977).

(18)      Ford, D.L., "The Applicability of Carbon Adsorption in the
               Treatment of Petrochemical 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,
               Tennessee, (September 1972).

(19)      Ford, D.L., "Current State of the Art of Activated Carbon
               Treatment," Proceedings, Open Forum on Management of
               Petroleum Refining Wastewaters. sponsored by the Environ-
               mental Protection Agency, American Petroleum Institute,
               the National Petroleum Refiners Association, and the
               University of Tulsa, Tulsa, Oklahoma, (January 1976).

(20)      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).
                                   205

-------
BIOGRAPHY        Davis L.  Ford

         Davis L. Ford holds a  B.S. degree in
Civil  Engineering from Texas A & M University and
a M.S. and Ph.D. degrees in Environmental Health
Engineering from fhe 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 insutries, the United Nations
(WHO and PAHO), the EPA and various state and
municipal agencies.
                                      TABLE 1
                    CARBON PILOT-PLANT RESULTS FOR POLISHING
                PETROCHEMICAL AND REFINING WASTEWATERS  (REF. 17)
                            Design
                              Q        Influent COD   Effluent COD   Percent
  Type  of  Wastewater       (MGD)         (mg/1)          (mg/1)      Removal

  Petrochemical              3           150              49           67
  Refinery                  26           100              41           59
  Refinery                  28           300              50           83
  Refinery                   8           100              40           60
  Petrochemical             29           150              48           68
                                        206

-------
                                                         Table  2

                         Estimated Effluent  Quality for  the Activated Sludge, Carbon, and Combined
                                        Treatment  of Refinery Wastewaters*(Ref. 19)
Constituent
Mean Value Range
Primary Effluent
Activated Sludge
Effluent
Total
Carbon
Effluent
Combined
Activated Sludge-
Carbon Effluent
Remarks
     COD
500-700 mg/1
100-200 mg/1   100-200 mg/1   30-100 mg/1
                             Exact COD residuals vary with
                             complexity of refinery & design
                             contact times in the Act.S. and
                             Carbon Treatment Plants.	
    BODC
250-350 mg/1
 20-50 mg/1     40-100 mg/1    5-30 mg/1
                             BOD residual depends on BOD/COD
                             ratio which characterizes rela-
                             tive biodegradability of waste-
                             water.               	  	
    Phenols       10-100 mg/1
NJ
                        mg/1
                  <1 mg/1       <1 mg/1      Phenols(ics)  are generally amen-
                                             able to biological and sorption
                                             removal.              	
     PH
 8.5-9.5
7-8.5
7-8.5
7-8.5
pH drop in Act. S. systems attri-
buted to biological production of
C02 and intermediate acids. pH
change in carbon columns depends
on preferential adsorption of
acidic and basic oreanics.
     SS
 50-200 mg/1
 20-50 mg/1
  <20 mg/1      <20 mg/1
              Primary effluent solids depend on
              design and operation of oil removal
              units. Act.  S.  effluent solids
              depend on effectiveness of  second-
              ary clarified.   Low effluent  solids
              characterize carbon column  effluent.
TDS 1500-3000 mg/1
NH3-N 15-150 mg/1
1
1500-3000 mg/1
1-30 mg/1
1500-3000 mg/1
10-140 mg/1
1500-
3000 mg/1
1-30 mg/1
TDS is essentially unchanged
through all three treatment systems
Exact concentration depends on pre-
stripping facilities, nitrogen
content of crudge charge, cor-
rosion additive practice and
biological nitrification.
                   1-10 mg/1
                     
-------
  FIGURE 1  EFFECTS  OF  SLUDGE AGE AND TEMPERATURE ON BIODEGRADABILITY
IUO
Jsso
•_•
< 60
o
S40
£E
g20
O
0
<
3O°C




D
^*~~
/
/


-
^^— 	



i^^^HBBMM
... •
IO°C



••••I*
l^*~~





i




1OMESTI



cmsn



"WATER








234 56789 10
                       SLUDGE AGE ,0C, days

 FIGURE 2  EFFECTS OF SLUDGE AGE AND TEMPERATURE ON BIODEGRADABILITY
           (REF. 9)
  100
                                         CHEMICAL WASTEWATER
                      34567
                       SLUDGE AGE ,0C, days
8
     10
FIGURE 3  EFFECTS OF SLUDGE AGE AND TEMPERATURE ON BIODEGRADABILITY
          (REF. 9)
  100      v
                                         PETROCHEMICAL WASTEWATER
                      34567
                      SLUDGE AGE, 0C, days
8
9
10
                                 208

-------
                                       FIGURE 4
 EFFECT OF  POLYELECTROLYTE ADDITION
 ON  EFFLUENT  TSS
 FLOW  PROBABILITY  DISTRIBUTION
        5   10   20 30 40 50 60 70 80  90  95
       % OF THE VALUES LESS THAN STATED VALUE
98 99
KEY
O WITHOUT POLYELECTROLYTE
A WITH POLYELECTROLYTE

-------
FIGURE 5  EFFECT OF EFFLUENT TSS OF CARBON ADDITION TO AERATION
          BASIN - PETROLEUM REFINERY (REF. 13)
240
                                     POWDERED CARBON
                                     ADDITION
          5   10      30    50   70      90  95     99

           % OF VALUES EQUAL  TO OR LESS THAN

FIGURE  6  EFFECT OF  EFFLUENT COD OF  CARBON ADDITION TO AERATION
         BASIN - PETROLEUM REFINERY (REF. 13)
280
240
^ 200
o>
J
Q 160
0
o
1- 120
LU
n! 80
U.
LU
40
0




•
«
•







+**






-—*
^-*^






x""
.-






^
-





BA,
X
-





9£C
4

~~





4S£
/







/
.«*





/

V ^^^




^^«
POWDER ED L
ADDITION




r



^
CARBON










5 10 30 50 70 90 95 99
           % OF VALUES EQUAL  TO OR LESS THAN
                         210

-------
              Figure 7
POST FILTRATION PERFORMANCE-
PILOT SCALE DEEP BED AND SHALLOW BED
DOWNFLOW FILTERS WITH  AND WITHOUT
POLYMER ADDITION
  PRACTICAL POST
I— FILTER LIMITATION
             40      60     80
              INFLUENT TSS (mg/l)

                  211

-------
                  Figure 8

     POST FILTRATION  PERFORMANCE-

     PILOT SCALE DOWNFLOW  FILTERS
  100
o>
CO


h-

UJ
ID
   50
          NOTE: DATA DEVELOPED

            AT HYDRAULIC LOADINGS

            OF 2-5
                                  100
                 INFLUENT  TSS(mg/l)
                     212

-------
             Figure 9


  OIL AND  GREASE  PROBABILITY

      BIOLOGICAL  TREATMENT
               (REF. 15)
  500
N,

O>
LJ
CO

LJ
a:
CD

o

<
    O.I   I    10     50    90    99  99.9

  % OF THE VALUES LESS THAN STATED VALUE
              213

-------
               Figure 10
   OIL AND GREASE PROBABILITY
    BIOLOGICAL TREATMENT WITH
         POST-FILTRATION
               (REF. 15)
100
 50
o>
JE
UJ
UJ
oc
a
<
 10
  I
 0.1    I     10     50      90    99
    % OF THE VALUES LESS THAN STATED VALUE
                                        99.9
                 214

-------
                SESSION VI

FUTURE CONSIDERATIONS IN BIOTREATMENT


            Chairman

            George W. Reid

            Regents Professor of Civil Engineering
            University of Oklahoma, Norman, Oklahoma


            Speakers

            J. F. Grutsch

            R. C. Mailatt

            "Design and Operation:  Bases for an Activated Sludge
            Route to BAT (1983) Water Quality Goals"


            Paul Goldstein and Robert W. Griffin
            "Considerations in Reuse of Refinery Wastewater"
            Milton R. Beychok
            "State-of-the-Art in Sour Water Stripping"

            Irv Kornfeld and Jay G. Kremer
            "Refinery Discharges to a Large Municipal Sewerage
            System"
                    215

-------
BIOGRAPHY            George W.Reid

       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 recent Urban Water Systems Conferences and
seminars in Colorado, New Mexico, Bogota, Lima,  Europe,
etc.
                                       216

-------
             DESIGN AND OPERATION:  BASES FOR AN ACTIVATED
            SLUDGE ROUTE TO BAT (1983) WATER QUALITY GOALS

                             J. F. Grutsch
      Coordinator, Environmental Projects, Standard Oil (Indiana)

                             R. C. Mallatt
 Manager,  Environmental & Energy Conservation, Standard Oil (Indiana)


ABSTRACT

     Since its debut 65 years ago, the activated sludge process (ASP)
has been developed into a widely applicable process with a tremendous
capacity for water purification.  Unfortunately, the role of the ASP has
become essentially totally identified with secondary treatment, that is,
best practicable control technology currently available (BPT) goals for
1977.  When effluent water quality superior to that normally associated
with BPT levels is a goal, for example, best available technology eco-
nomically  achievable (BAT), an add-on stage of granular carbon treatment
is the typical response for organics reductions.

     Current work is demonstrating that viewing the ASP solely as a
secondary  treatment process is shortsighted and, in fact, the ASP pro-
vides the  preferred means for achieving BAT water quality goals when
compared to alternatives.  A cost effective route for using the ASP to
achieve BAT effluent quality requires that the ASP be operated at
unusually  high sludge age and enhanced with high  surface-area activated
carbon.

     Other reports at this Second Open Forum will describe the success-
ful enhancement of the ASP using powdered activated carbon; this paper,
therefore, will focus on describing:  1) How ASP operation at very high
sludge age is achieved; 2) The support biological and kinetic data give
to defining, as being optimal, the ASP operating conditions required to
operate at very high sludge age, and 3) ASP process-design optimization
at very high sludge age.

INTRODUCTION

     At the May, 1977, API meeting, two papers (1,2) presented an alter-
native to  granular carbon adsorption treatment of activated sludge unit
 (ASU) effluent to achieve "best available technology economically achiev-
able" (BAT) proposed for 1983.  The alternate approach to BAT incor-
porates the use of newly developed, ultra-high  surface-area powdered
active carbon (PAC) in the ASU of the "best practicable control tech-
nology currently available" (BPT) model sequence for 1977.  The alternate
eliminates essentially all the capital costs of the granular carbon_
system. Further, operating costs for PAC can be minimized by optimizing
operation  of the end-of-pipe sequence.
                                 217

-------
     Optimization of the end-of-pipe sequence for  the PAC  alternate to
granular carbon treatment requires that the role of  the ASU and other
end-of-pipe treatment elements be defined as we described  at  the First
Open Forum (3).  A summary of the operating guidelines follows:

     1.   A systems optimization of a refinery end-of-pipe treatment
          sequence points to the ASU as the key element.

     2.   To achieve maximum water quality, system optimization points
          to reversing the historic work-horse role  of the ASU; i.e.,  it
          should be used only for the removal of essentially  soluble
          contaminants.

     3.   Reversing the role of the ASU yields a dramatic  series of
          beneficial effects.  At very high sludge age:

          a.   The SVI characteristics of the activated sludge mass are
               excellent.

          b.   Process control is greatly simplified.

          c.   The need for many process control tests is  eliminated.

          d.   An exemplary effluent low in TOG and  other  contaminants
               is produced.

          e.   The net cell yield is remarkably low.

          f.   The population dynamics of the sludge mass  improve.

          g.   Maximum ASU capacity for purification is achieved.

     4.   Systems optimization, wherein the key element is using the ASU
          for removal of only soluble contaminants,  permits clear  defini-
          tion of the roles the other elements play; i.e., colloidal and
          suspended matter must be essentially all removed in pretreat-
          ment 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 ASU.

     7.   Cell genetics, wherein inducible enzymes or mutant  species are
          required, support operating the ASU at very high sludge age
          for maximum purification capacity.

     8.   Bacteria are essentially enzyme factories.  Enzymes are sensi-
          tive to temperature, pH, excessive concentrations of heavy
          metals, oxidizing agents, salinity, UV, and other radiations.
                                 218

-------
         Using  the ASU for removal of only soluble contaminants  is
         consistent with obtaining a stable system less influenced by
         transient changes in environmental conditions since the endo-
         enzyme 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.

    The objectives of this paper are:

     1.   Outline how an ASU can be operated at high sludge age (50+
         days)  by outlining and describing in detail the chemistry of
         colloid destabilization which must be addressed if the  pre-
       ••- treatment (filter and dissolved air flotation) are to remove
         colloidal solids essentially completely.

     2.   Propose a model for chemical destabilization of negative
         colloids by weakly anionic polyelectrolytes.

     3.   Describe how the electrical properties of the activated sludge
         floe  impact on the mechanical design of the unit.

KINETIC  AND BIOLOGICAL DATA SUPPORTING ASU OPERATIONS ON SOLUBLE  CON-
TAMINANTS AT HIGH SLUDGE AGE

Wastes Properties

     An  intrinsic property of solids in the presence of water is  an
electrical  surface charge.  When colloids are being considered, the
electrical  charge is called zeta potential (ZP).  Almost all matter
dispersed in spent process water such as oil particles, silt, biocol-
loids, inorganic colloids, etc., has a negative ZP.  This repulsive
coulombic charge causes many particles to resist aggregation and  settl-
ing in primary  clarification facilities.  In the case of refineries, the
chemical oxygen demand (COD) of the contaminants in the effluent  from
API separators  average about 50 per cent soluble and 50 per cent  sus-
pended matter (Figure 1).

     By  comparison, the soluble COD in the effluent from a municipal
primary  clarifier is substantially less, being about 15-30 per cent
(Figure  2).  Thus, the colloids and suspended matter represent the major
COD component entering secondary treatment.
                                219

-------
ASU Kinetic Effects Achieved by the Removal of Colloids  and  Suspended
Solids from the Feed Stream

     Provides a Response to Biological Oxidation Kinetics.   A recent
kinetic equation  (equation 1) by Adams, Eckenfelder,  and Hovious  (4, 5)
applicable to a completely mixed ASU predicts that best  effluent  quality
(lowest Se) will be achieved with the lowest feed strength
            Se = Si (Si - Se)/KMt       '        '       Equation  1

     Where, Se = Soluble organics in effluent  (mg/1)
            S^ = organics in influent  (mg/1)
            K  = kinetic constant
            M  = biomass (mg/1)
            t  = time

     The authors points out that Si/Mt = F/M, and letting F = Si/Mt,
equation 1 becomes:

            Se = Si/(KF~1 +1)                          Equation  2

     Equation 2 clearly points out that a low F/M ratio  (high sludge
age) and low feed strength are associated with optimized ASU operations.
Removing colloids and solids in pretreatment facilities  has been demon-
strated to be an effective means to achieve dramatically better  ASU
effluent quality (1, 2, 3).

     Reasons for improved effluent quality with reduced  feed strength
can be visualized better with reference to Figure 3.  Transport  of sub-
strates through the cell membrane counter to concentration gradients is
achieved by enzyme transport systems (permeases) .  Once  inside the cell,
the substrate molecules are acted upon by a coordinated  and sequential
series of enzymes; there are several thousand endoenzymes in a single
bacterium (6) .  A small part of the enzyme sequence may  be as pictoria-
lized in Figure 4.

     Enzymic reactions are reversible.  Accumulation of  products affects
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 concentration of sub-
strate.  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 the 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, the 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  affect
the rate of reaction.  In many instances, enzyme-catalyzed reactions
                                 220

-------
appear to proceed in only one direction because the  equilibrium point is
very far in that direction.  In other  cases,  one or  more of the end
products may be removed constantly by  some mechanism so that the equi-
librium is never reached.  Under normal conditions in the  living cell
enzyme reactions are constantly pushed in this manner toward one or the
other side of the reactions.  Theoretically,  reversible reactions cannot
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 (7, 8).

     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 con-
trol called feedback inhibition.  Thus, if a  component of  the waste
substrate periodically is the same as  biological intermediate 814, the
internal concentration of 8^4 increases which increases the concen-
tration of the preceding  intermediates causing more  8^3, and Sg to be
transported externally, and less So to be transported internally; i.e.,
high residual organics are observed in the solvent phase.

     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 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. The result is,  of course, higher
residual organics in the  solvent phase.  This is called feedback repres-
sion.  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 accumulation of end products
 (feedback repression)  in  the enzyme sequence  (Figure 4).   In a sequen-
tially operating system,  the end product of each enzyme can be the
inducer of the next enzyme in the series and  the inhibitor or repressor
of  the preceding enzyme,  thus carrying forward the work of the enzyme
factory (9).

     If the principle  component of the substrate being treated is S0,
various biological intermediates (identified  as S6,  S13, S^, and S22J
may also have a propensity to be transported  externally by the enzyme
transport system.  Further,  if  the biological intermediate is three
carbon atoms or less,  it  can diffuse across the cell membrane,  inus,
                                 221

-------
with increasing concentration of So for fixed biomass,  the concentration
of internal biological intermediates will increase.   Since enzyme  equi-
libria are similar to chemical equilibria, increasing the  concentration
of internal biological intermediates (i.e., 85, 813,  8^4,  and  822)
causes their transport outside the membrane as a response  and  the  concen-
tration of external organics increases.  Starved systems (high sludge
age, low F/M) are predicted from biochemical principles to be  lowest in
residual soluble substrates.

     On the other hand, a complex substrate, So, could,result  in unex-
pectedly high residual soluble substrates by intermittently supplying a
biological intermediate of a more complex substrate which  shuts down the
degradation of the complex substrate by feedback inhibition or repression.

     Permits Operation at High Sludge Age.  Minimizing the feed strength
by removing colloids and suspended matter not only yields  a purer ef-
fluent but provides the means to respond to sludge age.  Outlined in the
schematic of Figure 5 are properties of the BOD/COD materials.  The
nonorganic and slowly bioxed materials increasingly accumulate in the
sludge mass with increasing sludge age causing deterioration of the
sludge settling properties (3). >This deterioration of sludge settling
properties is apparently caused by changing the electrical properties
(zeta potential) of the sludge which is discussed in  a later section.
ASUs can be operated easily at a very high sludge age (50+ days) if the
process is protected from colloids which are inert or only extremely
slowly biodegraded.  Successful operation at high sludge age responds to
those soluble components requiring acclimation of the biomass or having
a slow biox rate.

     The improvement in ASU effluent quality with increasing sludge age
typically is uniformly demonstrable up to about 20 days sludge age; for
operation higher than 20 days sludge age, improved effluent quality may
or may not be observed.  This appears to be related to operating tem-
perature and the nature of the substrates.  For example, with reference
to Figure 6, Curve A is typical of the ease with which organics in
municipal and many industrial effluents are removed biologically.  ASU
operation at low SA is very effective, increasing operation to 20 days
SA yields modest effluent-quality improvement and beyond 20 days, no
improvement is observable.

     At the opposite end of the spectrum are organics that yield a
removal curve like E; i.e..similar to nitrification  in a  refinery
effluent.  This curve indicates that microorganisms needed for such slow
reproducers that high sludge age operation is needed  to accumulate
sufficient appropriate organisms for organics removal.

     Curve B typifies an organic originally removed at a high rate but
which generates a biological intermediate that is removed  at a much
slower rate.  The biological intermediate is rate limiting and  increas-
ing sludge age is more effective for decreasing organics in the effluent
for this example than for an organic typified by curve A.
                                 222

-------
     Curve D is illustrative of organics whose rate limiting reaction is
probably the initial, or an early, enzyme reation in the degration
sequence and increasing sludge age increases the removal by increasing
the enzyme concentration.

     Curve C is probably representative of a sequential enzymic reaction
series wherein a series of rate limiting reactions are encountered.
Thus, increasing sludge age has the  effect of increasing the supply of
enzymes needed for the reactions in  the later stages which measures as
reduced organics in the effluent.

     As operating temperatures decrease, biological oxidation rates
decrease and more microorganisms are needed to achieve the same sub-
strate removal.  Thus, the lower the operating temperature, the more
pronounced the impact of increasing  sludge age.  In an Amoco study (1)
with parallel units operating at a median temperature of 14°C (57°F)
compared to a control unit at 20 days  sludge age, the percentage improve-
ment in effluent quality by prefiltration and increasing sludge age from
20 to 60 days is:

                         Per Cent Improvement in Effluent Quality By
                           Prefiltration and Increased Sludge Age
                                                  Prefiltration and
     Parameter           Prefiltration          Increased Sludge Age

     SOC                      9.4                      20.3
     SCOD                    20                        36.3
     NH3-N                    -                        58.7
     Phenolics                -                        29.6

     On the other hand, when operating at 85°F, Crame (2) reported
similar residual organics at sludge  ages beyond 20 days.

     Minimizes Waste Sludge.  Minimizing the feed strength by removing
colloids and suspended matter not only yields a purer effluent but
provides the means to achieve operation at very high sludge age which
provides for minimum waste biosludge production.  Bacteria in the acti-
vated sludge mass use the energy available in the substrate for two
general purposes.  As shown in Figure  3, the substrates can be used for
1) cell maintenance energy and 2) the  repair and generation of new cell
material.  The cell maintenance energy requirement gets first call on
the substrate resources.  If there is  substrate left over, the bio-
logical system proceeds to increase  in response to the available food
supply.  Thus, to minimize production  of waste sludge, it is obvious
that the ASU should have a large mass  of activated sludge relative to
the food supply; i.e., a high sludge age (SA) which is, of course, the
same as a low F/M.
                                 223

-------
     The relationship between SA, F/M, maintenance energy, and cell
generation has been known for some time (10):

     1^ = a (dF/dt) - b                            Equation 3
     SA       M

     where, SA = sludge age (days)
             a = cell yield coefficient
             b = cell maintenance energy coefficient
             M = biomass, Ibs.
             F = COD, Ibs.
             t = Time

     Using this relationship and data from Crame (2) and Grieves et al
(1) which are optimized, completely mixed ASUs operating on the soluble
substrate in refinery effluents, the following values were calculated
for the coefficients -a and b at two different temperatures:

                    Temp. (F)        a         b
     Refinery A        57          .317      .015
     Refinery B        85          .3        .03

     Comparison with literature data for various wastes suggest these
are appropriate coefficient values.

     These coefficients and equation 3 can be used to estimate the
equilibrium biomass inventory and the resultant waste sludge generation
as a function of sludge age as shown in Figure 7.  Considering the
curves for refinery B to be most typical operating conditions, it is
difficult to rationalize the prevailing practice of operating ASUs at 5-
15 days SA, a condition generating maximum waste biomass with its atten-
dant high disposal cost.  The waste sludge can be reduced by 2/3 by
operation at very high SA.  The time span to 150 days SA is included
because Amoco is operating a test unit at this condition.

     Simplifies Process Control.  The sludge age method of process
control has many advantages which are discussed in detail by Walker (11).
Operating at a very high sludge age is even simpler.  For example, at 50
days sludge age, 2 per cent of the inventory-activated sludge is wasted.
This is such a small amount that sludge wasting can be done on about a
twice/week basis.

ASU Biological Effects Achieved by the Removal of Colloids and Suspended
Solids from the Feed Stream

     Improves Flocculating Capacity of the Activated Sludge.  As shown
in Figure 8, activated sludge has a bimodal floe size distribution.  A
good flocculating and settling sludge has a preponderance of large floe
mass compared to the fine particle fraction.  The large floe mass is, in
                                 224

-------
essence,  a biopolymer analogous to a weakly anionic polyelectrolyte-type
polymer.   Thus, the large floe mass is a biopolymer flocculant  and the
fine particles represent the phase to be flocculated.  Qualitatively, -
when the  ratio of the two phases overwhelmingly favors the biopolymer as
illustrated by the solid line, a good flocculating sludge results.  When
the fine  particles predominate as illustrated by  the dotted line, their
surface area overwhelms the capacity of the biopolymer for flocculation,
and "arms and legs" (turbidity) is observed in the supernatant.

     Inert and slowly biologically-oxidized colloids and suspended
matter contribute to increasing the amount of fine particles in the
sludge mass.  Increasing SA increases their accumulation, thereby con-
tributing to the deterioration in sludge flocculating properties.
Removing  colloids and suspended matter before ASU treatment results in
an excellent activated sludge at very high sludge age.

     Improves Zeta Potential Probability Distribution of Activated Sludge
Particles.  Another way of looking at the settling properties of acti-
vated sludge and the impact of the fine particle  fraction on the settling
properties that is much more enlightening in terms of understanding why
solids are lost over an ASU final clarifier weir  is to examine the zeta
potential probability distribution of the activated sludge particles.

     Zeta potentials are typically reported as averages which is mis-
leading,  always, and particularly in the case of  activated sludge.  For
example,  the probability distribution curves in Figure 9 show the median
ZP values of a good settling sludge to be -11 mV  and a poor settling
sludge to be -12 mV.  This difference is not only difficult to measure
but is really insignificant.  What is significant is the slope of the
distribution curve and the ZP of the highest-charged particles.  If, for
example,  a ZP of -14 mV or more negative provides enough repulsive force
that the fine particles will not flocculate, fully 8 per cent of the
solids in the poor settling sludge mass resist flocculation and settling.
On the other hand, essentially all the particles  in the good settling
sludge are well below -14 mV ZP and flocculate well.  Limiting the
accumulation of colloids in the activated sludge  by effective phase
removal in pretreatment facilities is a principal means to control the
ZP probability distribution of activated sludge.  '

ACTIVATED SLUDGE UNIT DESIGN

Mechanical

     Mechanical elements that impact most strongly on the ASU are aera-
tion units and pumps—both have been observed to  be a principal cause of
process failure through shear of sludge, increasing the detritus, and
poor SVI, ZP properties.

     Aeration.  High-speed aeration and brush aerators used in mixed
liquor tanks have been observed to disrupt sludge severely causing tne
generation of colloidal fines.  This upsets the bimodal floe size distri-
bution as previously discussed, and the biopolymer component cannot

                                 225

-------
supply the flocculant capacity to completely reflocculate the detritus
into the sludge mass; i.e., the ZP probability curve slope increases
because of the colloidal detritus.  Low-speed aerators and compressed
air systems are alternates that do not have these undesirable properties.

     Pumps.  High-speed centrifugal pumps and high-pressure drops across
valves, etc., wreak havoc on recycled, activated sludge floe.  Low-speed
recessed impeller, low head pumps have been successfully applied in
systems with fairly long aeration tank retention times of 12 hours or
more.  These long retention times may have contributed to the success by
supplying reflocculation time.  An attractive sludge recycle system is
one used in the municipal sector, air-lift pumps.  Since these are low
head pumps, proper hydraulics are required and retrofit is not attrac-
tive in many existing plants.

Process

     Reactor Design.  Much has been written about completely mixed and
plug-flow reactor designs; completely mixed gaining in popularity over
the earlier plug-flow design.  Plug flow theoretically yields a slightly
better effluent on well-equalized feed streams, and completely mixed
systems are less sensitive to some toxic transient loadings.

     One area ignored is the role reactor design plays in excess sludge
generation; the plug-flow design inherently yields more excess sludge.
As an example, Lau (12) recently reported a comparison of results from a
completely mixed unit and volumetrically equally sized three-staged
reactor completely mixed unit.  Treating high-strength, readily biode-
gradable wastes at various F/M (COD/MLSS) ratios, Lau observed the
following waste sludge yields:

                    Sludge Yield, Per Cent of Yield for Single
                         Stage Unit at F/M of 0.5	
     F/M             Single Stage                 Three Stage

     0.5                 100                           150
     0.83                400                           650
     1.0                 400                           750

     These data support the previous discussion that there is a minimum
of substrate necessary to maintain cell integrity, active transport, and
other mechanical events referred to as maintenance energy.  When sub-
strate beyond that needed for maintenance energy is supplied, it is used
for cell growth.

     Additionally, in the front end of a plug flow or multiple stage
system, the microorganisms are exposed to a high concentration of sub-
strate compared to a single stage completely mixed unit.  They respond
to the increased substrate supply by generating more cell mass.  Once
the substrate energy has been committed to new cell mass by the micro-
organisms, most of the commitment is irretrievable and the result is
greater sludge production.  In aerobic stabilization of waste sludge


                                 226

-------
where some small reduction in cell mass  is  observed, probably the mono-
mers and polymers in the cell which  are  the normal  building blocks for
RNA, DNA, enzymes, peptidoglycan, polysaccharide, and  lipid formation
that can be salvaged for maintenance energy functions  accounts for the
loss in mass.

     Preferred ASU reactor design for comparatively weak refinery ef-
fluent treatment is probably two stage with the  lead reactor preferably
80+ per cent of the total reactor volume.

     Clarifier Design.  A clarifier  is usually considered to provide for
separation of activated sludge with  process design  considerations being
chiefly overflow and recycle rates.   Actually, this is short sighted,
and a clarifier design should consider in addition  to  solid flux rates
the elements of:  flocculation, capture  of  solids trapped by surface or
interfacial tension, and a stage of  reaction.  All  of  these latter
elements are provided for in a wide-well clarifier  design where the
diameter wide well is about one-half the diameter of the clarifier
(Figure 10).

     It has been well established that gently flocculating activated
sludge before clarification improves greatly the sludge settling pro-
perties.  Flocculation provides for  aggregating  the fines into the main
floe mass and taking advantage of the biopolymers naturally in the
system.  The wide-well zone provides a region to achieve f locculation,
and the improvement in sludge settling more than compensates for the
clarifier surface area reduction due to  the wide well.

CHEMICAL DESTABILIZATION REQUIREMENTS FOR OPTIMIZING THE PERFORMANCE OF
DISSOLVED AIR FLOTATION AND GRANULAR MEDIA  FILTRATION  UNITS

     Addressing water chemistry principles  determine the phase separa-
tion efficiency of the air flotation and granular media filtration
processes.  Fundamentally, maximizing phase separation efficiency by the
air flotation process requires recognizing  that  an  intrinsic property of
solids in the presence of water is a negative, electrical-surface charge
(zeta potential).  Flotation air bubbles also have  a negative zeta
potential as does the surface of granular media  in  a filter.  Maximizing
phase separation efficiency requires that these  coulombic repulsive
forces be controlled by controlling  the  properties  of  the dispersed
phase (13).

Air Flotation and Granular Media Filtration Definitions

     A proper place to start a discussion of air flotation is a defini-
tion that recognizes the principal limitation of the process; it will
not efficiently achieve the phase separation of  colloidal solids.  The
definition, therefore, should spell  out  the application more carefully
than done heretofore:
                                 227

-------
     Dissolved air flotation may be defined as clarification  of a
     suspension of flocculated material by contact with minute'
     bubbles that attach to the solids constituting the suspension
     causing the suspensions to be separated from the clarified
     water by flotation.

     The definition puts proper emphasis on the fact that the material
being separated should not be colloidal with its inherent high, repul-
sive, negative zeta potential required to maintain the colloidal state,
but a flocculated suspension.  A flocculated suspension implies a proper
chemical pretreatment consistent with the needs of a colloidal system.

     Similarly:

     Granular media filtration may be defined as clarification
     of a suspension of dispersed material by passage through a
     bed of porous media that separates and retains within the
     media the solids constituting the suspension.

Properties of Suspended Solids

     Refinery effluents from aerated lagoons are similar to surface
waters, API separator effluents, fire, and cooling water ponds, etc.,  in
that the suspended materials usually are predominately colloidal or a
combination of colloidal and very slightly flocculated suspensions. The
stability of these colloidal systems relates to the fact that the indi-
vidual particles carry like electrical charges causing their mutual
repulsion.  Except for some isolated examples, the charge on organic,
inorganic, and biocolloids  is  negative when suspended in water.  Col-
loidal destabilization by chemical treatment has the objective of neutra-
lizing or reducing the electrical charge so that mutual repulsion is
reduced to the extent that individual particles can approach each other
close enough for van der Waals and/or chemical forces to become effec-
tive.  The attractive van der Waals1 forces cause the particles to
aggregate into agglomerates which facilitate their removal by sedimen-
tation, air flotation, or filtration processes.  The surface charge on
colloidal particles may be estimated by electrophoretic, electroosmotic,
streaming, and sedimentation potential techniques.

     We have found that the electrophoretic procedures and equipment of
Riddick (14) permits the rapid determination of colloidal charge to be
made and all our investigations involved use of Zeta Meter.  Accordingly,
electrokinetic values reported herein are zeta potentials (ZP).

Electrokinetie Charge on Colloidal Particles

     Microorganisms, dispersed oil colloids, and inert suspended matter
such as inorganic sulfides, silt, coke fines, etc., that are  present in
refinery effluents are negatively charged.
                                 228

-------
     The efficiency of air flotation for phase removal depends on the
state of subdivision of the suspended matter; higher capture and removal
efficiencies are achieved with discrete particles of substantial size
and increasingly poorer efficiencies are observed with increasing col-
loidal solids fraction.  Investigation revealed that the negatively-
charged flotation microbubble was repulsed by the negatively-charged
colloidal solids such that poor bubble adherence for flotation occurred.
Consequently, the key to maximizing the effectiveness of the air flota-'
tion process was found to be essentially neutralizing or reducing the
electrical charge of the colloidal particles to eliminate the coulombic
repulsion due to like charges.

     Source of Charge.  The net electrokinetic charge, i.e., zeta poten-
tial, on colloidal particles is a result of (1) ionization, (2) ion
adsorption, and (3) ion dissolution mechanisms.

     The amino acids, proteins, and polysaccharides constituting the
surface of biocolloids, for example, acquire their charge mainly through
ionization of functional carboxyl and amino groups to give -COO" and -Nflt
ions.  The degree of ionization of these functional groups and, thereby,
the net charge on the particle depends on the pH of the solution.  The
pH at which the ZP is zero is called the isoelectric point; at a lower
pH, the ZP is positive, and at a higher pH, it is negative.

     Solids dispersed in water typically have negatively-charged surfaces
because cations have a greater tendency to become hydrated and reside in
the aqueous solution than do anions which are smaller, less hydrated,
and more polarized, thereby having the greater tendency to be adsorbed.
The net surface charge, i.e., ZP, may be acquired by the unequal adsorp-
tion of oppositely-charged ions; however, ion adsorption may be positive
or negative.

     Ionic substances may acquire a surface charge because the ions of
which they are composed dissolve unequally in solution.  In the case of
the aluminum and iron primary coagulants, for example, hydrogen and
hydroxyl ions are in equilibrium with the solid phase hydrous oxide.
With excess hydrogen ions, the surface of the solid phase is positively
charged with excess hydroxyl ion; the surface is negatively charged.
Since the concentrations of the hydrogen and hydroxide ions determine
the chart at the particle surface, they are called potential determining
ions.  This concept may be illustrated as follows:
                     IT
                     OH"
                                                              OH
                                                         Fe
                                                             OH
                                 229

-------
     As early as 1879, Helmholtz  (13) described  the  charge on colloidal
surfaces in terms of an "electric double layer"  and,  subsequently, Guoy
and Chapman (14) described a diffuse electric double layer model  that
permitted more quantitative treatment of electrokinetic  data.  Verwey
and Overbeek (15) set forth the classic work containing  the mathematical
foundation for showing that the tendency of fine particles^ to  remain
dispersed was due to the mutual repulsion of their electrical  double
layers being sufficient to overcome the van der  Waals1 attractive forces
pulling them together.  The Russian investigators, Landau  and  Derjaguin,
had much the same idea and published the electrical  double-layer  repul-
sion part in 1941.  The total treatment is now known as  the DLVO  Theory
(16).

     Electric Double Layer.  The electric double layer may be  regarded
as consisting of two regions:  (a) an inner region which may include
adsorbed ions and (b) a diffuse region in which  ions  are distributed
according to electrical forces and thermal motion.

     Stern (17) proposed a model in which the boundary of  the  inner
region (Stern layer) was located by a plane (the Stern plane)  about a
hydrated ion radius from the surface.  Adsorbed  ions  attached  to  the
surface by electrostatic and/or van der Waals' forces may  be dehydrated
in the direction of the surface.  A certain amount of solvent will also
be bound to the charged surface in addition to the adsorbed ions.  The
shear plane, therefore, is probably located farther  from the surface
than the Stern plane.  Ions with centers beyond  the  Stern  plane are
considered to be in the diffuse part of the double layer.   These  con-
cepts are illustrated in Figure 11.

     Electrokinetic potentials relate to the mobile  part of the particle,
therefore, the electrokinetic unit consists of the volume  enclosed by
the shear plane which is rather inexactly known.  The potential differ-
ence between the surface of shear and the solution is called the  zeta
potential (ZP).

Colloid Destabilization Mechanisms

     Destabilization of the waterborne suspended solids  may involve four
mechanisms:  (1) colloid entrapment or removal via the sweep floe mecha-
nism, (2) reduction in surface charge by double-layer repression, (3)
charge neutralization by adsorption, and (4) bridging by polymers.

     Colloid Entrapment.  Colloid entrapment involves chemical treatment
with comparatively massive amounts of primary coagulants;  the  amount of
coagulant used is typically so great in relation to  the  amount of col-
loidal matter that the nature of the colloidal material  is not relevant.
The amount of primary coagulant used may be 5 to 40  times  as much as is
used for charge neutralization by adsorption.  The rate  at which  the
primary coagulants form hydrous metal oxide polymers (Figure  12)  is
relatively slow and depends chiefly upon water temperature and pH.
Coupled with the high concentration used, all negatively-charged  colloidal
                                 230

-------
material is initially exposed  to  charge neutralization by  the transient
cationic species.  The polymer matrix  is 3-dimensional and voluminous as
illustrated by Figure 13, providing  for entrapment  of solids   As the
polymer contracts, freeing solvent water molecules  and settles  the
suspended solids remain enmeshed  in  the settling  floe and appear to be
swept from the water; hence, the  description of the process as a "sweep
floe" mechanism.  This destabilization mechanism  can result in the
generation of large amounts of wet alum (or  iron) sludges which are
difficult and costly to dewater.  Even though it  is by far the most
widely used mechanism for water clarification, it is not recommended
because of the sludge problem  and because the use of other mechanisms
result in significantly lower  operating and  capital costs.

     Double-Layer Repression.  Reduction in  surface charge by double-
layer repression is caused by  the presence of an  indifferent electrolyte
which in refineries is chiefly sodium  chloride from brackish water usage
or salt water ballast.  For water and  monovalent  electrolytes, the
thickness of the double layer  is  as  follows:

     Thickness of Double      NaCl Concentration       Specific Cond.,
     Layer, Angstroms           M         mg/1          Micromhos

          1,000                .00001          0.6
            600                 -            1.0
            320                .0001           6
            230                 -           10                25
            100                .001           59               115
             75                 -         100               200
             32                .01         585             1,000
             23                 -        1,000             1,900
             10                .1           -               8,800
              7.3               -       10,000            15,000

     For double-layer repression  of  colloid  surface charge in brackish
waters, the sodium ions of the indifferent electrolyte which surrounds
the colloid particles in order to electrically balance their negatively-
charged surfaces have less tendency  to diffuse away from the colloid
surface as the salinity increases.   Some salt concentration may even-
tually be reached such that the thickness of the  double layer may be
small enough that two colloids approach each other  closely enough that
van der Waals' forces cause aggregation.   An important aspect of double-
layer repression is that the quantity  of colloidal  charge is not signi-
ficantly reduced but just the  extent to which it  extends out from the
colloid surface.  This relates to the  nature of the destabilizing chemi-
cal (salt) and its mode of action; i.e.,  the sodium ions remain free in
the solvent and cause rapid dissipation of the charge as the distance
from the colloid surface increases.
                                 231

-------
     The above-mentioned principles can be visualized by  the represen-
tations of the double layer in demineralized, fresh, brackish (or  salt)
waters illustrated in Figure 14, 15, and  16 respectively.

     For demineralized water there are not many  ions available;  there-
fore, the charge on the particle surface  is not  reduced much by  adsorbed
ions in the Stern layer or counter-ions in the diffuse layers.   As a
result, the zeta potential is high and extends for a considerable dis-
tance into the solvent; i.e., the double  layer is thick.

     Fresh water, on the other hand, comparatively may contain many
salts.  As illustrated, the presence of counter-ions may lead to some
adsorption and potential drop across the  Stern layer.  The concentration
of counter-ions in the diffuse layer is much greater than the demineral-
ized water example and causes the charge  to dissipate more rapidly;
i.e., the double layer is much thinner than the  demineralized water
example.

     In the example for brackish or salt water,  the comparatively high
concentrations of sodium ions discourages their  diffusion away from the
particle surface.  The counter-ions occupying the Stern layer cause an
apparent reduction in potential but are not strongly adsorbed and,
therefore, do not permanently alter the surface  potential charge to the
much lower charge of the Stern potential  and the charge actually measured,
the zeta potential.  The high electrolyte concentration causes any
residual charge to dissipate rapidly; i.e., the  double layer  is  very
thin.

     Charge Neutralization.  Charge neutralization by adsorption of the
destabilizing chemical to the colloid is  a key mechanism for  optimizing
removal of waterborne solids from brackish waters by direct  filtration.
The colloidal charge may not only be reduced to  zero, but beyond zero,
i.e., reversed.  Charge neutralization by adsorption infers  that the
colloid-water interface is changed and, thereby, its physicochemical
properties.  It doesn't require much extension of one's imagination to
see how this destabilization mechanism can explain those cases where
optimal chemical dosages were found and overdosing resulted  in a deter-
ioration in, or failure of, direct filtration.   This phenomenon  is more
typically experienced using very low molecular weight polyelectrolytes
or surfactant-type molecules with little  bridging properties.  Some
examples of charge neutralization mechanisms are shown schematically in
Figure 17.

     Bridging.  Bridging by organic and inorganic polymers describes the
destabilization mechanism where the molecules of the added chemical
attach onto two or more colloids causing  aggregation.  There are two
kinds of bridging; polyelectrolyte bridging between dissimilarly and
similarly charged materials.  An example  of the  first kind is the
bridging of negatively-charged colloids by cationic polyelectrolyte.
Because of the coulombic atraction involved, this destabilization is not
difficult to perceive.  On the other hand, weakly anionic organic poly-
mers are negatively charged; however, they are especially useful for

                                 232

-------
aggregating and binding together  some negatively-charged aggregates into
agglomerates that resist redispersion.   Thus,  in this  instance, attrac-
tive forces of a chemical nature  seem to overcome electrostatic repul-
sion forces due to like charges.  An electrical  model  proposing a
mechanism for this seeming anomaly  has been offered by Grutsch and
Mallatt (18).  Bridging by polymers proved  to  be an important destabili-
zation mechanism for optimizing phase removal  processes.

DLVO Theory

     The DLVO Theory quantifies particle stability in  terms of energy
changes when particles approach one another.   The total energy is deter-
mined by summation of the attraction (London-van der Waals' forces) and
repulsion (overlapping of electric  double layers)  energies in terms of
interparticle distance.  The general character of the  resulting inter-
action energy-distance curve illustrates the very significant conclusions:
1) attraction will predominate at small  and large distances and 2)
repulsion may predominate at intermediate distances depending on the
actual values of the two forces.  An important purpose of the chemicals
used for destabilization is to reduce or eliminate the repulsion force
at intermediate distances so that attractive forces will predominate and
the particles will aggregate.

     These principles are readily illustrated  by the interaction energy
curves in Figure 18.  The energy  of attraction curve (Va) and the energy
repulsion curve (Vr) are summed and yield the  interaction energy curve
(Vt).  The interaction energy curve shows a repulsive  energy maximum
(Vm) which is an energy barrier to  coagulation and the formation of
stable aggregates of particles by attainment of  interparticle distances
which permit attractive forces at the primary  minimum  to react.  This
example illustrates another characteristic  feature of  these energy
curves and that is the existence  of a secondary  minimum at relatively
large interparticle distances.  If  this  secondary minimum is moderately
strong, it can give rise to a loose, easily-reversible form of floccu-
lation.  For small particles with a diameter less than about 200 A°, the
secondary minimum cannot achieve  this loose, reversible flocculation in
those cases where the energy barrier (Vm) is large enough to prevent
normal coagulation into the primary minimum.   Thus, for complete desta-
bilization of systems composed of fine particles,  a chemical approach
that responds to zeta potential charge neutralization  is required.

     A graphic illustration of how  zeta  potential charge neutralization
leads to destabilization is illustrated  in  Figure 19.  Letting VRJ. be
the energy of repulsion curve for the particle system, summation of VR1
with the energy of attraction (Va)  gives the curve for the energy of
interaction  (VT1).  This curve demonstrates that a repulsion always
exists which accounts for the stability  of  the original colloidal
system.
                                 233

-------
     The effect of adding an increment of chemical which reduces the
zeta potential by charge neutralization is illustrated by summing the
new energy of repulsion curve (Vj^) with Va which yields the energy of
interaction curve VT2«  This curve has the secondary minimum which
provides for loosely-flocculated aggregates as previously described for
Figure 18.  However, an energy barrier still exists which precludes
flocculation of particles smaller than 200 A° diameter.

     Further addition of chemical which lowers the energy of repulsion
curve to that described by VRg yields the interaction curve V^ when
summed with V^.  This interaction curve has a large secondary minimum,
no repulsion barrier, thus, all particles are coagulated into aggregates
by the attraction energy of the primary minimum.

     Double-layer repression, therefore, can improve solids removal by
direct filtration, but this mechanism does not achieve the best results
and can conceal definition of optimal chemical pretreatment to achieve
best filtration results if the interference of this destabilization
mechanism is not recognized.  Our refinery experience indicates that the
colloidal aggregates destabilized by double-layer repression appears
analogous to loose flocculation by the secondary minimum and the aggre-
gates are readily redispersed by hydraulic forces as if the net binding
forces are very weak.  A simple procedure to identify the existence of
the double-layer repression mechanism in brackish waters so that it can
be avoided and optimized chemical treatment can be achieved has been
described (19).

Destabilizing Chemicals

     Primary Coagulants.  Efficient destabilization of colloidal suspen-
sions using salts of iron and aluminum as primary coagulants must recog-
nize the properties of these primary coagulants.  The chief properties
of concern are the ZP-pH relationships and hydrolytic reactions.

     Stumm and O'Melia (20) describe the equilibrium composition of
solutions in contact with precipitated primary coagulants in the interest-
ing manner shown in Figures 20 and 21.  These diagrams are calculated
using constants for solubility and hydrolysis equilibria.  The shaded
areas, A and B, we have added in each figure are approximate operating
regions for air flotation and clarifiers by colloid entrapment (region
A) and direct filtration by charge neutralization (region B).  Both
regions are assumed to cover a pH range of 6.0 to 8.5.  The coagulant
dosage ranges from 33 to 200 mg/1 in region A and 3.3 to 20 mg/1 in
region B.  These figures are useful in the interpretation of some of our
filtration and air flotation unit results.

     With reference to Figure 21, the isoele^tric point for ferric
hydroxide coincides with the region of minimum solubility, and the
operating regions for water treatment (destabilization) yield a hydro-
lyzed, primary coagulant with a desirable positive zeta potential.
                                234

-------
     In many refinery situations, however,  it  is difficult to use this
attractive condition because the presence of sulfides and strongly
reducing conditions cause the reduction of  ferric  to ferrous iron and
the formation of mixed iron sulfides with no coagulation powers.  In
fact, in some refinery waters, the use of iron coagulants at modest
dosages may contribute to stabilizing solids rather than destabilizing
them.

     While alum has no redox or sulfide chemistry  comparable to iron,
its amphoterism and solubility pose definite limitations on alum usage.
With reference to Figure 20, a substantial  portion of operating region's
lies in the area where alum is soluble and  the predominant equilibrium
Species is negative, Al(OH)^.  In the more  acidic  part of region B,
however, the concentration of equilibrium ionic species is very much
lower and much less negative.  Considering  these data, it is not unex-
pected that investigators consistently report  optimal coagulation/
flocculation results with alum at a pH of 5-6.

     With inspection of Figure 20, one may  question why alum is effec-
tive at all for neutralizing negatively-charged colloids in the indicated
operating regions.  One approach to explaining observed performance
requires understanding that the data are equilibrium data; but before
equilibrium is reached, substantially different conditions exist.

     Alum very readily hydrolyzes to form polymers in a complex manner
not well defined.  The hydrolytic pathway and  reaction rates are affected
by pH, temperature, other ions, etc.  One hypothesized route which
includes different aluminum hydrolysis products which are known to exist
is outlined in Figure 12.  When alum is added  to water in amounts which
exceed the solubility limits, sequential kinetic reactions occur until
the ultimate precipitate is formed and the  ionic species appropriate to
the pH equilibrate with the precipitate.  The  hydrolytic reactions are
formed which are available for colloid adsorption.  The hydrolyzed
species have enhanced adsorption capabilities, possibly due to larger
size and less hydration and the presence of coordinated hydroxide groups
(20).  In solutions more alkaline than the  isoelectric point, the posi-
tively-charged polymers are transient and at equilibrium, anionic poly-
mers prevail.

     In modestly alkaline solution, the transient  positively-charged
polymers appear to contribute to destabilization of colloids.  On the
other hand, in solutions more acidic than the  isoelectric point, the
positively-charged polymers prevail at equilibrium and destabilization
of colloids may be achieved at significantly lower coagulant treatment
levels.

     In Figure 22, the zeta potentials of colloidal iron hydroxide
solutions are plotted as a function of pH.  The zeta potential decreases
in positive charge as the pH increases until the isoelectric point^is
reached at the pH of 8.3 at which the charge reverses.  In the vicinity
of the isoelectric point, the charge may vary  as indicated.  Alum has


                                235
a

-------
similar zeta potential-pH relationship as shown in Figure 23.  The zeta
potential may be negative or positive over the pH range of 7.0 to 7.8.

     Salts appear  to interfere with the coagulative powers of alum by
anion penetration of the alum polymer by chloride ion.  Anion penetra-
tion, the replacement of a coordinated group such as aquo, hydroxo, or
another anion, can be visualized with reference to Figure 12.  Chloride
ion is a highly mobile, nonhydrated, electronegative ion that at high
concentrations penetrates the hydrous aluminum oxide polymer, impairs
the olation of alum to polymers, and reduces the formation or charge of
the transient cationic alum polymers which are very important to colloid
charge reduction.

     Surfactants.  Certain substances, even when present in very low
concentrations, possess the unique property of altering the surface
energy of their solvents to an extreme degree.  Almost always, a lower-
ing rather than an increase of the surface energy is affected.  Sub-
stances or solutes possessing such properties are known as surface-
active agents or surfactants and their unique effect is known as surface
activity.

     By broad definition then, surface-active chemicals are soluble
substances whose presence in solution markedly changes the properties of
the solvent and the surfaces they contact.  They are categorized accord-
ing to the manner in which they dissociate or ionize in water and are
characterized structurally by possessing a molecular balance of a long
lipophilic, hydrocarbon "tail" and a polar, hydrophilic "head."

     Surfactants owe their physicochemical behavior to their property of
being adsorbed at the interface between liquids and gases (where they
contribute to the electrical charge on the DAF bubble) or liquid and
solid phases  (where they may contribute to the zeta potential).  Surfac-
tants tend to concentrate in an oriented manner, at the interface, in
such a way that almost entirely, they turn a majority of their hydro-
philic groups toward the more polar phase and a majority of their
lipophilic groups away from the more polar phase and, perhaps, even into
a nonpolar medium.  The surface-active molecule or ion, in a sense, acts
as sort of a bridge between two phases and makes any transition between
them less abrupt.

     There are three types of chemical surface-active agents which are
classified according to their dissociation characteristics in water.
These are:

     1.   Anionic Surfactants—Where the electrovalent and polar hydro-
          carbon group is part of the negatively-charged ion when the
          compound ionizes:

          ANIONIC        CH3(CH2)16COO~Na+
                                 236

-------
     2.    Nonionic Surfactante-Where the hydrophilic group is covalent
          and polar and which dissolves without ionization:

          NONIONIC       CH3(CH2)16COO(CH2CH20)XH

     3.    Cationic Surfactants—Where the electrovalent and polar hydro-
          carbon group is part of the positively-charged ion when the
          compound ionizes:

          CATIOKIC       CH3(CH2)17NH3+CL~

     Surfactants are powerful charge neutralizers (and charge reversers).
In the petroleum industry, anionic surfactants are used as emulsifiers
for asphalt by imparting a zeta potential on asphalt particles ranging
from -30 to -80 mV.  Cationic types impart a zeta potential ranging from
+18 to 128 mV.  Each surfactant possesses a distinct characteristic
capability of imparting quantitatively to asphalt during emulsification,
a specific zeta potential.

     Surfactants have not found wide use for destabilizing colloidal
systems.  In fact, they are an important cause for the existence of
colloidal systems, particularly in primary municipal effluents.  The
principal organic colloidal destabilizing chemicals are polyelectrolytes.

     Polyelectrolytes.  Polyelectrolytes used as water-treating chemi-
cals are macromolecules having many charged groups and may be classified
as cationic, anionic, and nonionic depending upon the residual charge on
the polymer in solution.  Examples of the structural types are shown in
Table 1.

     In solution, the polyelectrolytes are dissociated into polyvalent
macroions and a large number of small ions of opposite charge (counter
ions).   The macroion is highly charged, which is the cause for the
characteristic properties of the polyelectrolytes.  Most of the macro-
ions are long, flexible chains, their size and shape depending on the
macroion charge and interaction with counter ions.  With increasing
charge,  the macroion extends; with decreasing charge, the macroion
assumes  a contracted random coil.  The source of the charge is illu-
strated  by the polyacrylates, a widely-used polymer.  In distilled
water, polyacrylic acid's carboxylic functional group is only slightly
dissociated.  The addition of NaOH reacts with the carboxylic acid
groups causing them to dissociate leaving a charge on the macroion and
producing sodium counter-ions as shown in Figure 24.

     The Dimensions Involved.  The dimensions of the various components
involved in colloid destabilization vary a million fold, from a few A°
to more  than 10 A° as shown in Table 2.
                                 237

-------
     Where color is not a significant factor, the problem  is usually one
of causing colloidal particles down to about 1000 A°  in diameter  to
aggregate.  When a clarifier or DAF is used for phase separation, it is
desirable to build aggregates to fairly large size, say greater than 10^
A°.  On the other hand, when filters are used, simply destabilizing the
colloidal particles is sufficient because the destabilized particles
will build aggregates in the filter bed as the destabilized suspension
passes through the media and the particles impinge and adhere to  the
media or trapped suspended matter.

Systematic Approach to Determining Chemical Treatment Requirement

     Broad experience in refinery effluent treatment  led to outlining
the condition response schematic for chemical treatment of waterborne
colloids shown in Figure 25.  In phase removal by filtration, or even
DAF, we are not concerned with, and indeed it is desirable to avoid, the
use of (1) the "sweep floe" of colloid entrapment and (2) the double-
layer repression mechanisms for colloid destabilization.  Destabilization
efforts must focus on the charge neutralization and bridging mechanisms.
Charge neutralization correlated with plant performance as the optimum
destabilization mechanism.  For plant control  of  direct filtration,
charge neutralization has been the key test parameter  correlating with
performance of refinery filters.  Brackish water required that charge
neutralization be measured after dilution with distilled water to sepa-
rate the effects of double-layer repression and charge neutralization;
i.e., under plant conditions of high salinity, the addition of desta-
bilization chemicals could reduce the ZP to approximately zero by a
range of chemical treatments; however, when double-layer repression was
the cause of reduced ZP, reduced filter run lengths and performance were
observed.  Reducing the ZP to approximately zero, as  measured by means
responsive to charge neutralization, point out more definitively the
required destabilization chemical treatment and resulted in optimum
filter performance.

     Waterborne colloids subject to chemical destabilization and phase
separation fall into two general categories:  relatively inert substances
such as clays, sand, and organic materials; and microorganisms or bio-
colloids.  Both categories of colloidal matter may be stabilized because
they are charged and/or are highly hydrophilic.  Both categories of
colloidal matter also may vary in response to treatment by destabili-
zation chemicals and within each category, the state  of subdivision
seems to require additional consideration; i.e., extremely small col-
loidal particles are sometimes more difficult to aggregate for removal
by filtration.  Typically, destabilization of biocolloids, such as are
in aerated lagoon effluents, is a more demanding problem.

     In the case of polyelectrolytes, some counter ions at high con-
centrations screen the charged functional groups with an ionic cloud as
previously described.  Salinity, hydroxide, phenolics, sulfides,  etc.,
are examples of the kinds of counter ions found to affect various cat-
ionics.  Each waste water application of cationics must address the
contaminants present if the most cost-effective polyelectrolyte is to be

                                238

-------
used (21).

PROPOSED CHEMICAL AND BIOLOGICAL DESTABILIZATION MODELS

     Anionic polyelectrolytes are frequently used to flocculate nega-
tively-charged colloidal systems.  Even better acknowledged is the
capability of the activated sludge process  to flocculate negatively-
charged sewage and industrial colloidal systems; an extreme case being
the Contact Stabilization variation  of the  activated sludge process.
Because the activated sludge mass and the anionic polyelectrolytes are
both negatively charged, both of these destabilization examples seem to
be counter to theory.  A possible explanation for these extremely valu-
able properties lies in their electrical characteristics and the environ-
ment in which they operate.

Electrical Characteristics of Polyelectrolytes

     As discussed previously ionizable groups on the polyelectrolyte are
the source of an electric field.  Neglecting the effect of counter ions,
the field about an extended polyelectrolyte is shown qualitatively in
Figure 26.  In this example, there are potential maxima in the region of
the charged functional groups.  There is a  lesser potential field,
outside the region of potential maxima, that might be described as a
"potential tunnel."  This is illustrated isometrically in Figure 27.
When the polyelectrolyte is in the random coil conformation as shown in
Figure 28, there is an additional weak potential region (B) in the
polyelectrolyte's sphere of influence.  A fourth potential region is the
solution where there is no potential effect due to the polyelectrolyte.

     Each potential region has a different  effect on counter ions.  In
the three potential regions within the polyelectrolyte's sphere of
influence, counter ions can be considered as being bound to the poly-
electrolyte.  In the region of potential maxima, the bound counter ion
may be localized at charged functional groups forming ion pairs.  In the
potential tunnel region, the bound counter  ions are mobile, as they are
in the weak potential region.  Mobile counter ions establish an equili-
bria, therefore, between the potential tunnel, weak potential field in
the sphere of influence, and the solution.  An especially unique
property of polyelectrolytes ±s that the bound counter ions in the
potentialTunnel area can move parallel _t£  the polyelectrolyte molecule
in the apparent volume occupied by the potential tunnel; thus, counter
ions can "flow" in the potential tunnel areas of Figures 27 and 28
analogous to water in a garden hose.  Polyelectrolyte solutions, there-
fore, show an extremely large dielectric constant.

     The dielectric constant or polarizability of polyelectrolytes is
determined by the volume of polyelectrolyte in which counter ions are
retained (not by the charge density  of the  polyelectrolyte) as long as a
region of bound but mobile counter ions is  formed.  The dielectric
increment depends on the geometry of the polyelectrolyte; extended
polyelectrolytes give much larger dielectric increments than polyelec-
trolytes in the random coil conformation.   Further, the dielectric

                                239

-------
increment of polyelectrolytes in the extended (or helix) conformation
increases rapidly with increasing length.  Uniform and continuous dis-
tribution of charged groups and the mobility of bound counter ions are
essential for large polarizability (22).

Polyelectrolyte Destabilization Model

     The means for negatively-charged polyelectrolytes to destabilize
negatively-charged colloids is hypothesized to lie in the electrical
properties of the polyelectrolyte.  The proposed model in Figure 29
illustrates, approximately to scale, the colloid, its electrical double
layer, the anionic polyelectrolyte, and its potential tunnel.  Counter
ions are cations in both cases.

     Cations in the double layer are subject to at least two opposing
electrical forces, coulombic attraction to the colloid particle and,
also, to the solvent to maintain an electrically-balanced system.

     When the sphere of influence of the anionic polyelectrolyte approaches
the electrical double layer of the colloid, repulsion due to encountering
like charges might be expected.  However, in this instance, apparently:

     1.   The polyelectrolyte's electrical sphere of influence shields
          the counter ions in a localized area of the colloid double
          layer from the attractive, electrical solvent forces.

     2.   The resulting electrical imbalance results in an increase in
          the negative coulombic potential in the localized area on the
          colloid.

     3.   The increase in negative coulombic potential attracts the
          mobile counter ions from the potential tunnel region of the
          polyelectrolyte.

     4.   The potential tunnel of the polyelectrolyte serves as a conduit
          for counter ions which neutralize the surface charge in the
          localized area of the colloid surface.

     5.   The charge neutralization achieved reduces the energy of
          repulsion at the localized site sufficiently so that the sum
          of the energy of repulsion and energy of attraction curves
          yields an interaction energy of attraction at the localized
          site (Figure 18).

     6.   The positive interaction energy at the localized site permit
          attractive London-van der Waals' forces at the primary minimum
          between the polyelectrolyte and colloid to react.

     The key to this hypothetical model is the mechanism by which the
anionic polyelectrolytes implement the initial shielding action of the
colloid surface from the electrical solvent forces.  This capability
lies in the special properties of the polyelectrolyte:  polarizability,

                                 240

-------
bound but mobile counter ions ,in  the  potential tunnel, and equilibria
between counter ions in the potential tunnel  and  sphere of influence
established by coulombic forces.

     Repulsion due to like coulombic  charge is illustrated in Figure 30.
In Figure 30A, the double layer overlap  for energy  of repulsion curves A
(polyelectrolyte) and B (colloid) result in a net repulsion indicated by
curve C.  If this were a system of lyophobic  colloids, it would be a
stable colloidal system.  Polyelectrolytes are extremely polarizable,
however, particularly the high molecular weight anionics.  The polarizing
polyelectrolyte can supply an essentially infinite  quantity of counter
ions via its potential tunnel to  a localized  site near a colloid particle.
This influx of counter ions reduces the  repulsion energy curve A by
double-layer repression because of the equilibria established by the
bound mobile counter ions, and the surfaces can approach more closely as
shown in Figure SOB.  The influx  of counter ions  to the localized charged
region between the polyelectrolyte and colloid surface also would be
expected to reduce the energy of  repulsion (curve B) for the colloid in
the localized region.  The distance between the charged surfaces can be
diminished further for this additional reason as  illustrated in Figure
30C.

     Interaction between the two  surfaces is  achieved at localized sites
if the counter ions reduce the energy of repulsion  curves such that the
energy of interaction curve provides  for a secondary minimum (see Figure
18).  Particles larger than about 200 A° are  flocculated.  Further
reduction in the energy barrier of the repulsive  energy maximum permits
flocculation of the particles less than  200 A°.

     Sterically, when the two surfaces approach closely as illustrated
in Figure 30C, the polyelectrolyte provides a shielding or insulating
effect to the localized colloid site  from the coulombic solution forces.
Negating the effect of solution forces results in an increase in attrac-
tion of counter ions to be localized  colloid  site and contributes further
to polyelectrolyte polarization and double-layer  charge repression at
the site.

Electrical Characteristics of Microorganisms

     The microorganisms constituting  activated sludge also have a nega-
tive zeta potential.  Glucuronic  acid has been proposed as the source of
the electrical charge.

     As shown in Figure 31, the outermost surface of most bacteria is a
slimy capsule varying in thickness up to 100,000+ A°.  The capsule com-
position varies with species and  may  consist  of polymers of glucose or
other sugars, amino and acid sugars,  or  polypeptides.  Capsules generally
consist of about 98 per cent water.   The function of the capsular layer
is proposed as serving the microorganisms as  an osmotic barrier, i.e., a
mechanism for guarding against too rapid an influx  or efflux of water
(23).  We propose that the capsular layer performs  an equally if not
more important function of providing  the potential  tunnel region.

                                 241

-------
     The zeta potential is the charge difference between the plane of
shear and the solution.  In the case of microorganisms the shear plane
is some small, but indeterminate distance beyond the outer boundary of
the capsule.  It is hypothesized that the region beyond the shear plane
is analogous to the diffuse layer of counter ions in the colloid double-
layer model (Figure 11), whereas the capsule on the capsule side of the
shear plane serves a function analogous to the potential tunnel region
of polyelectrolytes; i.e., it provides for a region of bound, but mobile
counter ions.  The acidic polysaccharides in the outer layer of the soft
layer are at least partially responsible for the negative, electric
charge on the bacteria surface.  As in the polyelectrolyte example,
mobile counter ions establish an equilibria between the capsule layer
(potential tunnel) and the solution and the bound counter ions in the
capsule layer can move parallel to the cell wall of the microorganism.
The chief difference between the polyelectrolyte potential tunnel and
the biological potential tunnel (capsule) is the comparatively large
volume of the capsule-a desirable property.

Biological Destabilization Model

     As for polyelectrolytes, the means for negatively-charged micro-
organisms to destabilize negatively-charged colloids also is hypothe-
sized to lie in the electrical properties of the microorganisms.  The
model proposed for polyelectrolytes (Figure 29) is analogous to the
model for the biological system.  Once again, cations are counter ions
in both cases.  Dimensionally, the activated sludge floe and the capsule
volume are larger than the comparative volumes occupied by the poly-
electrolyte and its potential tunnel at the very low concentrations of
polyelectrolyte used (frequently less than 1 mg/1).  As the diffuse
counter ion layer of the double layers overlap (Figure 30A), compared to
polyelectrolytes, the massive size of the microorganisms floe more
readily shields the counter ions in the colloid double layer from
coulombic solvent forces.  The microorganisms supply counter ions to
satisfy the electrical imbalance from their potential tunnel (capsule).
The localized increase in counter ions reduces the energy of repulsion
curve analogous to anionic polyelectrolytes and the particles flocculate.

     The new technology, which is the subject of this article,
     forms the basis for a number of U.S. and foreign patent
     applications.  Williams Brothers Waste Controls, Incorporated,
     of Tulsa, Oklahoma, has been licensed to employ this tech-
     nology and to sublicense others throughout the world.
                                 242

-------
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, blend-
ing 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 & 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
 Chemists, Chief Process  Engineering
 and Technical Services Superintendent
 of the Amoco Oil refinery at Whiting,
 Indiana.
                                 243

-------
                         REFERENCES
 1)   Grieves,  C.  G.,  Stenstrom,  M.  K.,  Walk, J. D., and Grutsch, J.  F.,
     "Effluent Quality Improvement  by Powdered Activated Carbon in
     Refinery Activated Sludge Process," (Preprint 28-77), API 42nd  Mid-
     Year Refinery Meeting,  May 11, 1977.

 2)   Crame, L. W., "API Bioenhancement  Study," ibid.

 3)   Grutsch,  J.  F. and Mallatt, R. C., "A New Perspective on the Role
     of the Activated Sludge Process and Ancillary Facilities," First
     Open Forum on Management of Petroleum Refinery Wastewaters, USEPA,
     API, NPRA, University of Tulsa, Tulsa, Oklahoma.

 4)   Eckenfelder, W.  W., "Activated Sludge Treatment of Petroleum Refinery
     Wastewaters-An Overview," ibid.

 5)   Adams, C. E., Eckenfelder,  W.  W.,  and Hovious, J.  C., "A Kinetic
     Model for Design of Completely Mixed  Activated Sludge Treating
     Variable Strength Industrial Wastewaters," Water Research, Vol. 9,
     pp. 37 (1975).

 6)   Dickerson, R. E., Geis, I., The Structure and Action of Proteins,
     Harper and Row,  New York (1969), p. vii.

 7)   Frobisher, et al, Fundamentals of  Microbiology, W. B. Saunders
     Company,  Philadelphia,  Pennsylvania (1974), p. 92.

 8)   Aiba, S.; Humphrey, A.  E.,  and Millis, N. F., Biochemical Engineer-
     ing, Academic Press, New York  (1965), pp. 38, 39.

 9)   Lehninger, A. L., "Energy Transformation in the Cell," Scientific
     American, (May,  1960).

10)   Heukelekian, H.  Oxford, Manganelli, "Factors Affecting the Quantity
     of Sludge Production in the Activated Sludge Process," Sewage
     and Ind.  Wastes. Vol. 23, p. 945 (1951).

11)   Walker, L. F., "Hydraulically  Controlling Solids Retention Time in
     the Activated Sludge Process," Journal WPCF. Vol.  43, No. 1, p. 30
     (1971).

12)   Lau, C. M.,  "Staying Aeration  for  High, Efficient  Treatment of
     Aromatic Acids Plant Wastewater,"  32nd Industrial  Waste Conference,
     Purdue (1977).

13)   Grutsch,  J.  F^and Mallatt, R. C., "Optimizing Granular Media
     Filtration of Refinery Waters  Requires that the Electrical Charge
     Phenomena Operative be Addressed," Chemical Engineering Progress,
     Vol. 73,  No. 4,  p. 57,  (1977).
                                244

-------
14)   Riddick,  T.  M., Control of Colloid Stability Through Zeta Poteni-ial
     Livingston Publishing Company, (1968).               "	'

15)   Verwey,  E. J. W. and Overbeek, J. Th. G., Theory of the Stability
     of Lyophobic Colloids, Elsevier, Amsterdam, (1948).

16)   Overbeek, J. Th. G., Colloid Science, Vol. II, Elsevier, Amsterdam,
     1952.

17)   Stern, 0., Z. Elektrochem, Vol. 30, p. 508, (1924).

18)   Grutsch, J. F. and Mallatt, R. C., "Electrochemistry of Destabili-
     zation," Hydrocarbon Processing, Vol. 55, No. 5, p. 221, (1976).

19)   Grutsch, J. F., Mallatt, R. C., and Peters, A. W., "Chemical Coagu-
     lation/Mixed-Media Filtration of Aerated Lagoon Effluent," EPA-
     660/2-75-025, June, 1975.

20)  Stumm, W. and O'Melia, C. R., Journal, AWWA, Vol. 60, p. 514,
     (1968).

21)  Grutsch, J. F. and Mallatt, R. C., "Approach to Chemical Treatment,"
     Hydrocarbon Processing. Vol. 55, No.  6, p. 115, (1976).

22)  Oosawa, F., Polyelectrolytes, Marcel  Dekker, Incorporated, New
     York, (1971).
                                  245

-------
DISCUSSION

Milton Beychok, Consulting Engineer;  Isn't the optimum sludge age for the
PAC addition related to equilibrium capacity of carbon? Why recycle spent
carbon?

Grutsch;  It doesn't seem to work that way, Milton. It seems to work as a
sequential operation with some bioregeneration of the PAC.  First, the
biological organisms appear to reduce the TOC/COD to a low level, and second,
the carbon then takes over and reduces the residual TOC/COD to even lower
levels as a direct function of carbon surface area. Bioregeneration of carbon
surface seems to be indicated.  For example, at soluble TOC levels of 25 mg/1
in an activated sludge control unit, a parallel unit with 2,500 mg/1 PAC sees
25/2,500 or 0.01 Ib. TOC/lb. PAC at start-up, but at  100 days sludge age the
equilibrium exposure is 1 Ib.  TOC/lb. PAC at 24  hours retention time and
increases with dilution rate. To achieve TOC reductions at these loadings in
this concentration range is not expected, therefore, some of the PAC surface
area must be bioregenerated to supply more surface area which, of course,
supplies more driving force for TOC reductions.

Experiments with active carbons with different surface areas indicates
equilibrium carbon surface  area in the activated sludge unit determines
effluent quality (at fixed activated sludge unit operating conditions);  the
larger the surface area, the lower the concentration of residual substrates.

George Reid, O.U. Professor:  When you go to a long-term retention time,
don't you increase the capital and operating cost?

Grutsch:  If by "long-term retention time" you mean high sludge age—the
answer for low and moderate strength wastewater typical of refineries is no,
you actually can decrease costs. An extended discussion of the advantages
of high sludge age activated sludge units with their required pretreatment
for reducing capital and operating costs is included in our paper in the
Proceedings of the First Open Forum on Management of Petroleum Refinery
Wastewaters.

Reid:  Unless I misunderstood, is there a possibility of using a low sludge
age system?

Grutsch:  I think that is the wrong way to go. It  is the wrong way to go
because low sludge age systems maximize the conversion of the organic sub-
strate to biological solids (see Fig. 7).  You don't want to do that because of
the dewatering and disposal costs. You want to remove them chemically, since
the art is available to remove them chemically.  And second, a front end
biological system such as a  trickling filter does not remove the troublesome
                                  246

-------
colloidal fraction that ruins the activated sludge electrical properties so that
you can cycle up to high sludge age.  I think chemical engineering principles
properly applied here give maximum returns in better effluent quality at lower
operating and capital costs.

Bob Carloni, Lion Oil Company;  Do I understand from your discussion that
to flocculate a negatively charged colloid, one should use an anionic polyelec-
trolyte in preference to a cationic?

Grutsch:  I don't want you to close your mind to that possibility.  For example,
you can use a two-chemical or a one-chemical system. With the proper
facilities, a two-chemical system may be cheaper. Using a two-chemical
system in a municipal plant, the addition of an  anionic polyelectrolyte at
0.3 to 0.5 mg/1 to the influent of the primary clarifier dramatically improves
the removal of most of the suspended matter. The residual colloidal material
can then be destabilized more economically by cationic polyelectrolyte charge
neutralization for essentially complete removal  in a following DAF or granular
media filter. By contrast, a one-chemical system involves use of only a
cationic polyelectrolyte  for charge neutralization and bridging.  One-chemical
systems may be more costly in chemical requirement than a two-chemical
system.

John Penniman, Pen Kern, Incorporated:  The paper industry, using the
dissolved air flotation process under careful zeta potential  control,  has
reduced particulate concentration from a save-all to well under 10 ppm.
Bearing in  mind that zeta potential can fluctuate in refinery effluent, even
after equalization, from-6mV to-18mV, and that automatic zeta potential
instrumentation is now available, would you  speculate on its applicability to
improving  the  reliability and economics of water clarification.

Grutsch:  Well, that  is hard for me to put into perspective. We like to follow
the KISS principle; that is, "keep it simple, stupid!" So we don't like to put
too big a demand on the plant personnel.  For example, in chemical addition
we would like the plant  operator only to look at the chemical addition pump to
determine if the pump is running and pumping, and  that is all he has to
concern himself about.

So we use the zeta potential determinations to screen various poly electrolytes
and look for a polyelectrolyte that is insensitive to the system that we are
trying to destabilize. Then we try to pick a treatment concentration where the
suspended  solids or  pH variation doesn't significantly change the chemical
dosage. There will be sensitive systems that might use feed-back control.
Of course,  if the chemical prices increase inordinately, we will have to get a
little bit more conservative in our use and that might pay off a feed-back
control system.
                                   247

-------
Is)
                          TABLE 1.   EXAMPLES OF CATIONIC,  NONIONIC AND ANIONIC  POLYELECTROLYTES

                Polyelectrolyte
                  Description      Structural Type     Functional  Group                      fficnole
H
1
Catloolc Anlnes — N — R
R
R
1
Quaternary — H— R
R

0
Nonlonlc Polyamide — C — NHg

Polyalcohol —OH

•(— CHg— CH— )*•
CH,

*•(— CHg— • CH— •)*•
0*0
IH,,
•*• ( — CHg — CH— )*•
OH
R
^^-K
CacO
0
— (— CHg — CH— ) •*

Rydrochlorlde
Poly(n-»ethyl-'*-
ylnyl pyrldlaiuB
chloride)


Polyacrylaalde

Polyrlqylalcohol

Poly(Mth)acryllc
Acid
Polyrliqrlaulfaoat*
                                                                                I _

-------
    TABLE 2.   DIMENSIONS INVOLVED IN COLLOID DESTABILIZATION
A.   SOME COLLOIDAL SYSTEMS                  DIAMETER. ANGSTROMS

     COLOR BODIES                            50 - 1000
     INERT COLLOIDS (CLAY, SILT,
       INORGANIC SALTS, ETC.)                1,000 - 30,000
     EMULSIONS                               2,000 - 100,000
     BACTERIA                                5,000 - 100,000
     ALGAE                                   50,000 - 8,000,000

B.   CATIONS

     Na+                                     1.9
     Ca++                                    2
C.   POLYELECTROLYTES

     POTENTIAL TUNNEL                         7-11
     CHAIN LENGTH, 100,000 -  15,000,000
       M.W.                                   250,000 - 40,000,000

D.   ELECTRICAL DOUBLE LAYER

     RANGE OF EXPECTED VALUES                5-100
     EXPECTED TYPICAL IN REFINERY             30

E.   SOLVENT

     H20
                               249

-------
         2000
                            ORaw Waste Load API Separator
                            • Waste Load after Intermediate Treatment
                                                                                       O
Ui
O
     -o
      I
      o>
     O
      OJ

      X
     O
1000
 900
 800
 700
 600

 500

 400
          300
      CD

      O
          200
          100
         I	I
             I     I
             0.5  1
                         10     20   30  40  50  60   70  80
                             % Probability less than Indicated Value
90    95
98  99 99.5
                        Fig. 1 Protoablli-ty Plots of COD 'Before and Af-tep In-fcermeaia-te Treatment

-------
300
                                                    315"
JO
(J1
          200
        T3
        C
        CO

        E
c
CD
O)

X
O

15
o

E
0)

CJ
           100
           50
                        Before
                        After
                        195
                         53
                                         110—
                  I    I   I   I   I  I  I  I   I  I    I   I   I
                        10         50         90 95 98

                    % Probability Less Than Indicated Value
        Fig. 2  Primary Municipal  Effluent - Before
                And After Filtration (WPCF, V.48 #7, p.1801)

-------
     Soluble Substrates
              Soluble Substrates
              from Actions of
              Exoenzymes    ,
                                                                        Cell
                                                                          mbrane


                                                                         Cell Wall
                                             I
CO2 and Other
Metabolic By-Products
Fig. 3   Schematic of a Bacterial Cell, Its Biochemical Activities,
         and Exoenzyme Solubilization of Insoluble Substrates
                                         252

-------
NO
I
1
1
1
I
1
1
EB ^ 0 £13 ^ 0 £12 ^
1 i3
i V
1 4
Ec^ 1

T*^" Sl4 -^^™^^^~ Sl5 -^*— •" etc-
\
\
\
1
^^"^x Cell Membrane
1E' JE7
-E* -** ^

L
E« *0
^ ^ 810
! /
" £20 £22 /


/
/
*l
Cell Membrane ^




Eo






          So, Si, 82, etc. — Organic molecules
          EA, EB, EC, etc. — Enzyme transport systems
          £1, £2, Ea, etc. — Endoenzymes
          Fig. 4  Sequential Endoenzyme Biological Oxidation Showing
                 Permease Active Transport

-------
                           Total
                           BOD/COD
             Soluble
             BOD/COD
       Difficult to
       Biodegrade
Readily
Biodegradable
 Requires
 Acclimation
                   Colloidal &
                   Susp. BOD/COD
Fig. 5  Impacts of BOD/COD Components On
        Activated Sludge Unit
                            254

-------
c
0)

i
i
E
(0
0)
c

'w
(0

£
u
                   10
 15



Days
20
25
Fig. 6  Biological Removal Of Contaminants With

      Time-Types Of Removal Curves
                        255

-------
to
Ui
             13,000 Ibs COD/D
             For Refinery A: Alpha
             For Refinery B; Alpha
015;°F= 57
03 ; °F = 85
.317; Beta
.3  ; Beta
   £ 2000

        0
                      50           100
                     Sludge Age (Days)
   Fig. 7  Activated Sludge Unit Waste Sludge
          Generation and Equilibrium Biomass
          Inventory Dependence on Sludge Age

-------
Fraction of Particles
 |_ Inert Colloids-]^
     Emulsions
                Flocculated
              .-Sheared
       10*    105
Particle Size, Am stroms
10
Fig. 8  Activated Sludge Bimodal  Floe Size
         Distribution
                        257

-------
       I     II     III
      Zeta Potential, mV
       2    5   10             50             90  95

       % Probability Equal to, or less than, Indicated Value
Fig. 9  Zeta Potential - Probability Curves for Poor and Good
        Settling Activated Sludges
                             258

-------
Fig. 10  Wide Well Clarifier
                                            Scum

                      259

-------
               0
                          ©
                  0

                                   ©
                                          ©
      Stern Potential -\
     Nernst Potential
       Zeta Potential
                                      Bulk of Solution
          Particle Surface
       Stern (Rigid) Layer—'
             Shear Plane
     Diffuse Layer of Counter-ions-*
Fig. 11  Representation of Electric Double Layer
         According to Stern's Theory
                         260

-------
                [AL(H20)6
                         3
                                                                                                  OH
Ni
[AL(H20)5(OH>]
                [(H20){OH) Al
               .OH^       OH,
                     Al
,3,«..,,->, ^        „,         Al         AI(OH)(H2O}3]
           OH^     OH^     OH^
                                OH,/ ^OH
                HH20)2(OH)2AI '       lAlx.      lAI(OH)3(H20)]
                                OH"\    OH
[(H20)2(OH)2AI
                                  .
                                OH
                                                AI(OH)3(H20)]
                               •°H<  ^°"\
                l(H20}(OH)3M -^     (  Al        JAI(OH)3(H20)]
                                            "7.
                                                                                       [(H20)4AI          AI(H20)4]
                                                                                                               +4
                                                                               OH"
                [For solid phase see isometric in following figure]
                                                                                             OH"
                                                                                      [(H20)3«OH)AI        ^AI(OH)2(H20)2
                                                                                                     OH
                 Fig.12 Sequential Formation of Hydrous Aluminum Oxide Polymers

-------
                                                                               o
NJ
O>
N>
           H20
                                                                           OH2
           Fig. 13  Example of Complex which may exist in Precipitated
                   Hydrous Aluminum Oxide Polymers

-------
  Particle Surface
    Stern Plane
      Surface of Shear
                Diffuse Layer
                Stern Layer
                Surface Potentia
                Stern Potential
                Zeta Potential
             Distance
Fig.  14 Representation of Electric Double
        Layer in Demineralized Water
                     263

-------
  Particle Surface

   Stern Plane

     ^Surface of Shear
  +
 $
10
       ©   dPn
        ooTo;;
       00^00
p
L ,



©0Gp°
1
>i
i
" •
Diffuse Layer
t '
Stern Layer
ourrace roiennai
• .
1
l
I
1
1
1
l
1
Q+Arrt Do+^n+iol
          -Zeta Potential
             l
           Distance
Fig. 16 Representation of Electric Double Layer

        in Brackish Water
                      264

-------
 'article Surface
  pStern Plane
  \ ^Surface of Shear

     ©     &  ©


•"^0eo
                       ©
                         o
             Diffuse

             Stern
             Surface Potential
           Distance
                              O
                                ©
Fig. 15  Representation of Electric

         Double Layer in Fresh Water
                265

-------
 No Specific Adsorption
Specific Adsorption of
Cations
                                                          Specific Adsorption of
                                                          Anions
Strong Specific Adsorption
of Cations

A   Rigid Layer Migrating with Particle as Single Kinetic Unit
B   Slipping or Frictional Boundary Layer of Ions
X   Distance from an Arbitrary Point Inside Solid Phase
i//   Electrical Potential
f    Zeta or Electrophoretic Potential
Fig.  17   Distribution of Ions and Potentials in the Double Layer
          Surrounding Colloids (After Overbeek 8)
                                             266

-------
          Primary
         .Minimum
          Secondary
          Minimum

VA - Energy of Attraction

V  - Energy of Repulsion
 R
VT - Energy of Interaction

VM - Energy Barrier

D  - Distance Between
    Charged Surfaces
Fig. 18 Potential Energy of Interaction
        for Two Charged Surfaces
                     267

-------
                   VA - Energy of Attraction
                   VR1" VR2' VR3' Decreasing
                      Energies of Repulsion
                   VT1- VT2-VT3- Interaction
                      Energies for VR1, VR2 & VR3
                   D - Distance Between
                      Charged Surfaces
Fig. 19  Influence of Electrolytes
         on Interaction  Energy
                      268

-------
NJ
O>
VO
            O
           -2
           -4
           -6
       o
D)
O
           -8
          -10
           -12
                        AI(OH)3 (s)
                       ///>'//*
                       \\VtK\\
i\
 \^\
M\ J-.V-
A - Operating Region for Air
  Flotation and Clarif iers
B - Operating Region for
  Direct Filtration
C - AI(OH)~

D - AI13(OH)|+
       A-L.
E - Al
    /
    3+
                                 pH
                               8
                  10
              12
Fig.
           20  Equilibrium Compositions of Solutions in Contact
              withAI(OH)3

-------
N)
-~J
O
                           u
                           -2
                     1     -6
-8
                          10
                          -12
               A - Operating Region for Air Flotation and Clarif iers
               B - Operating Region for Direct Filtration
               C-Fe(OH)4     F-Fe3+
               D - Fe(OH)2     Q . pe
               E - Fe(OH)2+
                                                  \\\A\\\
                                                  V//.6///A
                                                        Fe(OH)o (s)
                                    \G
                                            \   \  D
                                                  6   PH    8
                                            10
                                             i     i
12
                     Fig. 21  Equilibrium Compositions of Solutions in Contact
                             with Fe(OH)3

-------
bo
           +40



           +30




           +20


        >

        J  +10
.2
'+5

Q)
4-»
o
Q_

(0
+j
CD

N
  0




-10



-20



-30



-40
                                     Data Scatter
                    I	I    I     I
                             5   6
9    10
                            Iso Electric Point pH = 8.3
        Fig. 22  Zeta Potential of Colloidal Iron Hydroxide

                  Solutions Plotted As A Function of pH

-------
+10
  0
-10
        I     I     I     I
I
        3456789    10
                         pH
Fig. 23  Zeta Potential — pH Plot for Aluminum
                     Hydroxide
                     272

-------
           — C — C— + NaOH-*—C —
N>
•xl
OJ
H
1
— C—
1
C
0


H
1
C —
1
C
/\
r \
o
1
H
H
1
C
1
H



H
1
— C —
1
C
//\
H \
O O
1
H
                LJX
H
1
— C
1
H




H
1
— C —
1
C
/A
0 O
e
X*— N.
(Na+)
H
1
C
1
H




                                 H
O O
                                  e
     Fig. 24 Dissociation Of Polyacrylic Acid
            By NaOH

-------
                                            Waterborne
                                          Suspended Solids
                 1) Charge
                Neutralization
                 2) Bridging
                                         1) "Sweep Floe"
                                            Mechanism
                                         2) Double Layer
                                            Repression
              Inert
              Solids
                 Biological
               Cell Material
                 Cold Water
                                          Warm Water
   Fresh Water
Brackish Water
Fresh Water
Brackish Water
Fig. 25   Condition—Response Flow Schematic for
         Chemical Treatment of Waterborne Colloids
                                             274

-------
                                                          Potential Along Dashed Profile
 Extended Macroion
A      B
Potential
A = Solution Potential; B = Weak Potential Field;
C =• Potential Tunnel; and D = Potential Maxima
Fig. 26   An Extended Macroion and Its Associated Potential Field
                                                 275

-------
ho
~J
CT>
                  Charged Functional

                  Groups on Macroion
                  Potential

                  Maxima
                  Potential

                  Tunnel
                  Fig. 27   Isometric Showing Potential Maxima and Potential Tunnel Fields

-------
A = No Potential Effect; i.e. the Solution
B = Weak Potential in the Macroion's Sphere
    of Influence
C « Potential Tunnel
D * Potential Maxima Near Charged  Functional
    Groups
   28   Macroion in Random Coil Conformation Showing Four Potential Regions
                                                   277

-------
Electrical Double Layer
Counter Ions
Shear Plane
Stern Layer
                                                        30A°
A =  Anionic Polyelectrolyte Showing Potential Tunnel
     and Mobile Counter Ions
B =  Polyelectrolyte Flooding Double Layer with
     Counter Ions
C =  Chemically Bound Polyelectrolyte
Fie 29    Electrical Model for Destabilization of Negative
          Colloid with Anionic Polyelectrolyte
                                    278

-------
Potential of
Polyelectrolyte
or
Microorganism
Potential
Tunnel Surface
               A. Stable Colloid System
                                        B
Potential of
Colloid
Surface
                   • Distance Between Surfaces
               B. Polarizing Polyelectrolyte
                 or Bacteria Systems
               C. Double Layer
                 Repressions By
                 Influx Of Counter-ions
                                                A = Energy of Repulsion Curve
                                                     for Polyelectrolyte or
                                                     Microorganisms

                                                B = Energy of Repulsion Curve
                                                     for Colloid

                                                C = Summed Energy of
                                                     Repulsion
      30  Polarization Affects Distance Between Surfaces
                                       279

-------
                  Membrane
                  Periplasmic Space
                  Rigid Layer
                  :Soft Layer
                  Capsule-
          X "Increasing Negative Potential
  Zeta
 Potential —
             Capsule
            "Potential Tunnel'
            Distance-
Fig. 31    Activated Sludge Floe Potential Model
                          280

-------
              "CONSIDERATIONS IN REUSE OF REFINERY WASTEWATER"

                              Robert W. Griffin
                           Senior Staff Consultant
                  Cyrus Win. Rice Division, NUS Corporation

                               Paul Goldstein
                  General Manager, Cyrus Wm. Rice Division
                       Vice President, NUS Corporation
ABSTRACT

Refineries typically recover all usable heat at process units;  the waste
heat dissipated by evaporative cooling towers and air coolers is of such
low quality that recovery historically has been neither economic nor prac-
ticable.  Refinery effluent water reuse, however, presents a unique oppor-
tunity for utilization of this waste heat for the reduction of wastewater
volumes because the process technology used for evaporation can use this
low grade waste heat as an energy source.  Increasing present day and future
energy costs increases the attractiveness of the approach for those situa-
tions where it is needed.  Results from two successful experimental programs
conducted as part of studies involving refinery wastewater reuse are dis-
cussed.

DISCUSSION

In 1973, NUS Corporation initiated a project for the API to develop methods
of water reuse that could be employed to reduce wastewater flows from grass
roots oil refineries.  The studies were undertaken with the premise that all
characteristic refinery water use practices and patterns would be considered
so that comprehensive water management programs could be evolved.  In
essence, the multiplicity of refinery water and wastewater streams were
viewed as parts of a single system.  This approach permitted the development
of bases for defining optimum approaches to water use and reuse.

The advantage of studying a hypothetical grass roots refinery was the freedom
to employ any practical water use patterns and add treating equipment to
alter water characteristics without the necessity of considering backfit
penalties.

The model refinery was established with all of the normal unit operations
of class D refineries including the following:
                                    281

-------
    1.  Atmospheric and vacuum distillation
    2.  Hydrocracking
    3.  FCC
    4.  Sulfuric acid alkylation
    5.  Catalytic reforming
    6.  Coking
    7.  Lube processing
    8.  Gas recovery
    9-  Gasoline sweetening
   10.  Desulfurization
   11.  Sulfur recovery
   12.  BTX
   13.  Gasoline blending

During the course of these studies, operating refineries for  the most part
already had BPT end-of-pipe treatment systems in place or under construction
to meet the 1977 compliance date.  This specific project was  initiated look-
ing ahead toward the probability of more stringent controls for 1983.  It
was recognized that additional processes to further reduce pollutants might
be necessary and that treatment of large wastewater volumes would probably
be exceedingly costly.  On this basis effluent reduction could effect a dual
benefit, i.e., the improvement in treatment efficiency of current systems
and the reduction in future capital outlays and operating costs for 1983
BAT compliance.

The model refinery used for the studies had wastewater flows  and qualities
typical of a conventional 150,000 bbl/day class D refinery as a base case
situation.  The studies considered three different water supplies among many
variables.  Included was a low solids water typical of the Gulf Coast area,
an intermediate solids water similar to Mississippi River water and a high
bicarbonate alkalinity water similar to Lake Michigan.

A complete water balance was developed for each supply with resulting dis-
charge volumes from 31 to 36 gallons per barrel of crude for  the base con-
ditions.  In addition the following utilities requirements were assumed for
the studies:

    Steam Generation                     20,000,000 Ib/day
    Cooling Load                         50 x 109 BTU/day
    Electrical Load                      681,500 KW/hr/day

Pretreatment methodology and recycle schemes developed in the study showed
that reduced effluent volumes of 7 to 11 gallons per barrel of crude were
possible.

With this smaller volume of effluent as a basis for further investigations,
two additional processes for further effluent reduction were  applied.  Both
systems had the advantage of utilizing waste heat from the refinery.

The concepts employed to reduce flow from 7 to 11 gallons per barrel to
lower values included a concentrating cooling tower and a brine concentrator
as shown schematically in Figure 1.  Figure 1 also shows a non-specific

                                     282

-------
waste  treatment  plant upstream of the above referenced equipment.  Deter-
mining the  extent of waste treatment necessary in order to use wastewater as
makeup to either or both systems was an important objective of the studies.

Concentrating Cooling Tower Tests

A cooling tower  for concentrating wastewater could  theoretically be used to
reduce effluent  volumes by evaporating water up  to  the solubility limit of
the least soluble inorganic salt present.  The concept was considered to be
attractive  since the capital cost of such a tower was considerably less per
unit of water evaporated than other alternatives (such as a brine concen-
trator) and only slightly more expensive than normal cooling  tower capacity.
The cooling capacity of a concentrating tower would be substituted for other
refinery cooling and the operating cost (electrical power required) would be
essentially a trade-off.

In order to examine the feasibility of such a system a pilot  plant was tested
using refinery wastewater as makeup.  During the operation of the pilot plant,
scaling and corrosion data was of primary importance as was the fate of or-
ganics and  other volatile compounds introduced in the makeup  water.

The cooling tower selected for the tests was a coil shed type as depicted in
Figure 2.   It was in essence a normal induced draft tower with distributor
nozzles in the bottom of an elevated collecting basin which distributed water
to underlying cooling coils.  The heat load to the  tower was  steam at 35 psig
which admitted to the coils condensed and subcooled.

The pilot plant  tests were conducted at a refinery.  Extensive analyses of
 the wastewater to be used as makeup were conducted.  Inorganic analyses
 focused on the phosphate, carbonate, sulfate, silica content  which with cal-
cium could form scale.  Suspended solids was also considered  as a possible
depositing material.  It was determined from the analyses that phosphate and
silica would not be significant scaling materials and that calcium carbonate
scale could be avoided by careful pH control using  acid.  It  was simulta-
neously determined that calcium sulfate and suspended solids  would be the
 likely potential sources of deposits.

During test runs suspended solids were concentrated to values exceeding
500 ppm in the tower without deposition.  This left the deposition of calcium
sulfate of primary concern.

Figure 3 shows the solubility limit of calcium sulfate as a function of tem-
perature.  During tests, the bulk water temperature was 105°F as indicated
by Point No. 1.   This according to literature should have been the maximum
concentration of calcium sulfate attainable without scaling.  Point No. 2
 corresponds to a skin temperature of 200°F.  Point  No. 3 indicates the solu-
bility limit at  the skin temperature of the coil inlet equal  to a 285°F
steam temperature.  Scaling actually occurred at Point No. 4  conditions as
a result of the  film boiling.  Scaling started at the steam inlet to the
coil and progressed across the first tubes.
                                     283

-------
Although the coil shed tower was selected primarily  for  accessibility  of the
heat exchangers, the direct cooling of process fluids  in the  tower  limits
the usefulness of the concept and an intermediate  cooling loop  to eliminate
the occurrence of film boiling as well as to reduce  skin temperature is re-
quired to achieve higher concentrations of dissolved salts and  thereby ex-
tend the usefulness of the concept.

Figure 4 shows by arrows the conditions in the test  program where crystalline
deposits developed.  It was determined that a short  chain polyacrylate with
acid pH control utilizing hydrochloric for Test 7  and  sulfuric  for  Tests 9
and 10 was effective in suppressing crystalline deposit  formation on the
tubes.  It was therefore concluded that these materials  would be very  effec-
tive in assuring reliable operation of a system where  calcium sulfate  was
controlling.  With proper design, system pH control  and  with  the short chain
polyacrylate as an additive, it is realistic to conclude that calcium  sul-
fate concentrations of 2500 ppm could have been tolerated without crystalline
deposit formation.  During Tests 7, 9 and 10, only a light powdery  film
developed in insufficient quantities to sample,  pie material could be re-
moved easily by water sprays.

The full test program encompassed evaluations of the following  chemical con-
trol conditions:

    1.  No chemical addition

    2.  Sulfuric acid for pH control, pH 7.1-7.2

    3.  Repeat of No. 2, pH 6.8-7.1

    4.  Organic phosphate (100 ppm), sulfuric acid,  pH below  8.0

    5.  Organic phosphate - polyacrylate (100 ppm)

    6.  Combination short & long chain polyacrylate  (100 ppm) with
        sulfuric acid pH control

    7.  Short chain polyacrylate (100 ppm), HC1 pH control

    8.  HC1 alone, pH 6.9-7.1

    9.  Short chain polyacrylate, repeat of Test No. 7,  with
        sulfuric acid

   10.  Short chain polyacrylate, repeat of Test No. 9

During pilot plant operation, data was collected to  determine the fate of
the organics present in makeup water.  Test data is  included  in Table  1.
The concentration of both the phenols and ammonia  in the recycle water were
lower than those in the makeup water.

Data was collected under both sterile and unsterilized conditions with the
same results.  The reduction in total carbon, ammonia  and phenols was

                                     284

-------
therefore attributed to  air  stripping in the tower and not due to biological
oxidation.  Consequently,  it was further concluded that thorough and effec-
tive treatment of refinery wastewater makeup is necessary to prevent air
emissions due to stripping.   Conventional wastewater treatment with oil
separation, equalization,  air  flotation or filtration, biological oxidation
and final solids removal would be desirable.

Brine Concentrator Tests

A survey of refinery wastewater conditions indicated that a brine concentra-
tor could be economically  employed for effluent reduction if the stream to
be treated was of low  volume and if the device would operate successfully
on refinery wastewater without scaling and produce water of condensate
quality which could be used  as a substitute for boiler makeup water.  A small
pilot evaporator using the calcium sulfate seed slurry scale control process
was operated on a Texas  refinery wastewater.  A series of four tests were
conducted during which the volume of evaporator blowdown was successfully
reduced  to 1% of the evaporator makeup water.  Soluble salt concentrations
in recirculating brine reached 300,000 ppm and suspended solids reached
80,000 ppm during the  tests  without deposition on the heat transfer surfaces.

Oil in the makeup water  to the evaporator ranged from 1 to 3 ppm but caused
no problem.

No corrosion of  the 316  stainless steel evaporator occurred except during
an early run when the  feedwater was poorly deaerated which in conjunction
with iron  concentrations of  approximately 30 ppm led to slight pitting of
the stainless steel heat transfer surfaces.

The distillate water produced by the brine concentrator contained 60% to
70% of the phenolic materials, 25% of the freon soluble oils, 25% to 50%
of the ammonia and 50% to 60% of the total organic carbon introduced in
the feedwater.   It was therefore concluded that if the distillate was to
be employed as boiler  feedwater, conventional wastewater treatment processes
would be required upstream of the evaporator.

In summary, it was determined that with the refinery wastewaters tested,
both the concentrating cooling tower and brine concentrator are viable
methods  for reducing wastewater flow.

The concentrating cooling tower enjoys the universal advantage of utilizing
waste heat normally rejected in conventional cooling towers.  The brine
concentrator finds its most  economical application in plants where low
pressure steam that would otherwise be vented is available as an energy
source.   However, in some situations where the value of the recovered water
for boiler use is great  enough based upon alternative raw water treatment
costs, operation on the  vapor compression cycle using electric energy input
"lay be justified.
                                     285

-------
BIOGRAPHY        Paul Goldstein

         Paul Goldstein is Vice President and
General Manager of the Cyrus Wm. Rice Division
of NUS Corporation.  He holds a B.S. degree in
Marine Engineering from the United States Mer-
chant Marine Academy.  Prior experience  in-
cludes techanical management with Combustion
Engineering, Inc. and Senior Research Engineer
with Foster Wheeler Corporation.
         Robert W. Griffin received a B.S.
degree in Chemical Engineering from Grove
City College. He is currently Manager of
Special Projects fro NUS Corporation, Pitts-
burgh, Pennsylvania.  He has been with  NUS
Corporation since graduation in 1950. Mr.
Griffin is also a Consulting Engineer.
                                        286

-------
PAPER DISCUSSION

Question - Bill Ruggles, Phillips Petroleum Company

What corrosion rates were  observed during use of highly concentrated waste-
water during cooling tower tests?

Answer

Each test conducted was  of short duration making it impractical to collect
meaningful corrosion data.  We were concentrating on the scaling data and
once the tubes were scaled it was necessary to lower pH dramatically to
remove  the deposit.  The coil was made with four passes and the tubes were
two rows of stainless, one row each of Admiralty and 90-10 cupro-nickel.
After the test runs there  was no evidence of corrosion on the tubes either
generalized or pitting.  We would expect corrosion rates on carbon steel
to be similar to  those experienced on brackish waters where somewhat higher
inhibitor concentrations are required.

Question - Jeffrey Chen, Dravo Corporation

What do you do with the  concentrated brine?

Answer

There is no specific answer to the problem of brine disposal, however, it
is somewhat site  specific.  The brine concentrator decreased effluent volume
to one  or two gallons of water per barrel of crude processed.  It could be
decreased further by a drier.  Of the total dry salt produced over 50% came
from ballast water  (150,000 gal/day) which was included in the study and
represented a salt on a  dry basis of 15 to 20 tons per day.  Answers to
the question depending on  the location could be deep well disposal, ocean
disposal, on-land in arid  regions or possibly reprocessing.
                                      287

-------
                               TABLE 1

                     ANALYSES  DURING TEST RUN
Date

CARBON

2/6
2/17
2/20
2/26
3/1
3/5
3/10
3/19
3/26

AMMONIA

2/6
2/17
2/20
2/26
3/1
3/5
3/10
3/19
3/26

PHENOLS

2/6
2/17
2/20
2/26
3/1
3/5
3/10
3/19
3/26
Makeup Water
    92
    79
    81
    69
    60
    74
    69
    60
    68
     3.7
     5.4
     6.1
     8.0
     1.9

     2.6
     3.2
     3.3
     0.2
     0.26
     0.20
     0.254

     0.299

     0.330
     0.495
Recycle Water
   185
   374
   458
   278
   326
   461
   323
   380
   300
     4.8
    13.0
     0.84
     1.3
     0.4
     0.82
     3.0
     1.2
     2.9
     0.1
     0.25
     0.26
     0.092
     0.229
     0.239
     0.02
     0.286
     0.411
Calculated From
	Cycles
     506
     671.
     753,
     607.
     531,
     488.
     828
     834
     595
      20.35
      45.9
      56.73
      45.75
      14.63

      31.2
      44.5
      28.9
       1.1
       2.21
       1.86
       1.5

       1.97

       4.59
       4.33
                                288

-------
                        FIGURE  I
                PROPOSED EQUIPMENT USE
                     REFINERY
to
oo
                                                 CONDENSATE
                      OIL
                   SEPARATOR
    WASTE
 TREATMENT!
I		J
                                                   BLOWDOWNl^-cb
                                                   TO WASTEJ    f

-------
K>
                   FIGURE  2
          CONCENTRATING  COOLING TOWER

-------
       25OO-
                  FIGURE  3
          THEORETICAL AND ACTUAL
  CALCIUM SULFATE SOLUBILITY VS.TEMPERATURE
FOR CONCENTRATING COOLING TOWER DEMONSTRATION

                      ANTICIPATED AT BULK WATER TEMP 105 °F
                      PRODUCT. INLET TEMPERATURE 200°F
                      STEAM INLET TEMPERATURE 35ps!g 280 °F
                   0 ACTUAL SOLUBILITY  280°F DUE TO FILM BOILING
       2000-
     Q-
        1500-
10
\D
     u.
     _J
     ±>'
     en
     O
     _J-
     <
     O
        1000-
         500-
            100
             200
300
400
                                       TEMPERATURE (°F)

-------
IO
                                              FIGURE 4

                       PERORMANCE DATA  CONCENTRATING  COOLING TOWER TEST

                           -»• POINT OF CRYSTALLINE SCALE FORMATION

                              CALCIUM CARBONATE  AND CALCIUM SULFATE
       2500-
     10
     O
     o
     o

     0 2000-
     o.
     Q.
co  1500
UJ
z
o
cc
<
X


-------
          STATE-OF-THE-ART IN SOUR WATER STRIPPING

                         M. R. Beychok

                      Consulting Engineer
     Sour water  effluents from refining and petrochemical plants
originate primarily from the use and subsequent condensation of
process steam. The  condensation usually occurs in the presence
of a hydrocarbon vapor phase containing various amounts of NH3
and H2S. Thus the condensed steam often contains NH3 and H2S in
amounts ranging  from 1,000 to 10,000 ppm which imparts the un-
pleasant odor characteristic of sour waters. Some sour waters,
particularly from hydrocrackers, may contain as much as 30,000
to 50,000 ppm of NHs and H2S.

     Sour waters may also contain significant amounts of CX>2>
phenols, cyanides,  fatty acids and other contaminants. Fortun-
ately, the principal contaminants, NH3 and H9S, can be removed
by relatively simple steam distillation (stripping). Tradition-
ally, the refining  and petrochemical industries have stripped
NH3 and H2S from their sour waters by steam distillation at
5-10 psig and 230-240 °F. The stripping steam is either injected
directly into the distillation tower or generated in reboilers.

     A systematic study of sour waters and a tray-by-tray design
method for sour  water strippers was first published in 1967(1).
At that time, the typical refinery sour water stripper involved:

     • About 8-10 trays
     • A stripping  steam rate of about 0.8 pounds of steam per
      gallon of raw feed (Ibs/gal RF)
     • 69 % (or  more) average NH3 removal
     • 95 % (or  more) average H2S removal
     • Tray efficiencies of about 40-50 %

In 1972, the American Petroleum Institute (API) undertook a de-
tailed survey of sour water strippers(2). The results of that
survey are summarized in Table 1 herein. Briefly, the 1972
survey indicated that the average sour water stripper involved:

     • 15 trays
     • A stripping  steam rate of about 0.8 Ibs/gal RF
     « 78 % NH3  removal
     • 96 % H2S  removal
     « Tray efficiencies of about 45 %

                              293

-------
     Clearly, very little had changed in the design or perform-
ance of sour water strippers between 1967 and 1972 other than a
trend toward using more trays (15  rather than 10,  as an average).

CURRENT DESIGN REQUIREMENT

     In recent years, a number of  25 to  30 tray sour water strip-
pers have been built. This trend toward  a dramatic increase in
trays has been necessitated by ever  more stringent environmental
regulations on the NHa content of  discharged effluent waters.
Whereas the traditional designs of sour  water strippers had emph-
asized H2S removals, environmental regulations now make it nec-
essary that strippers be designed  primarily for NH3 removal.

     As an order of magnitude, Table 2 illustrates that a typical
125,000 BSD refinery (within the EPA's cracking category) may
require sour water stripping down  to a level of 25-60 ppm of  NHs
to meet the EPA's refinery effluent  guidelines for Best Available
Technology Economically Achievable (BATEA).  That level of 25-60
ppm of NH3 in the stripped water is  based on these criteria:
     — The BATEA 30-day average guidelines for BOD and NH3
        translated to annual averages using the EPA's variability
        factors for BOD and NHa
     — Assuming the 125,000 BSD cracking re-finery has a sour
        water rate of 300-350 gpm
     — Assuming that 50-100 % of  the BOD is removed in a bio-
        treater (operating at 90 % efficiency) which consumes
        4 pounds of Nitrogen per 100 pounds of BOD removed
     — Assuming that 70 % of the  nitrogen entering the bio-
        treater comes from stripped  sour water

     Given a sour water containing 7,500 ppm of NH3, it will
require 99.2 to 99.7 % NH3 removal to achieve a stripped water
NH3 level of 25-60 ppm. Regardless of the precise  accuracy of
the illustrative case in Table 2,  it is  fairly obvious that the
current design requirement for NH3 removal in sour water strip-
pers should be at least 99 % and perhaps in excess of 99.5 %.
Achieving such NH3 removals requires more trays and/or steam as
compared to the average stripper in  the  1972 API survey.  In the
current era of high fuel costs and emphasis  on energy conserva-
tion, it is important that the design engineer carefully evaluate
the tradeoff between trays and stripping steam "in  achieving a
desired NH3 removal. Figure 1 is from a  recent publicationt3) of
a stripper design study, and it illustrates  the tradeoff between
incremental trays and incremental  stripping  steam.. It shows that
99.4 % NH3 removal could be achieved,  for the specific study, by
either of the combinations of equilibrium stages and stripping
steam listed below (from which we  can obtain the tradeoff between
incremental trays and incremental  steam):
                               294

-------
         Stages	            Stripping Steam

  Equilibrium  Actual         Ibs/gal RF  Ibs/qal TF

      9.3         21              1.1         i.o
      6.0         13              1.8         1.5

Thus an increment  of  8  actual trays could replace an increment
of 0.7 Ibs of steam/gal RF.  For a stripper raw feed rate of 350
gpm, that amounts  to  353,000 Ibs/day of steam savings as the
tradeoff against using  8 more trays. The savings are in fact
even larger since  the higher steam rate would increase the size
of the stripper tower,  reboiler and overhead condenser. In this
specific case, it  is  obvious that the correct design choice would
be to use more trays  rather  than more stripping steam.

RECENT AND ONGOING RESEARCH  ON SOUR WATERS

    After completing the stripper survey in 1972, the API re-
tained the Bechtel Corporation to:

    •  Determine  if  phenols or cyanides affected the stripping
       of synthetic  sour waters in a bench-scale stripper
    •  Determine  if  actual  refinery sour waters stripped in
       the same manner as did synthetic sour waters
    •  Evaluate the  validity of Van Krevelen's vapor-liquid
       equilibrium (VLB) data for the NH3-H2S-H20 system
The  results of Bechtel's work, have been published by the API(4)
and  summarized in  an  excellent paper by Gantz(5). Briefly,
Bechtel found that:
    — Fresh synthetic solutions of NH3 and H2$, as well as
       actual refinery sour waters, both exhibited rapid and
       pronounced oxidation. This resulted in a "fixed" amount
       of NH3 residual in stripped waters which could not be
       removed by intense stripping or batch boiling. Similar
       results had been reported earlier by Dobrzanski and
       Thompson(6).
    — Fresh synthetic solutions of NHa and H2S (protected from
       oxidation) could readily be stripped to very low levels
       of both components.  Levels of 10-15 ppm NH3 and 0-5 ppm
       H2S were achieved with 5-10 bench-scale trays and strip-
       ping rates of 1.0-1.8 Ibs/gal RF. (The 5-10 bench-scale
       trays were probably  equivalent to 10-20 plant scale
       trays).
    — The addition  of as much as 800 ppm phenols and 120 ppm
       cyanide did not affect the ability to strip NH3 and H2S
       from the fresh  synthetic sour waters.
    — Actual refinery sour waters all contained a varying
       amount of  "fixed" residual NH3 after stripping which
       was attributed  to either oxidation or some unknown acid-
       ic compounds  present in the sour waters.
    — Caustic injection into the stripper could be used_quite
       effectively to  release the "fixed" NH3 and to achieve


                               295

-------
        low levels of NH3 in stripped waters.
     — The Van Krevelen VLB data were valid for  synthetic  and
        actual refinery sour waters, if proper  allowance  was made
        for any "fixed" NHa residual as determined  by a batch
        boiling test.

     Following Bechtel's work, the API and the  EPA  jointly  funded
Stanford Research Institute (SRI) during 1976-1977  to:

     •  Determine what acidic material were present in sour water
        and if there were other causes for NH3  fixation.
     •  Study caustic injection strategies.
     •  Determine the behavior of cyanides during stripping.

The results of SRI's work have been described by  Bomberger  and
SmithC"7). A final report for publication is in  preparation.
Briefly, SRI found that:
     — There were numerous problems associated with the  standard
        analytical procedures when applied to sour  waters.  In
        particular, the NH3 determination procedures need much
        further developmental work.
     — The analytical procedures for cyanide were  inadequate and
        the cyanide stripping studies were therefore not  con-
        cluded.
     — Heavy metal contents were so low in the stripped  waters
        that ammonia-metal complexes were eliminated as a cause
        of NH3 fixation.
     — Most of the refinery sour water samples had significant
        amounts of oxidized sulfur compounds, and the oxidation
        had occurred in the refineries.
     — 12-40 % of the organics in the refinery sour waters were
        phenol and cresols.
     — NH3 fixation was caused by weak sulfidic  acids, weak
        organic acids and strong sulfidic acidsi

        Refinery
           B
           C
           D
           F
           G
           H

     — The optimum caustic injection strategy  for  sour water
        stripping was single-point injection in the tower feed.
        This was more effective than single-point injection at
        the tower middle or bottom, and more effective than
        multiple point injection.
     — The optimum caustic injection strategy  (in  the tower
        feed) freed practically all of the fixed  NH3 and  did not
        interfere with H2S removal.
                               296
Weak acids
(meq/1)
<2.0
<2.0
2.9
3.9
<2.0
<2.0
Strong acids
(meq/1 )
___
4.8
0.5
1.0
_ —
3.4
Total acid
(meq/1 )
2.0
6.8
3.4
4.9
2.0
5.4
Fixed NHs
(meq/1 )
1.0
11.0
2.1
6.2
2.1
4.2

-------
    The API has  also funded work by,Dr. Grant Wilson at the
Thermochemical  Institute of Brigham Young University (BYU).  The
objectives of that  program were:

    •  To obtain new and additional VLB data for the NH3-H2S-H70
       system. Since Van Krevelen's work was essentially at
       atmospheric pressure and 70-140 °F, BYU was to obtain
       data ranging from 150-250 psia and 175-250 °F. Whereas
       Van Krevelen's VLB data could only be extrapolated and
       used for  NHa/J^S molar ratios above 1.5, the BYU results
       would provide VLB data applicable at any NH3/H2S molar
       ratio.
    •  To provide  a correlation that would calculate system pH's
       as well as  VLB data.
    •  To extend the VLB data base to include other species such
       as mercaptans, cyanides,  phenols, C02 and others.
    The  BYU results have been published in a series of draft
reports,  the latest of which is dated May 1977(8). ^ finai re_
port is in preparation. The BYU results are summarized below:
     — New VLB data have been obtained as follows:

         	System	                  Data points

         H2S-H20                                    9

         HCN-H20                                    8

         C2H5SH-H20                                 6

         NH3-H2S-HCN-H20                            4
         NH3-H2S-H20-Phenol                         6
         NH3-H2S-H20-Xylenol                        6

         NH3-H2S-H20                               18*
          (* 176-248 °F and 15-242 psia total pressure)

     — In general, when the BYU data points for the NH3-H2S-H20
        system  are  compared to temperature-extrapolated Van Krev-
       elen data WHERE THE NH3/H2S MOLAR RATIO IS ABOVE 1.5 AND
       VAN KREVELEN IS THEREFORE APPLICABLE, the BYU and the
       Van Krevelen correlations agree within about 20 %. Any
       other comparisons at NH3/H2S molar ratios of less than
       1.5 are meaningless I1'.
     — The BYU data should yield a VLB correlation that would
       apply at  any NH3/H2S molar ratio. Such a correlation
       should  be very useful in designing sour water fraction-
       ators including both a rectifying and a stripping sect-
       ion. It would also be useful in designing high pressure
        (200 psia)  sour water systems.
     — BYU has developed a VLB correlation based on their meas-
       ured data as well as data from Van Krevelen and many
       other literature sources which is combined with a method
       of calculating system pH as well.


                               297

-------
The BYU experimental data work has  greatly extended our VLB data
base, but it requires a complex computer  program for application.
In most cases, for typical refinery sour  water strippers,  the
Van Krevelen VLB correlation as recently  modified(3; still pro-
vides a simple and reliable design  basis.

     Having been associated on a consulting basis with almost all
of the research programs discussed  in this paper,  it is my own
opinion that the most useful aspect of  all the work has been the
determination of the optimum caustic injection strategy for free-
ing the fixed NH3 in sour water strippers.  It  is also my opinion
that the experimental VLB work at BYU will prove to be  very
useful.
                            REFERENCES
(1) Beychok, M. R. , "AQUEOUS WASTES FROM PETROLEUM AND PETRO-
    CHEMICAL PLANTS", John Wiley and Sons, 1967

(2) "1972 Sour Water Stripping Survey Evaluation", API Publica-
    tion 927, June 1973

(3) Beychok, M. R., "Program Calculators For Design Study",
    Hydrocarbon Processing, Sept. 1976

(4) "Sour Water Stripping Project", API Publication 946, June
    1975

(5) Gantz, R. G., "Sour Water Stripper Operations", API Meeting,
    Chicago, May 1975. (Also, Hydrocarbon Processing, May 1975)

(6) Dobrzanski, L. T. and Thompson, W. J., "Performance Eval-
    uation Of Sour Water Strippers", 76th Annual AIChE Meeting,
    Tulsa, March 1974

(7) Bomberger, D. C. and Smith, J. H., "An Experimental Study Of
    Ammonia Fixation In Sour Water Strippers", API Meeting,
    Chicago, May 1977

(8) Wilson, G. M., "A New Correlation Of NH3, C02 and H2S
    Volatility Data From Aqueous Sour Water Systems...", draft
    report to API, May 1977
                               298

-------
DISCUSSION

N.F. Seppi:   How much caustic is typically added to raw sour water stripper feed?

Milton Beychok:    SRI tested caustic injection at the top, the middle and the bottom of
tneir laboratory stripper, as well as combinations of those injection points.  Their caustic
injection rates ranged from 60% to 120% of the stoichiometric amount based on the
fixed ammonia present.  As I recall, 90% to 100% of the stoichiometric amount was
enough to free and to remove essentially all of the fixed ammonia.
Ed Bienhoff:  Are you indicating that there is a shift away from two-stage strippers
wfiere caustic is injected between stages?

Milton Beychok:   I did not realize that I had even mentioned that aspect. As you
know, some refiners have used two-stage strippers to  remove the ammonia and the f-LS
as separate streams and to dispose of them separately. Basically,  those operations use
two conventional strippers in series. Acid is injected into the first stripper to maximize
the removal of H«S and  caustic is injected into the second stripper to maximize ammonia
removal.  Personally, I  think that it is better to use one of the proprietary licensed
processes for that purpose,  such as Chevron's fractionating process of U.S.  Steel's
PHOSAM process.  My reason for preferring those processes over the two-stage system
with separate acid  and caustic injection is that it must be quite difficult to achieve
good pH control in the two-stage system without  running into a lot of corrosion problems.

N.F. Seppi:   I must have  misunderstood you.  Were you referring to caustic injection
into a single tower or into the feed of the second tower of a  two-stage system?

Milton Beychok:  In terms of the SRI work, I was referring to a single conventional
stripper with caustic  injection into  the feed entering  that single stripper (i.e. the top
of the column), or the middle of that stripper, or the bottom of that stripper as well
as combinations of those injection points.  Feed  injection proved  to be the best.
                                         299

-------
BIOGRAPHY            Milton R. Beychok

         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 registered 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  from 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.
                                       300

-------
                             TABLE  1
                1972 SOUR WATER STRIPPER SURVEY*
PERFORMANCE

   Average NHs removal = 78.1 %   (50 towers reported)
   Average H2S removal = 95.8 %   (51 towers reported)

TRAYED TOWERS

   Average number of trays =15   (44 towers reported)
   Average number of trays, excluding three highest and three
             lowest values = 15

   Average tray efficiency = 46 % (12 towers evaluated)
   Average tray efficiency, excluding one highest and one
              lowest value = 45 %

PACKED TOWERS

   Average packed height = 15 ft  (14 towers reported)

STRIPPING STEAM

   Average of all towers = 0.8 Ibs/gal of raw feed
   Average of all towers removing more than 90 % NH3
                         = 1.2 Ibs/gal of raw feed

TOTAL STEAM
   Average of all towers = 1.1 Ibs/gal of raw feed
   Average of all towers removing more than 90 % NH3
                         = 1.4 Ibs/gal of raw feed
API Publication No.  927,  June 1973 (73 survey questionnaires)
                              301

-------
                             TABLE 2

         ALLOWABLE NH3 CONTENT IN STRIPPED SOUR WATER
                  TO MEET 1983 BATEA LIMITS
                (Typical 125,000 BSD Refinery)


EPA category	 cracking
Process configuration 	 6.32
Process factor 	 1.09
Size factor 	 1.47
Stripped sour water = 300 gpm = 3,600,000 Ibs/day

BATEA limits (30-day average):
      BOD =1.0 lbs/1000 bbls crude oil
    NH3-N =1.2 lbs/1000 bbls crude oil

Variability factors:
      BOD =1.7
    NHs-N =1.5
BATEA allowable discharges (equivalent annual averages)i
      BOD = (1.09)(1.47)(1.0)(125)/1.7 = 118 Ibs/day
    NH3-N = (1.09)(1.47)(1.2)(125)/1.5 = 160 Ibs/day

Assuming a biotreater removes 90 % of the BOD and consumes
4 Ibs of Nitrogen/100 Ibs of BOD removed* :

Nitrogen consumed = 118(0.9/0.1)(4/100) = 42 Ibs/day
Nitrogen entering = 160 +42            = 202 Ibs/day

Assuming 70 % of Nitrogen entering the biotreater comes from the
stripped sour waters

Allowable NH3 in stripped water

          = 202(0.7/3.6)(17/14) = 48 ppm as NH3

Repeating the above calculations for a range of values:
          stripped   % of NH3 entering
         sour water  biotreater coming    allowable ppm NH3
           (gpm)        from SWS	    in stripped water

                                                 61
                                                 48
                                                 34
                                                 53
                                                 41
                                                 29
300
300
300
350
350
350
90
70
50
90
70
50
                                         range: 30 to 60 ppm
* If only half of the BOD is removed
  in a biotreater, the range is 25 to 50 ppm
                              302

-------
8
CO
        a
w
w
I
CO
11


10


 9


 8


 7 (-


 6 -


 5 -


 4 -
        g
                   Fresh feed

                  Reflux drum
              -   FIGURE  1
3


2
                                  T
                                                               T
                          	1	1	
                    10,OOO ppm H2S and 7,500 ppm NH3
                    200 OF after exchange with bottoms
                    2.3 psig with a 2.0 psi pressure drop
                             across condenser and 0.5 psi
                             pressure drop per equilibrium stage


                                             — --o-
                                                                   <=>  -
                                                                 *1> '
                                                                           x
              -  o-
             —O  Stripping steam @ 1.0 Ibs/gal of TF (1.1 ibs/gal RF)
         O—— —O  Stripping steam @ 1.5 Ibs/gal of TF (1.8 ibs/gal RF)
                                              (1) Includes reboiler equilibrium
                                                  stage. Excludes overhead cond-
                                                  enser equilibrium stage.
                                              1
                                                  1
                     95
                      96           97

                         % REMOVAL OF
                                                  98
                                                          99
                                                     AND

-------
                     PETROLEUM REFINERY DISCHARGES TO A
                          LARGE SANITATION DISTRICT

                                Irv Kornfeld
       Lead Project Engineer, Sanitation Districts (Los Angeles County)

                                Jay G. Kremer
  Head, Industrial Waste Section, Sanitation Districts (Los Angeles County)


INTRODUCTION

     The Los Angeles County area is one of the major oil producing and
petroleum refining areas in the United States.  Half of the reported 26
U.S. petroleum refineries discharging to publicly owned treatment works
(POTW) are served by the County Sanitation Districts of Los Angeles County
(Districts or LACSD) sewerage system  (Figure 1).  Refinery industrial waste-
water discharges to the Districts total in the neighborhood of 17 to 20
million gallons per day (mgd), (64,345 to 75,700 m3/d).  It has been
estimated that this volume constitutes over 50% of all refinery discharges
to POTW in the United States.

     Eight of the larger refineries discharging to the Districts are
refineries whose processes can be classified in EPA Category B; that is,
those refineries with topping and cracking operations  (see Table 1).
The processes of the five smaller refineries can be classied in Category A;
that is, those refineries with topping and crude distillation operations
only (see Table 2).  The crude capacity of these 13 refineries totals
over 800,000 barrels per stream day (b/sd), (127,176 m3/sd).

SEWERAGE SERVICE IN LOS ANGELES COUNTY

     Consisting of 27 individual districts, the Sanitation Districts provide
sewerage service to the major portion of Los Angeles County outside of the
City of Los Angeles.  Fifteen of these Districts collectively own and operate
the Joint Outfall System which provides a common sewerage system for
over 3.8 million people, 750 square miles (1943 km2) of area, and
approximately 8,000 industrial companies.  The Districts own, operate and
maintain over 1100 miles (1770 km) of trunk sewers, and treatment facilities
for wastewater flows of 420 mgd  (1,589,700 m3/d).  The local 72 cities
within the Districts provide and maintain the small collection sewers.
Methods of wastewater treatment include primary, secondary and tertiary
at eleven treatment plant locations.
                                     304

-------
THE JOINT WATER POLLUTION  CONTROL PLANT

    All but one of the petroleum refineries discharge to the Districts'
Joint Water Pollution Control Plant (JWPCP) located six miles inland from the
Pacific Ocean, in the City of Carson.   The tributary sewers to this plant
accept primary and activated sludge from several upstream treatment plants,
in addition to domestic and industrial wastewater.  The JWPCP presently
utilizes primary treatment with polymer addition and advanced solids
recovery procedures to treat a daily average of 350 mgd (1,324,750 m3/d)
of wastewater, which includes approximately 70 mgd  (264,950 m3/d) of
industrial flow.  The petroleum refineries' wastewater discharge amounts to
approximately 25% of this  industrial flow and about 5% of the total JWPCP
wastewater influent.

     The JWPCP solid wastewater material, except for grit, is digested
anaerobically.  The digested material is processed by screening, centri-
fuging, and air drying for conversion to an innocuous end product suitable
for use as fertilizer.  The JWPCP wastewater effluent is discharged
directly to the Pacific Ocean about two miles off-shore from the Palos
Verdes Peninsula through  a system of tunnels and submarine outfalls.  A
commitment has been made  by the Districts and work is currently underway
to convert the JWPCP to a full secondary biological treatment plant.
Many components of refinery wastewater such as phenols, and oil and grease
will be more  adequately treated by such a plant.

THE DISTRICTS' INDUSTRIAL WASTE ORDINANCE

     In  1972, requirements defined in the Federal Water Pollution Control
Act Amendments  (PL 92-500) and the State of California "Ocean Plan" man-
dated that the Districts  establish a program which would control pollutant
levels  in  treatment plant effluents.  To meet the Districts' treated
wastewater quality goals,  control industrial pollutants, and to recover the
true cost  of  wastewater treatment from industrial companies, the Districts
adopted a Wastewater Ordinance on April 1, 1972.

     This  ordinance, which was amended on July 1, 1975, included a permit
program for industrial dischargers to the sewerage system.  Information
required  in the permits for major dischargers included:

       1.  Industrial process descriptions.

       2.  Industrial process equipment information and plans.

       3.  Description of wastewater pretreatment equipment.

       4.  Sewer plans of the industrial plant.

       5.  Pertinent wastewater constituent concentrations.

       6.  Wastewater flow volumes and methods of wastewater flow
          measurement.
                                     305

-------
       7.  Description of liquid waste materials  disposed  of  other than
           to the sewerage system.

Wastewater characterization information required  from the  refineries
includes the parameters listed in Table 3.  The permit  requirements for
the petroleum refinery industry (SIC 2911) specify minimum wastewater
pretreatment facilities of an oil-water separator and a sampling  point.

THE INDUSTRIAL WASTE SECTION

     The Districts' unit implementing the industrial  waste regulatory
program is the Industrial Waste Section.  Permit  processing,  field
inspection, and industrial waste engineering are  the  function designations
of the three subsections of the Industrial Waste  Section.  Graduate civil,
chemical or mechanical engineers fill over two-thirds of the  19 professional
positions in this section.  One function of the industrial waste  engineering
subsection is to provide technical support through a  project  engineer
competent in a specific field of industrial wastewater  engineering.  One
project engineer covers the oil producing and petroleum refining
industries.

WASTE DISCHARGES PROHIBITED BY ORDINANCE

     Listed in the Districts ordinance as prohibited  discharges and applic-
able to wastewaters from petroleum refineries are the following:

       1.  Any gasoline, benzene, naptha, solvent, or fuel oil.

       2.  Any waste containing toxic or poisonous solids, liquids
           or gases.

       3.  Any waste having a pH lower than 6.0 or having  any corrosive
           or detrimental characteristic that may cause injury to
           wastewater treatment or maintenance personnel.

       4.  Any solids or viscous substances of any size or in such
           quantity that they may cause obstruction to  flow in the
           sewer or be detrimental to proper wastewater treatment
           plant operations.

       5.  Any rainwater, stormwater, groundwater, or street  drainage.
                          J
       6.  Any water added for the purpose of diluting  wastes which
           would otherwise exceed applicable maximum  concentration
           limitations.

       7.  Any excessive amounts of petroleum or  mineral based cutting
           oils (commonly called soluble oils) which  form  persistent
           water emulsions;

       8.  Any excessive concentration of non-biodegradable oil,
           petroleum oil or refined petroleum products.


                                    306

-------
      9-   Any waste with an excessively high concentration of cyanide.

     10.   Any unreasonably large amounts of undissolved or dissolved
          solids.

     11.   Any waste with excessively high BOD, COD or decomposable
          organic content.

     12.   Any strongly odorous waste or waste  tending to  create
          odors.

     13.   Any waste containing dissolved suIfides above a concen-
          tration of 0.1 milligram per liter.

     14.   Any waste with a pH high enough to cause alkaline
          encrustations on sewer walls or other  adverse effects
          on the sewerage system.

     15.   Any waste having a temperature of 120°F or higher.

     16.   Any excessive amounts of deionized water, steam
          condensate, or distilled water.

     17.   Any waste containing substances which  may precipitate,
          solidify or become viscous at temperatures between
          50°F and 100°F.

     18.   Any blowdown or bleed water  from cooling towers or  other
          evaporative coolers exceeding 1/3 of the make-up water.

     19.   Any single-pass cooling water.

These regulations have had a major effect on the  ability of Districts'
treatment plants to handle refinery wastes and  the type of pretreatment
equipment required at each refinery.

INDUSTRIAL USER CHARGE

    Prior to the advent of the Districts' industrial waste ordinance,
the petroleum refineries and all other  large industrial dischargers
were charged only an ad valorem  (property) tax  for sewerage service.
This tax, however, did not cover the  true costs of industrial  wastewater
treatment.  The Districts' ordinance  established  an industrial user
charge or surcharge to obtain the revenue needed  to meet federal  law
PL 92-500 requirements.  Use was made of the charge parameters of total
wastewater flow, chemical oxygen demand, suspended solids  and  peak
flow in a Districts' surcharge formula  (Table 4).

    The surcharge rates for the last three fiscal years are  given on
Table 5.  The flat rate charge of $230  to $250  per million gallons can
be utilized by only industrial waste  dischargers  of less than  6 million
                                   307

-------
gallons per year.  However, this figure of  $250 per million gallons  is
the approximate amount that a large refinery will pay  the  Districts
for sewerage service.

     The fiscal year 1975-76 ad valorem (property) taxes and surcharge
from the refineries to the Districts totaled about $1,100,000 of which
the surcharge portion amounted to about $715,000.  The refineries'
surcharges amounted to approximately 16% of all industrial waste
surcharge payments received in fiscal year  1975-76.

SULFIDES AND THIOSULFATES

     In the 10 years prior to promulgation  of  the Districts ordinance,
a Districts' policy governing the use of trunk sewers  provided that
dissolved sulfides in wastewater discharged to the sewer must not
exceed a concentration^of 0.1 mg/1.  The petroleum refineries,  since the
early 1960's, had treated wastewater containing high dissolved sulfides
by oxidizing these sulfide compounds to thiosulfate.

     In the early 1970's, new State of California discharge requirements
for the Districts' JWPCP included more stringent bacterial standards.
These standards required substantial chlorination of JWPCP effluent
when large quantities of refinery thiosulfate  was present.   In order
to maintain chlorination effectiveness, the Districts  required that each
refinery discharging to the Districts' system  must reduce  its thio-
sulfate level to not more than 50 mg/1 as sulfur.  The refineries were
given a period of two years, until July 1,  1973, to complete the required
construction for the reduction in thiosulfate  concentration.

     Only the refineries in the EPA Category B, which  were high dischargers
of thiosulfate, were affected.  These refineries, totaling over 90% of
all refinery discharges, accomplished the task of reducing thiosulfate by
constructing sulfide strippers for refinery sour water streams.  These
sulfide strippers had a most significant effect in reducing overall
pollutant concentrations discharged from the refineries.

OIL AND GREASE

     Concentrations of mineral oil and grease  greater  than 75 mg/1 are
considered excesssive in industrial discharges to the  Districts' system.
Refinery wastewater dischargers are required,  as a condition of permit
approval, to pretreat so that the oil and grease content is below
75 mg/1.  All major refineries have used a  combination of  "good house-
keeping", improved oil water separators and dissolved  air  (or gas)
flotation units to obtain adequate oil and  grease removals.

RAINWATER

     It is the policy of the Sanitation Districts that rainwater will
not be permitted access to the Districts' sewerage system  due to
limited capacity, primarily designed for dry weather  flows, (separate
sanitary sewerage system).  In certain situations, where discharge  to the


                                     308

-------
storm sewer is not feasible, refineries  are permitted to discharge rain-
water to the sewer.  Rainwater permitted to be discharged with industrial
wastewater during a rain storm is  limited to the first 0.1 inch (2.54  cm)
of rainwater over the relevant surface area.   Normally,  rainwater diversion
devices discharge storm flow to surface  drains after 0.1 inch (2.54 cm) of
rain has fallen.  If greater than  this amount of rainwater is too polluted
for discharge to a storm drain, it may be stored on the refinery property
during the rain storm for later discharge to the sewer.   Discharge to
the Districts sewerage system can  only be made 24-hours after the
cessation of rainfall and then only  during the off-peak hours of sewer
flow, (10:00 p.m. to 8:00 a.m.).

TOXIC WASTEWATER CONSTITUENTS

     Limits for the discharge of toxic pollutants,  mainly heavy metals,
were presented to industry on July 1,  1975, with an enforcement date,
after an eighteen-month implementation period, of January 1,  1977 (see
Table 6).  The Districts believe that  these limits with some  minor
modifications will allow the Districts to comply with it's NPDES permit
requirements for toxic pollutants  discharged from its treatment plants.
The refineries' existing wastewater  discharges are generally  in compliance
with these toxic wastewater constituent  limits.   Typical concentrations
of these materials in refinery effluent  are shown in Table 7.

REFINERY DISCHARGE IMPROVEMENTS

     A significant reduction was noted in refinery wastewater pollutants
in  1974 following the start up of  refinery sour water strippers and the
resultant reduction in wastewater  thiosulfate to levels of less than
50 milligrams per liter.  Tables 8 and 9 indicate typical wastewater
quality at the same category refineries  during 1972-73.   Table 10 and  11
indicate typical wastewater quality  at the same category refineries
during 1975-76.  The mass emission totals and percentage reductions for
refinery wastewater pollutants between the years 1972-73 and  1975-76 are
listed in Table 12.

     In addition to thiosulfate, significant reductions were  obtained  in
other refinery constituent levels, including COD, ammonia, phenols, and
oil  and grease.  During the three  year period, the refineries improved
their primary treatment facilities and upgraded their environmental
wastewater management efforts.  These  activities served to significantly
reduce refinery wastewater pollutant levels.

     Reduction in refinery pollutant levels coincided with significant
reductions in BOD levels, about 70 mg/1, in the influent to the JWPCP.
An  investigation determined that these reduced BOD levels were mainly
attributable to the decrease in refinery thiosulfate discharges.

     It has been estimated that, since  1972, the cost to the  LACSD
refineries for capital wastewater  pretreatment improvements has amounted
to  over 20 million dollars.  Thiosulfate reduction, which mainly included
new sour water sulfur stripping facilities, accounted for approximately


                                    309

-------
80 percent of these costs.

REFINERY WASTEWATER PRETREATMENT SYSTEMS

     Pretreatment of wastewater at the refineries  discharging  to  the
Districts sewerage system involves mainly advanced primary  treatment.
A primary treatment system for a Category B refinery  includes  a sour
water stripper, a sour water oxidizer, an oil water separator,  and a
dissolved gas or air flotation unit.  A typical pretreatment system
for a large refinery is shown on Figure 3.  Small  Category  A refineries
have nearly the same pretreatment systems except the  sour water stripping
capability is usually not included.  The small refineries,  for  the
most part, refine low-sulfur crude which does not  generate  a high-
sulfide wastewater.

     Three of the large refineries discharging to  the Districts'  system
are able to discharge a portion of their low pollutant wastewater streams
to a storm water channel which flows to the Los Angeles Harbor  area.
These channel discharges are under the jurisdiction of the  local
California State Water Quality Control Board.  The Board limits the mass
emission of pollutants, including BOD and oil and  grease, which can be
discharged into a storm water channel.

REFINERY ODORS

     The improvements in refinery pretreatment systems helped  reduce
odor problems in the Districts' trunk sewers and in treatment plants
receiving refinery wastewater.  One problem prevalent in the past was
refinery odors being released in a Districts' treatment plant.

     With one company's cooperation, an in-plant refinery wastewater
survey was initiated in order to determine if an odorous refinery
wastewater stream could be selectively separated and  treated for  odor
removal prior to sewer disposal.  The refinery, at the time of  investi-
gation, discharged approximately 5 percent of the  influent  received at
the Districts' Los Coyotes Water Reclamation Plant (LCWRP), a  secondary
treatment plant.  At the time of the survey, the refinery did  not have
a sour water stripper and all sulfides were oxidized  to thiosulfate.

     It was found that the sour water oxidizer, operated for wastewater
sulfide removal, was the refinery's major existing odor removal facility.
Also, selectively removing particularly odorous sour  water  streams
from the total refinery wastewater was not practical  as over 60 percent
of the refinery sour water streams included highly odorous  sulfide
concentrations.  The installation of a sour water  stripper  facility, which
could reduce ammonia, sulfide, and thiosulfate discharges,  was found to
result in substantial wastewater odor reductions.   After the refinery's
sour water stripper began operation, wastewater odors at the LCWRP
were significantly reduced.  Optimum stripper operation required
removing a particularly odorous refinery spent caustic stream  from
sewer discharge and truck transportation of it to  a landfill for
disposal.


                                   310

-------
PETROLEUM REFINING PRETREATMENT STANDARDS

     The Environmental Protection Agency  (EPA)  established petroleum
refinery pretreatment standards (Part  419,  Federal Register,  March 23,
1977) in an interim final form for pollutants  discharged to POTW from'
existing sources.   Established were  two sets of pretreatment  standards.
The first set,  known as the prohibited discharge standards, is  for control
of gross problems  such as the discharge of  flammable materials  or
wastes that could  plug sewers.  The  Districts'  Wastewater Ordinance
also prohibits the discharge of such materials  to the sewer system.   The
second set of pretreatment standards establishes numerical limits on
ammonia and oil and grease and suggests local  restrictions on chromium
sulfide and phenol.

     The regulations established a maximum  one-day concentration of 100
milligrams per liter (mg/1) for ammonia and 100 milligrams per liter for
oil  and grease allowed to be discharged by  petroleum refineries to POTW.
In addition, guidance was provided to  the operators of POTW for control
of chromium, sulfide, and phenolic compounds,which may prove  harmful
to or not be adequately treated by the POTW.   The EPA recommended,
however, that sulfides, phenols, and chromium  be controlled only as
needed by the local agency.

     The Districts commented on the  development document which established
 these regulations.  It was recommended that the EPA either propose that
 sewerage agencies receiving refinery waste  establish suitable local
 source  control programs, or that any EPA  limits be established uniquely
 for  each of the 13 sewerage systems  receiving  such wastes.  It is
believed that sewerage agencies such as the Sanitation Districts have
 the  staff and technical competence to  operate  a cost effective industrial
 source  control program and should be given  the total responsibility
 for  such a program.

      Impact on the Districts and the local  petroleum industry for the
 items prohibited  is minimal as  the Districts'  ordinance has already
 established most  of the standards.   The mean concentrations of oil and
 grease, and ammonia in the refinery  studies made by EPA were stated to
 be well below the maximum pretreatment standard of 100 mg/1.   This is
 not  true for all  refineries discharging  to  the Districts' sewerage
 system  as two or  three may discharge daily  ammonia levels in excess of
 100  milligrams per liter.  If the Districts'  treatment plants can meet
 the  required NPDES permit conditions for  ammonia discharge, it is not
 cost effective to require more  severe  refinery ammonia restrictions
 than needed for environmental protection.   All of the Districts' refineries
 are  required to discharge oil and grease  below a daily level of 75
 (<100)  milligrams per liter.

 SUMMARY

      The Sanitation Districts of Los Angeles County provide sewerage
 service for 13 (50%) of the 26  U.S.  refineries reported to be discharging
 to publically owned treatment works  (POTW).  Eight refineries are


                                     311

-------
classified in EPA Category B and  five  in  Category  A.   Total refinery
capacity is 800,000 barrels per stream day  (127,176 m3/sd)  with a
wastewater discharge of about  17  to 20 million  gallons per  day  (64,345
to 75,700 m3/d).

     The San_tation Districts  have been able  to satisfactorily  treat
municipal wastewaters containing  significant  quantities of  refinery
wastes.  Wastewater discharge  requirements  for  refineries have  been
established by a Districts ordinance.   This has resulted in wastewater
pretreatment installations at  local refineries  costing over 20  million
dollars.  Most major refineries have pretreatment  equipment which
includes a sour water stripper, a sour water  oxidizer, an oil-water
separator, and a dissolved air (or gas) flotation  unit.

     A  significant improvement in the  quality of wastewater discharged
by refineries has occurred in  the last few  years.  Installation of
improved pretreatment equipment by the refineries  has  significantly
reduced some problems at Districts' treatment plants such as high odor
levels  and high wastewater organic content.

     In the Sanitation Districts  area, some pretreatment regulations
for  refineries are quite stringent, such  as 0.1 mg/1 of dissolved
sulfide and 75 mg/1 of oil and grease.  Local conditions require that
these limits be more severe than  proposed by  EPA in the recently
published pretreatment regulations.  Conversely, it appears that the
EPA  pretreatment limit of a one-day maximum concentration of 100 mg/1
of ammonia from refineries may not be  required  for meeting  pollution
control goals.

     The Sanitation Districts  have requested  that  EPA  permit responsible
local sewering agencies such as the Districts to establish  and  enforce
its  own industrial pretreatment regulatory  program.  Such a program,
aimed at meeting required receiving water quality, would comply with
pollution control goals without placing an  excessive cost burden on
industry.
 DISCUSSION

 Randy  Buttram;  Can a  1 trotted  amount  of storm water be discharged along with
 normal  process  effluent  during  a  storm?

 Irv Kornfeld: Rainwater  permitted to be discharged with  industrial waste-
 water  is  limited to the  first 0.1 inch (0.254 cm)  of rainwater over  the
 relevant  surface area.   It  is the policy of  the  Districts that rainwater
 will not  be permitted access  to the  Districts' sewerage  system.   This is
 because the system has limited  capacity as it is primarily designed  for dry
 weather flows.  However, in certain  situations where discharge to a  storm
 sewer  is  not feasible, refineries are  permitted  to discharge rainwater to
 the sewer 24-hours after cessation of  rainfall during the off-peak hours
 of sewer  flow;  that is, between 10:00  p.m. and 8:00 a.m.
                                     312

-------
BIOGRAPHIES
   Irv Kornfeld is a Lead Project Engineer
in the Industrial Waste Section of the
Sanitation Districts of Los Angeles County.
He holds a B.S. degree in Mechanical
Engineering from the Illinois Institute  of
Technology and a M.S. degree in Civil
Engineering from the California State
University at Long Beach.  He is a
Registered Mechanical and Civil Engineer
in the State of California and a member
of ASCE and WPCF.  Mr. Kornfeld has been
with the Sanitation Districts since 1972
and has prior employment with engineering
consultants and contractors to the petroleum
refining industry.
   Jay G. Kremer is Head of the  Industrial Waste
 Section of the Sanitation Districts  of  Los
 Angeles County.  He has a B.S. degree in Civil
 Engineering from Northwestern University,  a M.S.
 degree in Sanitary Engineering from  the Illinois
 Institute of Technology and a Masters of Public
 Administration degree from the University of
 Southern California.  He is a registered Civil
 Engineer in California and a member  of  ASCE,
 WPCF and AAEE.  He has been with the Sanitation
 Districts since 1963 and previously  worked for
 various consulting engineer and  governmental
 agencies.
                                     313

-------
UJ
                   LOS  ANGELES

                CITY   OF
	J
     REFINERIES
  I. ATLANTIC RICHFIELD CO.
  2. DOUGLAS OIL CO.
  3. GULF OIL  CO.
  4. MOBIL OIL CO.
  5. POWERINE OIL CO.
  6. SHELL OIL CO..
  7. TEXACO INC.
  8. UNION OIL CO.
  9. EDGINGTON OIL CO.
 10. FLETCHER OIL CO.
 II. GOLDEN EAGLE REFINERY CO.
 12. LUNDY THAGARD OIL CO.
 13. MACMILLAN RING FREE OIL CO.
                           — OCEAN OUTFALLS

           FIG. 1.  SANITATION DISTRICTS  OF  LOS ANGELES  COUNTY
                        PETROLEUM REFINERY  LOCATIONS

-------
cr
o
H
O
O
z
>

a.

H

D
5
                   RATIO  P/A
NOTE: Mathematical formula  for "M11 is: M = 2.5 log(P/A)
    2.  VALUES  OF  MULTIPLYING FACTOR "M1
                       315

-------
        SOUR WATER
         STRIPPER
SOUR WATER
 SOURCES
           I
        SOUR WATER
         OXIDIZER
DESALTER
 WATER
          25% VOLUME
 10%I VOLUME
                      NEUTRALIZATION BASIN
                            I
                      OIL WATER SEPARATOR
                       DISSOLVED AIR OR
                       GAS FLOTATION
                            i
                       DISCHARGE BASIN
                       FLOW METERING
                       SAMPLING POINT
                            f
                        PUBLIC SEWER
COOLING WATER
 SLOWDOWN
BOILER SLOWDOWN
MISC. OILY WATER
  65% VOLUME
                    SANITARY
                     WASTE
FIG. 3. TYPICAL  LARGE REFINERY WASTEWATER
        PRETREATMENT SYSTEM
                              316

-------
            TABLE 1  REFINERIES DISCHARGING TO THE LACSD,
              CATEGORY B (TOPPING AND  CRACKING PLANTS)

                                              Crude      Wastewater
                                             Capacity    Discharge
	Refinery	       Location          b/sd     	mgd

Atlantic Richfield Co.   Carson               186,000   > 4.0   < 6.0
Douglas Oil Co.          Paramount             48,000   > 0.2   < 0.5
Gulf Oil Co.             Santa Fe Springs      53,800   > 0.4   < 0.7
Mobil Oil Co.             Torrance             131,000   > 3.0   < 5.0
Powerine Oil Co.         Santa Fe Springs      46,000   > 0.2   < 0.3
Shell Oil Co.             Carson                93,000   > 2.0   < 4.0
Texaco, Inc.             Wilmington            79,000   > 0.6   < 0.8
Union Oil Co.             Wilmington           111,000   > 3.0   < 5.0
                                  Total       747,800   >13.4   <22.3


Metric Conversion

m3/d = (mgd) (3,785)
            TABLE 2  REFINERIES DISCHARGING  TO THE LACSD,
                     CATEGORY A (TOPPING  PLANTS)

                                              Crude      Wastewater
                                             Capacity    Discharge
	Refinery	    Location        b/sd          mgd	

Edgington Oil Co.             Long  Beach      31,000    >0.10   <0.30
Fletcher Oil & Refinery Co.   Carson         20,000    >0.05   <0.10
Golden Eagle Refinery Co.     Carson         13,000    >0.04   <0.07
Lunday-Thagard Oil Co.        South Gate       8,100    >0.02   <0.06
Macmillan Ring Free Oil Co.   Signal Hill    12.200   _^0.05_ _fOJ.08_
                                  Total       84,300    >0.26   <0.61
Metric Conversion

m3/d = (mgd)(3,785)
                                  317

-------
           TABLE 3  PETROLEUM REFINERIES WASTE  CHARACTERIZATION

                     Parameter         Quantity Values

                  Flow (Total)            gals/day
                  Flow (Peak)             gals/day
                  COD                       mg/1
                  Suspended Solids          mg/1
                  pH                        units
                  Ammonia  (N)               mg/1
                  Oil and Grease            mg/1
                  Phenols                   mg/1
                  Thiosulfate (S)           mg/1
                  Arsenic                   mg/1
                  Cadmium                   mg/1
                  Chromium                  mg/1
                  Copper                    mg/1
                  Lead                      mg/1
                  Nickel                    mg/1
                  Zinc                      mg/1
                  Cyanide     '--••	: -       mg/1
                         TABLE 4  LACSD SURCHARGE

     Surcharge = a (V) + b (COD) + c (SS) + dM (P) - TAX

and where surcharge equals, net,-annual industrial wastewater treatment
surcharge, in dollars.  No refund is made if a negative number results.

    V = Total annual volume of flow, in millions of gallons.
  COD = Total annual discharge of chemical oxygen demand in thousands
        of pounds.
   SS = Total annual discharge of suspended solids in thousands of
        pounds.
    P = Peak discharge rate over a 30 minute period, occurring between
        the hours of 8:00 a.m. and 10:00 p.m.
    A = Average discharge rate,  determined by dividing (V) by the total
        annual hours of operation and working time for the industrial
        discharger converted to gallons per minute.
    a, b, c & d = Unit charge rates adopted annually by the individual
        District based upon the projected annual total cost of wastewater
        collection, treatment and disposal, in dollars per unit.
    M = A multiplying factor accounting for increased Districts cost to
        high ratios of industrial discharge to obtain flow rates  (P/A).
        Factor M is obtained from Figure 2.
  TAX = The annual ad valorem taxes paid to the Districts during  the accrual
        years on land or property utilized for the generation of  industrial
        wastewater, in dollars.


                                    318

-------
TABLE 5  INDUSTRIAL WASTEWATER TREATMENT SURCHARGE RATES
Unit Rate
a
b
c
d
Flat Rate
Charge
Parameter
Volume
Millions of gallons
per year
COD
Thousands of
pounds per year
Suspended Solids
Thousands of
pounds per year
Peak Flow
Gallons per
minute
Volume
Millions of gallons
per year
Surcharge Rates
1974-75
$104.00
$ 6.25
$ 14.25
$ 18.75
$230.00
1975-76
$104.00
$ 6.25
$ 14.25
$ 12.00
$230.00
1976-77
$127.00
$ 6.60
$ 16.10
$ 14.60
$250.00
       TABLE  6   LACSD INDUSTRIAL WASTEWATER EFFLUENT
         LIMITATIONS  FOR JOINT OUTFALL DISTRICTS
          Constituent
       Arsenic
       Cadmium
       Chromium (Total)
       Copper
       Lead
       Mercury
       Nickel
       Silver
       Zinc
       Cyanide  (Total)
       Total Identifiable
       Chlorinated
       Hydrocarbons
Phase I Control Period
	(mg/1)	

           3
          15
          10
          15
          40
           2
          12
           5
          25
          10
   Essentially None
                               319

-------
       TABLE 7  TYPICAL REFINERY HEAVY METALS DISCHARGE LEVELS
                         1975-76 (typical day)
                    Constituents
                      Arsenic
                      Cadmium
                      Chromium
                      Copper
                      Lead
                      Nickel
                      Zinc
                      Cyanide
Concentrations
   in mg/1

    0.01
    0.01
    1.1
    0.06
    0.11
    0.08
    0.41
    0.85
                TABLE 8  LACSD REFINERY DISCHARGE LEVELS
                          1972-73 (Typical Day)
                          Category B Refineries
Refinery
Flow mgd

COD
Suspended Solids (SS)
pH
Ammonia (N)
Oil & Grease
Phenols
Thiosulfate (S)
Chromium (Cr)
A
4.968
mg/1
1,426
27.5
7.4
895
25
460
783
0.445
Ib/day
59,083
1,139
	
37,082
1,036
19,059
32,442
18.45
B
0.479
mg/1
2,640
97
8.9
1,075
165
80
865
1.10
Ib/day
10,546
388
	
4,294
659
320
3,455
4.39
C
3.297
mg/1
3,205
108
11.5
1,162
122
3.93
2,113
1.22
Ib/day
88,128
2,970
	
31,951
3,355
108
58,101
33.54
Metric Conversion
kg = (Ib)(0.454)
                                  320

-------
               TABLE 9  LACSD REFINERY DISCHARGE LEVELS
                          1972-73 (Typical Day)
                          Category B Refineries
Refinery
Flow mgd

COD
Suspended Solids (SS)
pH
Ammonia (N)
Oil and Grease
Phenols
Thiosulfate (S)
Chromium (Cr)
D
4.034
mg/1
2,956
595
8.7
181
460
1,617
688
1.06
Ib/day
99,450
20,117
	
6,089
15,476
54,401
23,148
35.66
E
0.739
mg/1
12,702
30
8.3
3,280
50
155
2,380
0.258
Ib/day
78,285
185
	
20,215
308
955
14,669
1.59
F
3.958
mg/1
1,795
8
6.9
776
9
30
1,427
1.34
Ib/day
59,252
264
	
25,615
297
990
47,105
44.23
Metric Conversion
kg = (Ib)(0.454)
               TABLE 10  LACSD REFINERY DISCHARGE LEVELS
                          1975-76 (.Typical Day)
                          Category B Refineries
Refinery
Flow, mgd

COD
Suspended Solids (SS)
PH
Ammonia (N)
Oil and Grease
Phenols
Thiosulfate (S)
Chromium (Cr)
A
5
mg/1
774
11
10.6
35
31
32
11
0.78
.513
Ib/day
35,587
506
	
1,609
1,425
1,471
506
36
B
0.530
mg/1
746
49
	
51
~47
14
29
0.9
Ib/day
3,297
217
	
225
208
62
128
4
C
4.028
mg/1
1,093
48
9.8
39
54
76
28
1.05
Ib/day
36,717
1,612
- —
1,310
1,814
2,553
941
35
 Metric Conversion

 kg =  (Ib)(0.454)
                                    321

-------
               TABLE 11  LACSD REFINERY DISCHARGE LEVELS
                          1975-76  (Typical Bay)
                          Category B Refineries
Refinery
Flow, mgd

COD
Suspended Solids (SS)
PH
Ammonia (N)
Oil and Grease
Phenols
Thiosulfate (S)
Chromium (Cr)
D
2.918
mg/1
1,223
53
8.5
547
96
78
18
1.1
Ib/day
29,763
1,290
	
13,312
2,336
1,898
438
27
E
0.703
mg/1
4,150
47
6.8
162
120
178
31
0.09
Ib/day
24,332
275
	
950
704
1,044
182
0.5
F
3.299
mg/1
282
32
7.1
92
5
18
30
0.69
Ib/day
7,759
880
	
2,531
138
495
825
19
Metric Conversion
kg = (Ib)(0.454)
         TABLE  12   LACSD  REFINERY DISCHARGE LEVELS (TYPICAL DAY),
                   Totals of  Refineries  A,  B,  C,  D, E, F
                     (Over 90%  of LACSD  Refinery  Flow)
     Flow
     COD
     Suspended  Solids
     Ammonia  (N)
     Oil  and  Grease
     Phenols
     Thiosulfate  (S)
     Chromium (Cr)
(SS)
                                      1972-73   1975-76   % Reduction
mgd
Ib/day
Ib/day
Ib/day
Ib/day
Ib/day
Ib/day
Ib/day
17.48
394,744
25,063
125,246
21,131
75,833
178,920
137.86
17.00
137,445
4,780
19,937
6,625
7,523
3,020
121.50

65
81
84
69
90
98
12
    Metric Conversion

    kg =  (Ib)(0.454)
                                   322

-------
          SESSION Vil

POWDERED ACTIVATED CARBON


      Chairman

      Thomas P. Meloy

      Director, Industrial and Extractive Processes
      Division, U.S.  E.P.A., Washington,  D.C.


      Speakers

      Francis L. Robertaccio
      "Combined Powdered Activated Carbon Treatment-
      Biological Treatment:  Theory and Results"

      Colin  G. Grieves and Michael  K. Strenstrom

      Joe D. Walk and James F. Grutsch

      "Powdered Activated Carbon Enhancement of
      Activated Sludge for BAT Refinery Wastewater Treat-
      ment"

      Paschal B. DeJohn and James P. Black
      "Case  Histories:  Application of PAC in Treating
      Petroleum Refinery Wastes"

      James  F. Dehnert
      "Case  History:  Use of PAC With a Biodisc-Filtration
      Process"
             323

-------
BIOGRAPHY           Thomas P. Meloy

       Dr. Thomas P. Meloy is Director of the
Industrial and Extractive Processes Division,  Office
of Research and Development,  U.S. Environmental
Protection Agency.  He holds an A.B. and B.S.
respectively from  Harvard and MIT as well as a Ph.D.
from MIT.  Formerly, he was Director of Engineering
Division at the National Science Foundation and
Vice-President for Research and Development at
Meloy Laboratories.  His specialized field is particulate
systems.  This fall he will become the Benedum Professor
at West Virginia University.
                                      324

-------
"COMBINED  POWDERED ACTIVATED CARBON - BIOLOGICAL TREATMENT:
                    THEORY AND RESULTS"

                   Francis L. Robertaccio
                      Senior Engineer
              E.  I, du Pont de Nemours & Co.,  inc.
          Central Research and Development  Department
                      Wilmington, Delaware

     The  purposes of this paper'are:  to acquaint you with an
overview  of some  of the theoretical aspects  of the  combined
powdered  activated carbon-biological treatment process,  and to
present recent start-up experiences and results from the  1.5 x
105 M3/day installation based on the process at the Du Pont
Chambers  Works Plant in Deepwater, New Jersey.

     PACT is Du Pont's name for a patented  process  for purifica-
tion of sewage and/or industrial wastewater  which comprises
subjecting the wastewater to an aerobic biological  treatment
process in the presence of powdered activated  carbon^' (Figure
1).  The  aerobic  biological treatment vessel(s) can have  many
geometric configurations.  Single or multiple  reactors can be
used.  The reactors can be plug flow, completely mixed,  or
somewhere in between.  Powdered activated carbon addition is
compatible with activated sludge, contact stabilization,  or
aerated lagoon systems; that is, any process in which the
carbon can be suspended.  The rate of powdered activated  carbon
addition  for a given wastewater is a function  of the effluent
quality desired.   When the rate of addition  is expressed  in
terms of  weight of carbon added per unit volume of  incoming
wastewater, the rate becomes a function of  the type of carbon
used.

     In addition, certain internal process  controls such  as the
solids retention  time, or sludge age, can be changed to  in-
fluence the rate  of application of a given  type of  carbon to
produce a desired result.(2)  Some of these  relationships are
illustrated in Figure 2.  Here sludge age and  carbon dose are
shown as  variables affecting effluent quality  as measured by
the total organic carbon  (TOC) test-  All data points represent
treatment conditions by which the effluent  BOD of the industrial
wastewater tested was reduced to negligible  concentration.  The
effluent  TOC is shown to be reduced by an independent increase
in either the sludge age or the carbon dose.   Note  that  the im-
provement in effluent quality, by increasing sludge age,  is
less apparent when carbon is absent.  We have  long  postulated
this phenomena results because the adsorbed  microorganisms
have the  sludge age rather than the relatively shorter hydraulic
detention time to biodegrade adsorbed and difficult-to-degrade
molecules.  It is important to recognize the economic advantage
associated with the ability to biodegrade these materials in
the biological reactor as an inherent advantage of  the process.


                              325

-------
The alternative for this type of  effluent  quality improvement  is
a more expensive add-on granular  carbon  adsorption step.

     Table 1 illustrates the effectiveness of  using carbon as
an adsorbent in a biological reactor.(3)   The  table shows the
dissolved organic carbon (DOC)  removal  from a  PACT unit and
comparable data from a biological  unit.  Also  shown is the
combined removal obtained from  a  separate  adsorption of the
biological unit effluent using  the  same  type and quantity of
carbon used in the PACT unit.   The  same  carbon combined with
bacteria in the PACT unit removed  more  DOC and exceeded the
quantity expected from separate isotherm determinations.

     Of course, as more molecules  are biode.graded, adsorption
sites are filled with molecules that are more  biorefractory and
a more rigorous form of regeneration is  needed if the spent
carbon is to be reused.  Alternatively,  more active carbons
 (i. e., higher surface area) can  be used in throw away doses.
Economic considerations grouped as  various capital and operating
expenses dictate the choice.  To  some extent the economics are
strongly influenced by the  carbon  usage  rate,  however, site
specific factors such as the local  costs of alternative sludge
disposal methods must be considered.  At the PACT treatment
facility for Chambers Works we  will thermally  regenerate  carbon
from PACT sludge but wet oxidation  can  also be used.
                                                 !
     The heart of the PACT  system  is a  matrix  of microorganisms
and powdered activated carbon.  Figure  3 shows the matrix.^ '
The photo on the left is powdered  activated carbon in the water;
the photo on the right the  PACT matrix.  The PACT matrix  has
some interesting properties.

     First, the carbon acts as  a  weighting agent.  Sludge
settling rates are vastly improved  as illustrated in the  series
of pictures in Figure 4.  Activated sludge and PACT mixed liquor
were taken from treatability units  operating on the same  waste-
water at the same sludge age.   The  series  of photographs  are
taken at different elapsed  settling times  shown on the timer in
the background.  Note that  the  PACT sludge settles better and
has a clearer supernatant.  The PACT sludge also compacts very
well.  The PACT sludge had  a mixed liquor  suspended solids
concentration of 7700 mg/1  (about  65% carbon)  and ,.a. sludge
volume index of 20 cc/g.  The activated sludge had a mixed
liquor concentration of 2400 mg/1  and a sludge volume index of
46.  We feel that the improved  sludge settling, achieved by
simple carbon addition can  result  in the processing of more
wastewater through existing hydraulically  overloaded treatment
plants.  This is often a viable alternative to rather expensive
capital equipment expansion programs to accomplish similar
results.  Of course, carbon addition will  improve effluent
quality at the same time.   A new  treatment plant can incorporate
                               326

-------
this  advantage  in its design by substantially  reducing  the  size
of secondary  clarification.  Another  illustration  of  the  im-
proved  settling of PACT sludge is  shown  in  Figure  5.  Here  the
initial settling velocity of activated sludge  is compared at
various concentrations to a family of PACT  sludge  curves  at two
carbon  levels and two temperatures.   The  numbers on the PACT
labeled curves  are the application rates  for carbon in  mg/1.

     While  on the topic of sludge  handling,  it should be
mentioned that  PACT sludge dewaters much  more  readily than
conventional  activated sludge.  The manifestation  of  this
property is reduced size of sludge dewatering  equipment.  The
need to dewater more sludge as a result  of  the presence of
carbon  is offset by reduced cycle  times.  Figure 6 compares the
specific resistance of activated sludge  to  two PACT sludges
at different  carbon feed doses. (3)

     A  second property of the PACT sludge matrix is that  it
contains an effective adsorbent.   We  have already  explored  one
aspect  of the role of the adsorbent - removal  of biorefractory
organic compounds - in the discussion of  carbon dose  and  its
relationship  to sludge age.  In that  discussion the effluent
total organic carbon content was a gross  measure of biore-
fractory material.  More specific  measures  of  biorefractory
materials which might require control in  specific  instances
include materials contributing to  final  effluent color, oil
and grease, surfactants, chlorinated  hydrocarbons, phenols  and
toxicity to fish or other trophic  level  measures of toxicity
in receiving  waters.  No matter how you  care to measure,  or are
told to measure these biorefractory materials,it is apparent
PACT can control these substances  to  levels  beyond the
capability of conventional biological systems.  In complex
waste situations control of these  substances at the PACT  treat-
ment plant is often a more viable  alternative  than biological
systems followed by granular activated carbon  columns or  source
treatment.

     Sometimes  biorefractory materials are as  much, or more  of a
problem within  a biological system as they  are in  its final
effluent.  Examples include materials toxic  or inhibitory to
biological reactor microorganisms, materials that  are periodic-
ally present  in high concentrations  (shock  loads), or materials
that cause severe foam, odor, or bulking  sludge.   Unlike  post-
biological separate granular activated carbon  treatment,  the
Presence of carbon in the aerator  often  controls these  in-
process problems as well as those  normally  associated with  the
biologically  treated effluent.  Over  the  years, we have had a
difficult time  sustaining bench scale biological treatability
units on Chambers Works wastewaters due  to  the periodic
Presence of toxic or inhibitory materials.(4)   However, PACT
treatability  units operated in parallel did  not experience


                              327

-------
similar upsets.  As a result  of  this  property of the PACT
process the full scale PACT facility  at  Chambers Works has no
upstream equalization.  We have  even  spiked Chambers Works
wastewater with various toxic  substances including pentachloro-
phenol  (190 1/min. pilot plant test)  and found the PACT system
can sustain efficient operation  when  the conventional biological
system completely fails.  Ferguson, et   al  ,  reported similar
findings in shock loading tests  involving trichlorophenol.(5)

     There are two intertwining  reasons  to  consider any waste
treatment process:  technical  merits  and economics.  We have
touched upon the technical merits  of  the PACT process and
summarized them in Table II.   In most instances any one of the
technical merits may be sufficient reason to  consider PACT.  In
most instances a number of technical  merits must be simulta-
neously applied to the consideration  of  the process at a
specific site.  The resulting  matrix  of  reasons results in a
difficult appraisal of the full  value of the  use of the PACT
process versus alternative processes.  Some of the economic
considerations are shown in Table  III.   At  Du Pont we are con-
vinced PACT is a versatile, economic  and technically viable
process.  We have about 100 man  years experience in PACT process
research and development.  At  the  Chambers  Works facility which
will be described next, we feel  PACT  represents a $7 million
capital savings and a $5 million/year operating cost savings
 (1972 dollars) over the next  best  alternative which was granular
carbon treatment followed by  activated sludge.(6)

     The full scale PACT facility  at  Chambers Works has been in
a start-up phase since mid-November 1976,  and  it  proceeded
smoothly through the coldest  winter in decades.  The liquid
train is on line and the solids  handling train is expected to
be fully operational fairly soon.   During March 1977 a half full
flowrate test, and during early  May a full  flowrate test were
conducted.  This portion of this paper will highlight the start-
up operation and describe results  of  the tests.

     Figure 7 shows the major  components of the PACT portion
of the Chambers Works treatment  plant.   Construction of the PACT
facility started in February  1974  and was completed in December
1976 at an estimated capital  cost  of  $22.5MM.  Primary effluent
is split equally to each of three  15MM liter  aeration tanks as
is the recycled PACT sludge.   Five 1,000 hp blowers supply air
to static mixers in the aerators.   Effluent from the aerators
is conveyed to the clarifier  flowsplitter and then to two
secondary clarifiers.  Treated effluent  (overflow) is dis-
charged to a basin and then to the Delaware River.  Secondary
clarifier underflow is returned  to the aeration tanks via two
2 meter screw pumps.  This part  of the system constitutes the
"liquid train" and includes feed and  unloading facilities for
carbon, phosphoric acid and polymer,
                               328

-------
    Waste  sludge can be removed  from  either  the  mixed  liquor
or recycle  lines and is pumped to a  sludge  thickener  where  it
is settled  to  7 + 2 weight percent solids.  The thickener
underflow serves as a feed stream for  the  filter  press  which is
designed to produce a 35 weight percent  dry solids  PACT cake.
The cake is mechanically conveyed to a five hearth  furnace
where  it is dryed, the biomass incinerated, adsorbed  organics
pyrolyzed and  the powdered carbon in the sludge regenerated for
reuse.   The furnace off-gases pass through  two  scrubbers and
an afterburner before being discharged to  atmosphere.   The  dry,
regenerated carbon recovered off  the bottom hearth  is slurried,
acid washed, and returned to the  carbon  feed  tanks  for  recycle'
to the  process.  No waste sludge  is  produced.   This part of the
system  is called the "solid train".

     The liquid train startup became evident  in November 1976
when 2.7 x  10-> kg of powdered carbon and 2.0  x  105  kg of
bacterial solids were added to one aerator.   Water  temperature
at the  time was 11-15°C.  During  January a  second aerator and
clarifier were brought on line.   During  February  the  carbon
regeneration startup began and during  March the sludge  press
was brought on line.  The more important operating  problems
encountered and solved during startup  have  been presented in
a recent paper by Flynn.*''   The  problems  were  of the type
found with  the startup of a conventional activated  sludge pro-
cess,  that  is, they were not at all  related to  the  uniqueness
of the  PACT process.  These problems went  through the classic
problem solving stages - initial  definition,  questioning of
assumptions, hypothesis forming,  reobservation  of the problem
in some cases, implementation of  a solution and feedback on the
success of  the solution.

     In March  a half-full flowrate test  was conducted.   Table  IV
compares operating conditions, feed  and  effluent  quality for
the full scale PACT facility and  various bench  scale  controls.
The effluent color and dissolved  organic carbon  (DOC) are
important control parameters.  During  this  test,  flowsheet
dosages of  virgin carbon (regenerated  carbon  not  available  at
this date)  reduced effluent DOC to 20  ppm  (43 ppm goal) for the
last seven  days of the test and an average  of 36  ppm  for the
entire test.  Effluent color was  310  (540  goal) despite the
feed color  being 42% over design.  The full scale,  half-flow
test results compare favorably with  the  PACT  bench  scale control.
This table  also presents insight  into  the  improvement in
effluent quality offered by PACT.  Note  the marked  decrease in
effluent DOC and color in the PACT full  scale or  bench  scale
units versus the biological bench scale  unit.

     In early  May a full flowrate test was  conducted.   Table V
Presents operating conditions, feed  and  effluent  quality for
the test.   The effluent color and dissolved organic carbon
                              329

-------
again met goals despite the feed color being  108%  over  design.
Once again this test was conducted using  virgin  carbon.

     In summary, this paper has presented an  overview of  some
of the theoretical aspects of the combined powdered  activated
carbon-biological treatment process  (PACT) and updated  recent
startup experiences from the 1.5 x 10^ M3/day installation
based on this process at the Du Pont Chambers Works.  This
process is a versatile, viable wastewater treatment  technology;
we expect its use to become an accepted solution to  a variety
of existing and future wastewater treatment problems.
                       REFERENCES CITED

1.  U. S. Patent 3,904,518
2.  Flynn, B. P., F. L. Robertaccio and L. T. Barry,
    "Truth or Consequences:  Biological Fouling and Other
    Considerations in the Powdered Carbon - Activated Sludge
    System".   Presented at the 31st Annual Purdue Industrial
    Waste Conference, West Lafayette, Indiana, May 5, 1976
3.  Heath, H. W., "Combined Activated Carbon-Biological
    ("PACT")  Treatment of 40 MGD Industrial Waste" presented
    to Symposium on Industrial Waste Pollution Control at
    ACS National Meeting, New Orleans, LA., March 24, 1977
4.  Robertaccio, F. L., D. G. Hutton, G. Grulich and H. L.
    Glotzer,  "Treatment of Organic Chemical Wastewater with
    the Du Pont PACT Process".  Presented at A. I. Ch. E.
    National Meeting, Dallas, Texas, February 1972
5.  Ferguson, J. F., G. F. P. Keoy, M. S. Merrill and
    A. H. Benedict "Powdered Activated Carbon - Biological
    Treatment:  Low Detention Time Process" presented at
    the 31st Annual Purdue Industrial Waste Conference, West
    Lafayette, Indiana, May 4, 1976
6.  Flynn, B. P. and L. T. Barry "Finding a Home for the
    Carbon:  Aerator (Powdered) or Column  (Granular)".
    Presented at the 31st Annual Purdue Industrial Waste
    Conference, West Lafayette, Indiana, May 1976
7.  Flynn, B. P. "Operating Problem Solving During a
    Secondary-Tertiary Treatment Plant Start-Up".  Presented
    at llth Mid-Atlantic Regional ACS Meeting, University
    of Delaware, Newark, Delaware, April 22, 1977
                              330

-------
DISCUSSION

Ed Sebesta,  Brown S Root:  What is  the  design  hydraulic  loading
   for secondary clarifiers at the  Chambers  Works  wastewater
   treatment facility?
Robertaccio  and B. P. Flynn, Du Pont;   The solid  flux  rate  is
   designed  at 250 Ibs. per day per  square foot which  is kind
   of high.   At that solid flux rate,you  should be  able  to  get
   an underflow concentration of  3-1/2  weight  percent.   In
   full scale testing we have been  able to generate 7-1/2
   weight percent solids which means we could  operate  without
   a waste sludge thickener and could  feed our filter  press
   directly  from our return sludge  line.  The  hydraulic  over-
   flow rate is in excess of about  1000 gallons/day/ft.2.   We
   have two  secondary clarifiers  but could send full flow
   through one secondary clarifier.

Leonard W. Crame, Texaco;  What does the  Du  Pont  PACT  process
   patent mean to the refining industry in terms  of using this
   process?
Robertaccio;  Du Pont will license  any  user  of the  process.
   The royalty rate will be reasonable  in order to  encourage
   use of the process.

J. E. Rucker, API;  Please comment  on  economics and feasibility
   of regeneration of powdered carbon  from PACT sludge.
Robertaccio:  Economics  first.  We  think  that  powdered carbon
   can be regenerated for an operating  cost  of about 5C  a
   pound.  Capital costs would depend  on  the size  of the
   facility  and the method used to  annualize capital costs.  At
   Chambers  Works capital costs would  add another  5C a pound.
   Now feasibility.  We put as much  effort into the regenera-
   tion part of the Chambers Works  facility  as we  did  to the
   PACT process.  The regeneration  system is being  brought  on
   line.  We have had some mechanical  problems but  we  don't
   expect to have any more trouble  solving these  as we had
   solving other problems.  Of course,  a  number of  thermal  and
   wet oxidation regeneration equipment manufacturers  will  tell
   you they  think regeneration of powdered activated carbon
   from PACT sludge is no problem.

Dave Skamenca, Envirotech:  Did you  pilot test mixing  the
   powdered  activated carbon - biomass  mixture with static
   mixers?
Robertaccio;  Yes.
                              331

-------
BIOGRAPHY           Francis L. Robertaccio

      Fran is a senior engineer for the DuPont Company
in Wilmington,  Delaware.  He has 12 years experience
in various industrial pollution control positions.  He
holds a B.S. and M.S.  in Chemical Engineering from
Clarkson College and a Ph.D.  in Environmental
Engineering  from the University of Delaware.  He has
authored a dozen papers on industrial pollution control/
holds several U.S.  and  foreign patents and is a member
of AlChEand WPCF.
                                     332

-------
TABLE I   -   SYNERGISTIC EFFECT ON  DOC  REMOVAL  WITH  "PACT"
DOC, mg/1

FEED TO
UNITS
183
178
167

BIO UNIT
EFFLUENT
80
70
79
BIO UNIT
EFFLUENT + BATCH
CARBON ADSORPTION (D
59
42
55

"PACT" <2)
EFFLUENT
44
18
25
TRIAL

  1
  2
  3

 (1)  Take 500 cc filtered Bio Unit  effluent,  add 150 ppm
     virgin carbon, stir 3  hours  at room temperature, filter,
     and analyze filtrate for DOC
 (2)  "PACT" unit operating  at 20°C  with 8.0 day sludge age
     and 160 ppm carbon addition
                              333

-------
TABLE II - REASONS TO CONSIDER PACT
    Existing Biological Treatment Plants
    A.  Need to improve effluent quality
           treatment plant hydraulically overloaded
           biological treatment unstable
           odor, foam
           new, more restrictive effluent limits
           biological treatment inherently incapable  of
            removing some control parameters
        •  water quality limitations not met in geographic
            area
    B.  Need to r.educe cost
        •  rising sludge disposal costs
        •  expensive chemicals used to aid biological  treatment
    C.  Need to expand treatment plant
        •  want to accept new customers
        •  currently overloaded
        •  want to accept new product's wastewater; afraid
            biological process will become unstable or in-
            capable of removing new waste constituents
        •  can no longer use off plant sludge disposal site
    D.  This treatment plant will eventually be abandoned
         (i. e., to join regional plant) but I have to  get the
        most out of what I have
 II  New  (potentially biological) Treatment Plants
    A.  Want cost effective process
    B.  Concern about efficiency of biological treatment
           have potentially toxic waste
           face strict effluent limits
           want stable process
           have wastewater from changing product line
           future regulations might outdate biological
            treatment capabilities
    C.  Have components in waste not currently regulated, but
        want them removed now.
    D.  Limited amount of land available for treatment
    E.  Want to minimize sludge disposal problems
        •  have undesirable components in waste that  will
            be concentrated in sludge; don't want these
            released to environment
        •  don't have land, or availability of ocean  disposal
 III New Advanced Waste Treatment Plants
    A.  Want cost effective process
    B.  Want flexibility to alter treatment plant
        •  as regulations change
        •  as product mix dictates different treatment
            need -
                    - over short intervals
                    - over a long period
    C.  Concerned about stability of alternate advanced
        treatment processes
                              334

-------
    D.   Want to minimize investment
IV  Miscellaneous
    A.   Pretreatment plus municipal disposal  route too
        expensive
    B.   PACT is a low risk process compatible with many
        existing waste processing schemes,  changes in
        product mix, or changes  in regulations
 TABLE III - ECONOMIC  CONSIDERATIONS OF POWDERED ACTIVATED
            CARBON  ADDITION	
 PRO                             CON
    Eliminates  granular carbon  •  Powdered Carbon Cost
    adsorption  equipment needs,     - virgin
    including initial  GAC            about 0.5-0.8C/1000 liters/
    inventory                         lOppm
                                     using 55  to 80£/kg carbon
                                •  Regenerated  (full cost*
                                   - about 0.1-0.2C/1000
                                     liters/lOppm
    Minimizes need  for equaliza-
    tion  facilities to control
    wastewater  variability
    Eliminates  separate second-
    ary  sludge  disposal if re-
    generation  is  used
    Reduces  or  eliminates need  •  May require use of flocculant
    for  antifoam,  odor control
    Protects biological system
    from inhibition or toxic
    upset
    Reduces  size  requirements
    for  secondary  sludge
    settling, thickening,
    dewatering
    Carbon  addition rate
    readily  changed for changes
    in wastewater  character-
    istics  or regulations
                               335

-------
TABLE  IV  -  CHAMBERS WORKS HALF FULL FLOWRATE  TEST
            TEST PERIOD 3/13/77-3/26/77  INCLUSIVE
Operating  Conditions
  Carbon  Dose(ppm)
  Aeration  Temp (°C)
  Hydraulic Residence
   Time  (hrs)
  Sludge  Age (days)
Feed  Quality
  BOD-Soluble  (mg/1)
  DOC  (mg/1)
  Color  (APHA)
Effluent Quality
  BOD-Soluble  (mg/1)
  DOC  (mg/1)
  Color  (APHA)
FULL SCALE
PACT PLANT

   182
    22

    14.6
   304
   214
  1416

    15.2
    35.7
   311
 BENCH SCALE CONTROLS
        CONVENTIONAL
PACT    BIOLOGICAL

 150         0
  22        22
   7.5
   8

 304
 214
1416

  19.3
  28.4
 369
   7.5
   8

 304
 214
1416

  13.8
  67.3
1900
 *no steady state material balance available
TABLE V - CHAMBERS WORKS FULL FLOWRATE  TEST
          TEST PERIOD 4/26/77-5/6/77  INCLUSIVE

          Operating Conditions
           Carbon Dose  (ppm)                189
           Aeration Temp  (°C)                28.5
           Hydraulic Residence Time  (hrs)     7.5
          Feed Quality
           BOD - Soluble  (mg/1)             300
           DOC -  (mg/1)                     214
           Color  (APHA)                    2080
          Effluent Quality
           BOD - Soluble  (mg/1)               9.6
           DOC -  (mg/1)                      28
           Color  (APHA)                     490
                              336

-------
                      Powdered Activated
                      Carbon
  Sewage
  and/or
  Industrial
  Wastewater
Aerobic
Biological
Treatment
Product
Water
to
                      Gas Containing
                      Oxygen
Fig.  1  A SIMPLE ILLUSTRATION  OF PACT

-------
                    .20
u>
OJ
00
                    * Numbers shown in the bottom face of the exploded cube are
                    applied carbon doses, mg/liter

                   Fig. 2 EFFECT OF SLUDGE AGE AND CARBON DOSE ON

                              EFFLUENT TOG AND APPARENT LOADING

-------
Left:
  Virgin Carbon in Water
            h—	1
                   Right:
                     PACT Sludge
Fig. 3 PHOTOMICROGRAPHS

-------

                    PACT
                   SLUDGE
                                  PACT
                                  ACTIVATED SLUDGE
             ACTIVATED
              SLUDGE
        PACT
       SLUDGE
        TKBp
    INTIAL
 SOLIDS. M6/L
  T708 1 896
  2412 £  150
ftCTIVATED  PACT
 SLUDGE  SLUDGE
SLUDGE VOLUME
INDEX. CC/GRAM
    20 ±2
    461 3
    ACTIVATED  PACT
     SLUDGE  SLUDGE
                                 ACTIVATED
                                  SLUDGE
uo
t-
o
 PACT
SLUDGE
ACTIVATED
. SL,U5?E SLUDGE
                                                    ACTIVATED
 PACT
SLUDGE
                    ACTIVATED   PACT
                     SLUDGE  SLUDGE
                         ACTIVATED
                          SLUDGE
    NOTE:  TIMER HANDS MOVE COUNTERCLOCKWISE.  READ ELAPSED TIME USING  SMALL  NUMBERS.
           LONG HAND INDICATES MINUTES, SHORT HAND INDICATES SECONDS.
           FOR EXAMPLE,  LOWER LEFT  TIMER READS I  MINUTE, 21 SECONDS.
           BOTH SLUDGES FROM TREATABILITY  UNITS  AT SAME SLUDGE AGE, HYDRAULIC DETENTION  TIME.

                                 .  Fig.4  COMPARATIVE SETTLING CHARACTERISTICS
                                             OF  PACT AND  ACTIVATED SLUDGE


-------
                          1.O
       Initial Settling Velocity,   n 1
                 Ft/Min
CO
                         0.01
PACT-ISO (winter, 10°C)
                                                                          I   I  I
                           0-1                           1.0
                                             Suspended Solids Concentration, %

                                 Fig. 5  SETTLING CHARACTERISTICS OF PACT
                                                & ACTIVATED SLUDGES
                     10

-------
              7.0
              6.0
              5.0
Bio Unit
C  =  .0047
M  =  .137
"r" =  128x1012 —
             gm
M  =  Slope t/v vs v
C  =  gm solids Ice filtrate
P  =  Filtration Pressure, 1.5 x 107
A  =  Filter Area, 38.3 cm2
V  =  Filtrate Viscosity, .010 poise
2PA2
 ™   = 4.40 x 1012
N>
             4.0
             3.0 -
             2.0 -
             1.0 -
                                             55 ppm "PACT"
                                             C  =  .0167
                                             M  =  .022
                                             "r" =  5.8x1012|£
                                   50
                                 100                 150
                                   Filtrate Volume in Cm3
                                                                      110 ppm "PACT"
                                                                      C  = .028
                                                                      M  = .0093
                                                                      "r" = 1.5x
                                                                                                       , cm
                                                                                                        gm
                                             200
250
                                  Fig. 6  SPECIFIC RESISTANCE, "r" OF  PACT vs
                                                       BIOLOGICAL SLUDGE

-------
   Carbon ;
   Unloading, Storage
      and Feed
Primary
Effluent
           -M
                     Five 1,000hpl
                       Blowers   I
                          I Air
Three
Parallel
Aerators

-/



                                        Polymer Unloading,
                                         Storage and Feed
                                          o-
                                                     Two
                                                   Parallel
                                                   Clarifiers
                            a
                               Two Parallel
                              Screw Pumps
                                                        Return
                                                       'Sludge"
                                 Waste Sludge

                                  "Liquid Train"
                                                                                 To Basin
                                                                                'and River
 yste
idge
kener
    To Primary
    Treatment
To Carbon Slurry
 Storage Tanks *
                                 Fuel Oil
                            Unloading, Storage
                                and Feed




Filter Procc



1

^
' 1
                                                                        Afterburner
                                 Dilution
                                 Water
               Carbon
               Slurry
                                L_L
                                          Acid
                                        Contactor
                                                      Multiple
                                                      Hearth
                                                      Furnace
I Quench Tank
                               "Solid Train"
                                                           HCI Unloading,
                                                          Storage and Feed
    Fig. 7 DUPONT  PACT  PROCESS:  CHAMBERS WORKS

-------
               POWDERED ACTIVATED CARBON ENHANCEMENT OF ACTIVATED
                 SLUDGE FOR BATEA REFINERY WASTEWATER TREATMENT

                                Colin G. Grieves
                      Research Engineer, Amoco Oil Company

                              Michael K. Stenstrom
                      Research Engineer, Amoco Oil Company

                                   Joe D. Walk
                Project Director, Standard Oil Company (Indiana)

                                James F. Grutsch
      Coordinator of Environmental Projects, Standard Oil Company (Indiana)
ABSTRACT

Pilot plant studies show that powdered activated carbon enhancement of
activated sludge is a viable alternate to and less costly substitute for
granular carbon tertiary treatment of refinery wastewaters.  Effluent quality
depends upon both the equilibrium concentration and the surface area of the
powdered carbon in the activated sludge mixed-liquor.

Operation at very high sludge ages—60 days or more—allows the carbon to
accumulate to high concentrations in the mixed-liquor even though only small
make-up amounts are added to the system.  Also, carbons with a high surface
area are especially efficient in adsorbing contaminants.  Consequently,
costly regeneration may be unnecessary because the spent carbon can simply be
discarded with the waste sludge.  Powdered carbons may thus eliminate the
need for the add-on granular carbon adsorption process that the Environmental
Protection Agency has recommended for meeting proposed 1983 standards for
Best Available Technology Economically Achievable (BATEA).

INTRODUCTION

According to the EPA guidelines for treating refinery wastewaters  , the
sequence shown in Figure 1 is recommended for meeting 1977 standards for Best
Practical Technology Currently Available (BPTCA).  For meeting 1983 goals for
Best Available Technology Economically Achievable (BATEA), the guidelines
recommend an add-on process using granular carbon adsorption.  However, this
approach may be both inefficient and very costly.  So far as is known, its
effectiveness has never been adequately demonstrated.  Moreover, preliminary
estimates indicate that capital and operating costs for the granular carbon
adsorption and regeneration facilities may equal or exceed those of the
entire current activated sludge process.
                                              9—7 c
By contrast, both patents and research studies     indicate that powdered
activated carbon may be a practical and economical substitute for granular
carbon.  For example, powdered carbon costs only about one-half as much as
# References inserted at end of text.

                                    344

-------
granular—$0.65/kg  versus $1.20/kg15.  In addition, recent studies have shown
that  powdered  carbon can be added directly to the mixed-liquor in activated
sludge  aeration  tanks Z1» 22» 23» 24.  Thus, appropriate alterations in
operating procedures may eliminate the need for regeneration by making it
economically feasible to discard the spent carbon with the waste sludge.

In general,  the  cost effectiveness of a powdered carbon process increases
with the concentration of carbon maintained in the mixed-liquor.  A mass bal-
ance of such a process is represented by the following equation:

                                                  (1)
where

     C  = Equilibrium mixed-liquor carbon concentration   (mg/1)
     Ci = Influent carbon concentration                   (mg/1)
     9C = Sludge age                                      (days)
     9jj = Hydraulic retention time in the aeration  tank   (days)

 Equation 1 reveals that the equilibrium mixed-liquor  carbon  concentration is
 proportional to the product of the influent  carbon  concentration  (carbon
 dose) and the sludge age.  Thus, equilibrium carbon concentration can be
 increased by increasing the carbon dose, or  the  sludge age,  or both.  There-
 fore, to keep carbon costs to a minimum, it  is desirable  to  operate at as high
 a sludge age as possible and not at an excessively  long hydraulic retention
 time.

 A possible drawback to operation at a high sludge age is  the increased risk
 that toxic, inhibitory, or inert materials will  build up  in  the aeration
 tank.  For example, a build-up of oily solids could reduce the oxygen trans-
 fer efficiency and inhibit both the nitrifying and  organic carbon utilizing
 organisms.  The dissolved oxygen concentration in the mixed-liquor could
 also become too low for effective nitrification, and  the  final clarifiers
 could become overloaded.  Therefore, it is desirable  in the  pretreatment step
 to remove as much solid material as possible from the wastewater  before it
 enters the aeration tank.

 To evaluate the effects of such variables in a process using powdered carbon,
 an extensive 15 month four-phase pilot plant study  was carried out at Amoco
 Oil Company's Texas City refinery.  Pilot plants operating in parallel with
 the refinery activated sludge process facility were fed the  same  wastewater
 for treatment.  Specific variables investigated  were:

     Carbon type, including surface area and pore volume

     Carbon addition rate

     Sludge Age

     Pretreatment of feed to remove oil and  solids
                                     345

-------
EXPERIMENTAL EQUIPMENT

Figure 2 shows the configuration of  the pilot  plants.   Each had a volume of
42 liters, and as many as eight units were  operated in parallel during  por-
tions of the study.  They were housed in  a  rain-tight  enclosure but  were
neither heated nor cooled.  Thus, the temperature of the mixed-liquor varied
from 4°C to 31°C.

Operating conditions and analytical  procedures are summarized in Table  1.
The pH was checked daily and controlled by  addition of caustic at a  constant
rate.  Dibasic potassium phosphate,  K2HP04, was added  to satisy the  phos-
phorus requirement of the microorganisms.

The wastewater feed, a slipstream from the  pressure filters of the refinery
treatment plant, was passed through  a pilot gravity sand filter before  being
fed to the pilot plants.

Table 2 summarizes the characteristics of the  five powdered carbons  evaluated.
Amoco's experimental high-surface-area carbons are designated as Al  and A2,
PX-21 and PX-23, respectively.  Those designated as B, C,  and D are  commer-
cially available carbons having a much lower surface area.   Carbon A2 (PX-23)
has the highest pore volume.

Effectiveness was judged on the basis of  the following effluent standards
proposed for a BATEA facility1:

                                                   Concentration,
                                                     mg/liter

     Total Organic Carbon  (TOC)                         15
     Chemical Oxygen Demand (COD)                       24
     Ammonia  (NH3-N)                                     6.3
     Phenolics                                           0.02

These standards are for a Class "C"  refinery and are based on the guideline
effluent flow rate of 0.46 m^/m^ of  crude throughput per stream day  (19 gal/
bbl).  Because the BATEA treatment sequence will undoubtedly result  in very
low concentrations of effluent suspended  solids, only  the soluble components
of the effluent were measured.

To obtain high sludge ages, effluent suspended solids  were allowed to settle
in 30-gallon plastic containers and  then  were  returned to the pilot  plants
periodically.  At any given sludge age, all plants were allowed to reach
steady-state operation over an extended period of time.  Then performance
data were taken over a 30-day period.

RESULTS AND DISCUSSION

The four phases of the study were carried out  in sequence, with the  design of
succeeding phases based on the results of the  preceding ones.  In summary,
they examined:
                                     346

-------
    phase                     	      Obi active           	

     I                        Effect of carbon type at an addition rate of
                              100 mg/liter and a sludge age of 20 days with
                              prefiltered feed.

     II                       Effect of carbon type at an addition "rate of
                              200 mg/liter and a sludge age of 20 days with
                              prefiltered feed.

     Ill                      Effect of increasing sludge age to 60 days and
                              reducing carbon addition rate to 25 mg/liter
                              with unfiltered and prefiltered feed.

     IV                       Effect of further increasing sludge age to 150
                              days while reducing carbon addition rate to
                              10 mg/liter.

Phases  I and II

The results of  Phases I and II, summarized in Table 3, indicate that powdered
activated  carbon significantly enhances the performance of a refinery
activated  sludge process.   Improvement in the quality of the effluents from
carbon-fed plants ranged from 65% for soluble organic carbon up to 95% for
phenolics.   At  the 200 mg/liter addition rate, the results usually satisfied
the BATEA  effluent quality goals.  The high surface area carbon AT was
significantly more effective than the other three.  The commercially avail-
able carbon B produced slightly better effluent than carbon C, which would be
expected if efficiency is proportional to surface area.  Because nitrifica-
tion was essentially complete in the control unit, carbon addition could not
improve ammonia conversion.  Carbon D, which is derived from wood charcoal
and has a  significantly lower pore volume than the others, performed so
poorly  in  Phase I that it was dropped from further consideration.  The per-
formance of carbon Al at 100 mg/liter dose was about as effective as carbon
B at 200 mg/liter, or about twice as effective as the best commercially
available  carbon tested.

Phase III

Table 4 shows the effects of sludge age and feed'filtration upon performance.
The plant  with  filtered feed performed better than one with unfiltered feed,
and a sludge age of 60 days was better than one of 20 days.  No deterioration
in the  settling characteristics of the mixed-liquor suspended solids was
observed at this higher sludge age.

At a sludge age of 20 days the plant with filtered feed performed marginally
better  than the one with unfiltered feed.  Undoubtedly, greater differences
in effluent quality would have been observed ;Ln a plant operated at a sludge
age of  60  days  with unfiltered feed.  (Not recorded in these data, however,
is the  complete failure of the plant fed unfiltered feed shortly after
cessation  of data gathering for this steady-state period.)
                                    347

-------
Table 4 also shows how pore size and surface area affect  the  performance of
the carbons.  Carbons Al and A2 have approximately  the  same surface area, but
carbon A2 has much larger pores.  Yet, at an equivalent addition  rate of 50
mg/liter, both carbons showed about the same performance.  Thus,  large pore
diameters are not required for effective treatment  of this refinery waste-
water.  Moreover, plants fed 50 mg/liter of either  Al or  A2 performed much
better than the plant fed 100 mg/liter of carbon B.  In fact,  these high-
surface-area carbons are between two and four times more  effective than
carbon B in enhancing SOC and soluble COD removal.

A comparison of the data in Tables 3 and 4 shows that a low carbon dose and a
high sludge age enhance an activated sludge process almost as much as do a
high carbon dose and a low sludge age.

It is possible that the difference in performance is solely due to difference
in temperature between the phases—mean operating temperature during Phase
III was only 14°C, whereas during Phases I and II temperature averaged 31°C
and 25°C, respectively.

Also observed during the lower operating temperature of Phase III was an
increase in the ammonia removal efficiency of the carbon-fed pilot plants.
This phenomenon was unexpected because activated carbon does not normally
adsorb ammonia.  Possibly, the increased removal rate is due to the adsorp-
tion of potentially toxic or inhibitory organic materials which would reduce
the rate of nitrification if left in solution.  The control plant in Phases I
and II had little difficulty in achieving full nitrification, perhaps because
of the higher temperature.

Phase IV

As shown in Table 5, Phase IV was designed to push  the  activated sludge sys-
tem to the limit by increasing sludge age to 150 days and decreasing carbon
addition to 10 mg/liter.  Further, in one of the plants, hydraulic retention
time was reduced to 7.5 hours, compared with 15 hours in  the other plants.

Despite similarities in influent quality during all four phases, during
Phase IV the effluent SOC and COD of the control increased by about 30-35%
over that observed during the first three phases, despite a mean temperature
of 27°c (c-f• 14°C during Phase III).  All pilot plants essentially nitrified
completely.

Remarkably, however, the plant with 10 mg/liter of  high surface area carbon
Al at a sludge age of 150 days produced an effluent whose soluble organic
carbon concentration was 50% lower than that of the control reactor and
slightly lower than that of all of the other pilot  plants.  The plant dosed
with 25 mg/liter carbon Al, with one-half the hydraulic capacity of the other
plants, produced the second best effluent.

The outstanding performance at a sludge age of 150  days indicates that
refinery activated sludge processes can be operated with  very  little added
                                    348

-------
carbon.  The  dose may be low enough so that the carbon need not be regen-
erated but be discarded with the waste activated sludge.  At a very high
sludge age, there will be smaller quantities of waste sludge to be disposed
of.

The data in Table 5 also indicates that powdered carbon can be used to
increase the  hydraulic capacity of an activated sludge plant, as proposed by
others13, or  to increase the effluent quality of an overloaded plant.  The
carbon-fed plant that operated at one-half the hydraulic retention time of
the control produced an effluent 50% better than that of the control.  Exper-
ience with pilot activated sludge plants operated at several of Amoco's other
refineries has shown that conventional activated sludge processes cannot be
operated successfully with a hydraulic residence time of only 7% hours.

Status  of Powdered Carbon Enhancement of Activated Sludge

The data from Phase IV indicate that the limits of the powdered carbon
enhanced activated sludge process have not been reached.  In addition, more
data are needed before economic studies can be made to weigh the possible
options for  achieving a given effluent quality:  high fresh carbon dose at
moderate sludge age (20-60 days) with regeneration of spent carbon; low
fresh carbon dose at high sludge age (60-150 days) with no regeneration of
spent carbon.  Cost analyses should be made for each of these extreme options,
and several  intermediate ones, and compared with those for tertiary treatment
with granular carbon technology.

Figure 3 shows the qualitative curves this pilot study has generated.  Of
course, the  one for the 150-day sludge age is purely speculative because
only one data point exists.  However, the trend of the data does show that
effluent quality is a function of mixed-liquor carbon concentration.  The
curves are probably asymptotic to a residual organic carbon concentration,
but over the range investigated an increase in mixed-liquor carbon concentra-
tion causes  a decrease in effluent soluble organic carbon.  Furthermore, the
relationship  between effluent quality, sludge age, and carbon dose is clearly
non-linear.   For example, to achieve an effluent quality of 12.5 mg/liter of
soluble organic carbon, the three options are:  100 mg/liter of carbon at a
sludge age of 20 days; 47 mg/liter of carbon at a sludge age of 60 days; 24
mg/liter of  carbon at a sludge age of 150 days.  If the relationship were
linear, the values calculated from a base case of 100 mg/liter at a 20-day
sludge age would be 33 mg/liter and 13 mg/liter at 60 days and 150 days,
respectively.

Apparently,  the process loses effectiveness because of incomplete microbial
regeneration.  Microbial regeneration of the spent carbon is probably not as
effective as  using fresh carbon; some materials adsorbed by the carbon are
undoubtedly  non-biodegradable, even after 150 days of contact with micro-
organisms  in  the pilot plant.  The ability to retain significant effective-
ness even at  150 days is the key to cost effective high sludge age operation
with powdered activated carbon.  Of course, there may be other reasons why
carbon loses  effectiveness at high sludge age, such as production of cell
lysis products which are then adsorbed by the carbon.
                                    349

-------
Effluent Variability

Variation in effluent quality over a 30-day  (or longer) period  is  extremely
important.  The EPA1 has set the daily maximum variability  equivalent to the
99% probability value and the 30-day maximum variability  to the 98%  level.
For BATEA the daily maximum variability factors for TOC,  COD, NH3~N, and
phenolics are proposed at 1.6, 2.0, 2.0, and 2.4, respectively.  The 30-day
maximum values are 1.3, 1.6, 1.5, and 1.7, respectively.

Figures 4, 5, 6, and 7 show probability data for the 30-day operating periods
during Phase III.  Table 6 shows the daily maximum  (99% probability) and
30-day maximum (98% probability) variability factors calculated from these
figures for the plant fed with 25 mg/liter of Carbon Al.  The EPA  guideline
values are also given.  The actual variability factor was calculated as the
99%  (or 98%) probability value divided by the target quality value.  In
general, the variability in effluent quality was higher than the guideline
values.

It is  important to notethat the proposed guideline variability factors are
unrealistic.The data base used by EPA1 for their production was  obtained
from limited pilot studies.-• In addition, BPTCA 30-day maximum  (98% prob-
ability) values were used as the BATEA 30-day maximum values.   Variability
factors will undoubtedly have to be amended  before BATEA  goals  become BATEA
standards.

SUMMARY AND CONCLUSIONS

A viable alternative to granular activated carbon tertiary  treatment of
refinery activated sludge effluent for meeting proposed 1983 BATEA effluent
quality standards has been demonstrated.  The proposed process  involves add-
ing  powdered activated carbon to the aeration tank of the activated sludge
process, achieving cost effectiveness by operating at a very high  sludge age
and  a  low carbon dose.  Effective removal of oil and colloidal  solids in
the  pretreatment step is necessary for successful operation.

Effluent quality depends upon both the equilibrium mixed-liquor carbon
concentration and the surface area of the carbon.  An experimental carbon
with a high surface area appears to be several times more effective than the
best commercial carbons in achieving an effluent quality  standard.  Pore
size of the activated carbon had no apparent effect upon  effluent  quality.

In general, the process can be used to meet  only the long-term  average
effluent quality proposed for BATEA.  Daily  maximum and 30-day  maximum var-
iability goals, as presently defined cannot  be met.

The  proposed process also enhances nitrification at low temperatures and
dampens effects of increased hydraulic flow  rate on the activated  sludge
factors.  Both phenomena will help to decrease effluent variability.
                                    350

-------
REFERENCES

 1.  U.S. Environmental Protection Agency,  (Draft) Development Document for
    Effluent Limitations,  Guidelines and New Source Performance Standards
    for  the Petroleum Refining Point Source Category, USEPA.  Washington.
    B.C.,  20460  (April, 1974).

 2.  C. P.  Derleth,  U.S. Patent 1,617,014,  February 8, 1927.

 3.  N. Statham,  U.S.  Patent 2,059,286, November 3, 1936.

 4.  F. L.  Robertaccio, D.  G.  Button, G. Grulich, and H.  L.  Glotzer,  "Treat-
    ment of Organic Chemicals Plant Wastewater with the DuPont PACT
    Process."  Presented at AIChE National Meeting, Dallas,  Texas, February
     20-23, 1972.

 5.  A. D.  Adams, "Improving Activated Sludge Treatment with Powdered Acti-
    vated Carbon."   Proc.  28th Annual Purdue Industrial Waste Conference,
     1972.

 6.   F. L.  Robertaccio.  "Powdered Activated Carbon'Addition to Biological
     Reactors."  Proc. 6th Mid Atlantic Industrial Waste Conference,
    University of Delaware, Newark 1973.

 7.   B. P.  Flynn, "Finding a Home for the Carbon Aerator (Powdered) or Column
     (Granular)." Proc. 31st Annual Purdue Industrial Waste Conference,
     1976.

 8.   A. B.  Scaramelli and F. A. DiGiano, "Upgrading the Activated Sludge
     System by  Addition of Powdered Activated Carbon."  Water and Sewage
     Works, 120:   9, 90, 1970.

 9.   0. Hals and  A.  Benedek.  "Simultaneous Biological Treatment and  Acti-
     vated Carbon Adsorption."  Pres. 46th Annual Water Pollution Control
     Federation Conference, Cleveland, 1973.

 10.   A. A.  Kalinske, "Enhancement of Biological Oxidation of Organic  Waste
     Using Activated Carbon in Microbial Suspensions."  Water and Sewage
     Works.  115;  7, 62, 1972.

 11.  P. Koppe,  et al, "The Biochemical Oxidation of a Slowly Degradable
     Substance  in the Presence of Activated Carbon:  Biocarbon Unit."
     Gesundeits Ingenieur (Ger) 95.:  247, 1974.

 12.  A. D.  Adams, "Improving Activated Sludge Treatment with Powdered
     Activated  Carbon."  Proc. 6th Mid Atlantic Industrial Waste Conference.
     University of Delaware, Newark, 1973.

 13.   J. F.  Ferguson, G. F. P. Keay, M. S. Merrill, A. H. Benedict, "Powdered
     Activated  Carbon-Biologica] Treatment:  Low Detention Time Process."
     Paper presented etc the Jist Annual Industrial Waste Cunleieuce.   IUJ.UUB
     University,  Lafayette, Indiana.  1976.

                                    351

-------
14.  F. B. DeWalle and E. S. K. Chian, "Biological Regeneration of Powdered
     Activated Carbon Added to Sludge Units."  Water Research 11;  439, 1977.

15.  F. D. DeWalle, E. S. K. Chian, E. M. Small, "Organic Matter Removal by
     Powdered Activated Carbon Added to Activated Sludge."  Journal Water
     Pollution Control Fed.  ^9_:  593. 1977.

16.  Anon, "Chemenator" Chemical Engineering £4:  1, 35.  1977.

17.  P. B. DeJohn, "Carbon from Lignite or Coal:  Which is Better?"  Chemical
     Engineering.  82:9, 13.  1975.

18.  D. G. Hutton and F. L. Robertaccio, U.S. Patent 3,904,518,
     September 9, 1975.

19.  P. B. DeJohn and A. D. Adams, "Treatment of Oil Refinery Wastewaters
     with Powdered Activated Carbon."  Pres. at the 30th Annual Purdue
     Industrial Waste Conference.  1975.

20.  J. A. Rizzo, "Use of Powdered Activated Carbon in an Activated Sludge
     System."  Proceedings of th&iOpen Forum on Management of Petroleum
     Refinery Wastewaters, January 26-29, 1976, EPA, API, NPRA, University
     of Tulsa, at the University of Tulsa, Tulsa, Oklahoma.

21.  M. K. Stenstrom and C. G. Grieves, "Enhancement of Oil Refinery Acti-
     vated Sludge by Additon of Powdered Activated Carbon."  Pres. at 32nd
     Annual Purdue Industrial Waste Conference, 1977.

22.  C. G. Grieves, M. K. Stenstrom, J. D. Walk, and J. F. Grutsch, "Effluent
     Quality Improvement by Powdered Activated Carbon in Refining Activated
     Sludge Processes."  Pres. at the 42nd Mid-year Refining Meeting, API.
     Chicago, Illinois, May 9-12, 1977.

23.  G. T. Thibault, K. D. Tracy, J. B. Wilkinson, "Evaluation of Powdered
     Activated Carbon Treatment for Improving Activated Sludge Performance."
     Pres. at the 42nd Mid-year Refining Meeting, API.  Chicago, Illinois,
     May 9-12, 1977.
                               f
24.  L. W. Crame, "Pilot Studies on Enhancement of the Refinery Activated
     Sludge Process."  Pres. at the 42nd Mid-year Refining Meeting, API.
     Chicago, Illinois, May 9-12, 1977.

25.  B. R. Kim, V. L. Snoeyink, F. M. Saunders, "Influence of Activated
     Sludge CRT on Adsorption.1!  Environ. Engr. Div., ASCE, 102;  55.  1976.

26.  J. F. Grutsch and R. C. Mallatt, "Optimize the Effluent System."
     Hydrocarbon Processing.  55;  3:  105-112.

27.  U.S. Environmental Protection Agency, Methods for Chemical Analysis
     of Water and Wastes, USEPA, Washington, D.C., 1974.
                                    352

-------
DISCUSSION

Piysuch Shah, Exxon Research and  Engineering Co.;  Would you please comment on the
effects of toxicant build up and on the performance of aged activated sludge units,
especially 100-150 days?  Also, what is the maximum concentration that can be allowed
in the feed?

Grieves;  Assume a very high sludge age of 150 days, a hydraulic residence time of 12
hours, 1 mg/liter of a toxicant (for example, chromium) in the feed, and 100% removal
of it by the activated sludge.  At steady state,  chromium concentration would build up to
300 mg/liter, which, in all likelihood, would be toxic to the microorganisms.  However,
we have data to indicate that even at 150 days sludge age, chromium does not accumu-
late to more than 30-50 mg/liter in the sludge.  We  certainly have not observed any
effects of toxicant build-up — on the contrary,  the  150-day sludge age reactor is the
most effective unit.

          As for other toxicants — for example oil and grease and inert suspended solids
-- if they are not effectively removed by prefiltration, or air flotation, they  could very
well accumulate to toxic or inhibitory concentrations in the mixed-liquor.  As well as
being toxic or inhibitory to microorganisms,  especially nitrifiers, oxygen transfer problems
will be encountered.  High inert solids concentrations may also cause overloading
problems in the  final  clarifier.

Ed Sebesta, Brown & Root,  Inc.;  The data indicates that nitrification occurred during
some phases of the experiments while nitrification did not occur during other phases.  Do
you have any comments about why this occurred?

Grieves:  If you have ever operated a refinery wastewater treatment  facility,  you will
know that frequently there are excursions with nitrification.  We achieved good nitrifi-
cation during the warm operating periods, phases I and II.  During phase III operation, it
was relatively cool — we recorded a mixed-liquor temperature  of 2°C on one occasion,
quite a severe winter for this part of Texas — and, as expected, nitrification  in the
control activated sludge plant was poor. This is reflected in  the probability plot (Figure
6) of the data.  However,  in the activated sludge pilot plants to which carbon was added,
almost complete nitrification was observed.  This was unexpected because, as you know,
carbon does not normally adsorb ammonia.

Bob Smith, Carborundum Co.;  Have you compared the cost effectiveness of the high
capacity Amoco carbon vs. the lower capacity carbons?

Grieves:   No, we have not made this comparison yet.  We have not  decided whether to
go commercial with our carbon or not.  However, if  and when we do decide to
commercialize our product, you can rest assured that it will be  cost effective with other
commercially available carbons.
                                        353

-------
BIOGRAPHIES
Colin G. Grieves is a Research Engineer in the
Water Conservation group at Amoco Oil Company's
Research and Development department in Naperville
Illinois. He has M.S. and Ph.D. degrees in
Environmental Systems Engineering from Clemson
University, Clemson, South Carolina, and a B.Sc.
degree in Civil Engineering from the University
of Newcastle Upon Tyne,  England.  Previous
employment was in the Public Health Engineering
Division of Babtie Shaw & Morton, Consulting
Civil and Structural Engineering in Glasgow,
Scotland.  Colin is the author of several papers
in the wastewater treatment field.
Michael K. Stenstrom is a Research Engineer in
the Water Conservation group at Amoco Oil
Company's Research and Development department in
Naperville, Illinois.  He has a B.S.  degree in
Electrical Engineering and M.S. and Ph.D. degrees
in Environmental Systems Engineering from Clemson
University, Clemson, South Carolina.   Mike is the
author of several papers dealing with various
aspects of municipal and industrial wastewater
treatment.
                                    354

-------
BIOGRAPHIES

 Joe D. Walk is Active Carbon Project  Director in
 the Corporate Development Department  of  Standard
 Oijl (Indiana), Chicago, Illinois.   He has a B.S.
 degree in Chemical Engineering  from the  University
 of Texas.  Previously he served as  Process
 Coordinator in air/water conservation, crude
 running and product treating for Amoco Oil's ten
 refineries.  Prior assignments  include Manager
 of Technical department at  Texas City refinery,
 as well as positions in New York City, New Orleans,
 and El Dorado, Arkansas, during his 31 years
 service with Standard Oil.
 James F. Grutsch  is Coordinator-Environmental
 Projects, Standard Oil Company (Indiana).   He
 holds undergraduate and  graduate degrees in
 chemistry from Indiana University.   Prior  to his
 present assignment with  Standard,  Jim served
 successfully 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 6 years at Indiana.
                                     355

-------
                          TABLE 1  OPERATING CONDITIONS

                             Pilot Plant Conditions
Aeration Zone Volume
Settling Zone Volume
Nominal Flow Rate
Nominal Hydraulic Retention Time
Nominal Settling Time
Air Flow Rate
PH
Caustic Addition Rate
Phosphorous Added to Feed
Temperature
36.7
5.7
2.45
15.0
2.33
300
6-8.5
0.12-0.30
3
liters
liters
liters/hr
hours
hours
liters/hr

liters/hr
mg/ liter
         Ambient (4-31°C)
Frequency
Daily
3 Times
a Week
   Analytical Work

      Analysis Performed^'» 28
Once a
Week
Influent and mixed-liquor pH, temperature,
influent flow rate, caustic addition rate.
Carbon addition and sludge wastage.

Influent and effluent total and volatile sus-
pended solids, soluble organic carbon, soluble
chemical oxygen demand, soluble ammonia nitro-
gen, and soluble phenolics.  Mixed-liquor
suspended solids and mixed-liquor volatile
suspended solids.  Sludge volume index.

Material balances to calculate quantity of
sludge to be wasted to maintain desired sludge
age.
                                    356

-------
                                     TABLE  2   PROPERTIES  OF  POWDERED ACTIVATED CARBONS
CO
Property
Carbon Designation
Experimental Amoco
High Surface Area -

Surface Area
BET, m2/g
Pore Volume, cc/g
> 15 A° Radius
< 15 A° Radius
Iodine Number
Methylene Blue Adsorption,
Phenol Number
Bulk Density, g/cc
Screen Analysis
Passes 100 Mesh, Wt.%
Passes 200 Mesh, Wt.%
Passes 325 Mesh, Wt.%
Al
Grade PX-21
3099
0.16
1.45
3349
mg/g 586
12.8
0.298
98.4
92.7
84.1
A2
Grade PX-23
3148
0.43
1.60
3375
550
12.6
0.228
99.1
93.4
80.8
Commercially Available Conventional Surface
Area Carbons
B
717
0.28
0.51
1790
100
34.1
0.610
99.2
86.7
60.6
C
514
0.38
0.11-0.42
920
83
22.9
0.576
100.0
94.4
68.3
D
532
0.03
0.25
888
50
23.8
0.484
100.0
97.9
91.8
      Molasses Number
10
205
103
85

-------
                                   TABLE 3

PHASES I AND II - EFFECT OF CARBON TYPE AND ADDITION RATE ON EFFLUENT QUALITY*
        50% Probability Data During 30 Days of Steady-State Operation
                            Sludge Age = 20 Days


            	Concentration, mg/liter	
            Filtered   	Pilot Plant Effluent	
Component   Influent   No Carbon   Carbon Al   Carbon B   Carbon C   Carbon D

                Phase I:  Carbon Addition Rate = 100 mg/liter
        Equil. Mixed-Liquor Temp •» 31°C. Carbon Cone = 3200 mg/liter
soc
SCOD
NHs-N
Phenolics
Equil
SOC
COD
NH3-N
Phenolics
72.0
230
25.8
4.35
22.0
73
0.5
0.018
Phase II: Carbon
. Mixed-Liquor Temp
70.0
230
25.4
4.06
26.5
58
0.2
0.020
12.5
28.5
0.2
0.003
Addition Rate =
= 25°C, Carbon
9
17
0.2
0.001
17.5
48
0.5
0.010
18.5
44
0.5
0.010
23.0
65
0.5
0.017
200 mg/liter
Cone = 6400 mg/ liter
13.5
24
0.2
0.001
15.5
28
0.1
0.003




* BATEA effluent standards in mg/liter are:  Soluble Organic Carbon  (SOC) 15
                                             Soluble COD  (SCOD)           24
                                             Ammonia Nitrogen  (NH3-N)     6.3
                                             Phenolics                    0.02
                                     358

-------
                                   TABLE 4
     PHASE III  -  EFFECT OF CARBON TYPE AND ADDITION RATE, SLUDGE AGE,
                AND  INFLUENT PRETREATMENT ON EFFLUENT QUALITY
        50% Probability Data During 30 Days of Steady-State Operation
        Equil. Mixed-Liquor Temp = 14°C, Carbon Cone. = 2400 mg/liter
                         Carbon
  Influent
Pretreatment

Filtered  Feed
         Addition
      Rate, mg/liter
                Influent Concentration, mg/liter

                SOC     COD    NH3-N   Phenolics
Unfiltered

Filtered



Filtered

Filtered

Filtered

Filtered

Filtered
B

Al

Al

A2
                73.5   294.5   19.3      3.95

                Effluent Concentration, mg/liter

Sludge Age = 20 Days

                32.0   103.5   12.1      0.027

                29.0   83.0    14.5      0.027

Sludge Age = 60 Days

                25.0   65.9

  100           16.0   40.3

  50            12.0   27.5

  25            16.0   50.3

  50            13.0   31.0
5.1
0.2
0.1
0.4
1.8
0.019
0.001
0.002
0.006
0.004
                                    359

-------
                                                    TABLE 5

                        PHASE IV - EFFECT OF HIGH SLUDGE AGE, LOW CARBON ADDITION RATE,
                          AND DECREASED HYDRAULIC RETENTION TIME ON EFFLUENT QUALITY*
                         50% Probability Data During 30 Days of Steady-State Operation
                                        .Equil. Mixed-Liquor Temp = 27°C

Type
B
Al
Al
Al
Carbon
Addition
Rate, mg/liter
-
25
25
25
10
Sludge
Age, days
60
60
60
60
150
Hydraulic
Retention
Time, hr
15
15
15
7.5
15
Equil. Mixed
Liquor Carbon
Cone, mg/liter
-
2400
2400
4800
2400
Effluent Cone,
mg/liter
SOC
29
22
18
17
16
COD
99
64
52
46
49
NH3-N
0.1
0.1
0.1
0.3
0.1
Phenolics
0.018
0.010
0.010
0.010
0.010
* Filtered influent contained 78 mg/liter SOC, 270 mg/liter COD, 29 mg/liter NH3-N, and 3.25 mg/liter
  phenolics.

-------
                              TABLE 6

PHASE III - BAT GUIDELINE AND ACTUAL VARIABILITY FACTORS FOR PILOT
                PLANT FED 25 mg/liter OF CARBON Al
BAT Guideline

Parameter

Soluble Organic Carbon
Soluble COD
NH3-N
Phenolics
Variability

Daily Max.
1.6
2.0
2.0
2.4
Factor

30 Day Max.
1.3
1.6
1.5
1.7
Actual

Variability Factor

Daily Max. 30
2.8
7.5
2.1
5

Day Max.
2.8
7.5
2.0
5

-------
    FIGURE 1.  SIMPLIFIED REFINERY  BPT WASTEWATER
                TREATMENT  SYSTEM
                  Chemicals
Refinery
Waste-~~"
water
 Gravity
Separator
     Slop Oil
     to
     Treatment
Dissolved
Air
Flotation
Equalizat-
ion
Granular
Media
Filter
 Final
-•••Effluent
                                         Recycle
      Sludge
      to
      Treatment

-------
       FIGURE 2.   SCHEMATIC OF ACTIVATED  SLUDGE  REACTOR
                    USED IN PILOT  PROGRAM
CO
                                                   Air
                            .Baffle
Air— »•«==


36.7 Liters
Aeration
Zone

=





/
\/
\&
* \
.v y|
Side
Feed -*•<=

i—^FfflMpnf-
— 5.7 Liters
Settline
Zone
	 Sludge
Blanket

mSm



<

1



<
Enc




i
I
r
r


\

a^ 	 Chemicals



Sampling
SS!""~Port
\ Dif fuser
Stone
Chemicals.
                 {r?r
          Feed.
                      o
                      o
                      Top
                         Plexiglass
                        ./Construction
                                  Effluent
                              >A.ccess
                               Hole

-------

c
o
I
o
o
O
O
I/)
0)
LU
             FIGURE 3. EFFECT OF MIXED-LIQUOR CARBON
                        CONCENTRATION ON  EFFLUENT SOC
                               Legend
                                20 days sludge age
                                60
                               150
carbon
dose
mg/l   100
                            treatment objective
                                -150 days
            1      2     3     A

           Mixed-liquor carbon concentration
                                    8

-------
     o
     o
            FIGURE  4  - SOLUBLE  ORGANIC CARBON -  PHASE  3
CONTROL, NO CARBON,   60 DAY SRT-X
AMOCO PX-21, 25 M8/1, 60 DAY S8T- +
AMOCO PX-21, 50 U8/L, 60 DAY SRT-A
CARBON B, 100 H8/L.   60 DAY SRT-O
LU
X
o
2
U
 •
o
   0.013
                    0.1     0.2   0.3  0.4  0.5  0.6  0.7   0.8
                          CUMULATIVE  PROBABILITY
                                                          0.9
                                                                    0.987
                                   365

-------
             FIGURE  5
      O
      O
   SOLUBLE  CHEMICAL  OXYGEN  DEMAND
             PHASE 3
CONTROL, NO CARBON,   60 DAY SRT-X
AMOCO PX-21, 25 U6/L, 60 DAY SRT- +
AMOCO PX-21, 50 UO/L, 60 DAY SRT-A
CARBON B, 100 HO/I,   60 DAY SRT-CP
CC
UJ
CD



 •

O
 •
O
 •
O
 •

CO
  0.023
                         0.2    0.3   0.4  0.5  0.6   0.7   0.8
                          CUMULATIVE PROBABILITY
                                                            0.9
                                           0.977
                                      366

-------
                FIGURE 6  -  AMMONIA-NITROGEN -  PHASE  3
                         CONTROL, HO CARBON,   60 DAY SRT-X
                         AMOCO PX-21, 25 IW/U, 60 DAY SRT- +
                         AMOCO PX-21, SO U6/L, 60 DAY SRT-A
                         CARSON B, 100 MS/I.   60 DAY SRT-O
CC.
LU
C!)
2
 I
ro
   0.013
                     O.t     0.2   0.3  0.4  O.J  0.6  0.7   0.6    0.9
                          CUMULATIVE PROBABILITY
                                                                     0.987
                                   367

-------
         FIGURE  7   -  PHENOLICS  -   PHASE   3
     10
     ro
CO
O
O
z
LU
•3Z
Q_
     O

  0.036
 CONTROL, NO CARBON,  60 DAY SRT-X
 AMOCO PX-21, 25 U6/L, 60 DAY 8RT- +
 AUOCO PX-21, SO M8/L, 60 BAY SRT-A
 CARBON B, 100 M8/L,  60 DAY 3RT - O
0.2   O.S  0.4  O.S  0.6  0.7   0.8
                      CUMULATIVE  PROBABILITY
0.9
                                      0.964
                              368

-------
                  "TREATMENT OF OIL REFINERY WASTEWATERS
                      WITH POWDERED ACTIVATED CARBON"


                             Paschal B. DeJohn
      Manager, Purification Sales, ICI United States Inc.  (Delaware)

                              James P. Black
         Industry Coordinator, ICI United States Inc. (Delaware)


INTRODUCTION

      The effectiveness of powdered carbon as an additive to improve activated
sludge treatment  has been demonstrated in a variety of industrial  and munici-
pal plants.   This type of treatment has gained wide acceptance in  the past
few years and is  currently an essential part of treatment at 60-80 plants.
These plants  range in size from 10,000 gpd package units located along the
Alaska Pipeline to the very sophisticated 40,000,000 gpd PACT treatment plant
at the DuPont Chambers Works in Deepwater, New Jersey.  At least four petrol-
eum refiners  currently use powdered carbon as an integral part of  their waste
treatment scheme.

HOW THE PROCESS WORKS

      The reason  powdered carbon has gained such acceptance treating a wide
variety of waste  streams is the extreme flexibility which can be employed in
its usage.

      • The  amount of carbon used can be varied to meet the
        treatment requirements as they change.

      • Higher COD or BOD removal than is usually obtainable
        by conventional biological treatment can be achieved.

      • The  combination of activated carbon in a biological
        system provides more effective treatment than either
        of the processes would if used singularly.

      Carbon  aids the biological process two ways:

      1.  By  direct adsorption of pollutants.

      2.  By  providing a more favorable environment for the micro-
         organisms to propogate.

      Adsorption  is an equilibrium phenomenon.  In general, carbon preferen-
tially absorbs higher molecular weight compounds.  Given a related series of
organic compounds;  for example, alcohols, one finds that the lower molecular
weight alcohols (methanol, ethanol) are not appreciably absorbed by carbon
while the higher  molecular weight alcohols are.  Fortunately, compounds which
are poorly absorbed (weakly held by the carbon) are usually compounds which
are the most  amenable to biological treatment.

                                    369

-------
      We can generally classify organic compounds into  three broad  categories
with respect to their adsorptability onto carbon.

      1.  Compounds which are readily adsorbed.  These  compounds are
          usually "tightly held" by the carbon.  And consequently,
          they are not readily desorbed.

      2.  Compounds which are adsorbed with difficulty.  These
          compounds are desorbed easily.

      3.  Compounds which are poorly adsorbed.

      Organic compounds can also be classified in terms of their susceptibili-
ty to biodegradation:

      1.  Compounds which are readily and rapidly biodegraded.

      2.  Compounds that are degraded slowly.

      3.  Compounds that are not biodegraded.  Many of  these can
          function as toxicants in a biological system.

      It is important for one to understand the interaction of carbon and the
microorganisms present in an activated sludge system.   Exhibit 1 does this by
considering how a carbon-biological system handles each of the above classi-
fications of organics compounds.  Relative adsorptivity and biodegradability
for  organic compounds was taken from an EPA source (Reference 1).

      The boxes in Exhibit 1 have been numbered from 1  through 9 and are
interpreted as follows:

                                    Degree of
          Box                    Biodegradability          Adsorptability

           1                          Rapid                    Strong
           2                          Rapid                   Moderate
           3                          Rapid                     Weak
           4                          Slow                     Strong
           5                          Slow                    Moderate
           6                          Slow                      Weak
           7                          None                     Strong
           8                          None                    Moderate
           9                          None                      Weak

The  classes of compounds which are represented by boxes 1, 2, and 3 would be
handled quite easily by the microorganisms in a carbon-biological system.
Those compounds which are represented by boxes 1, 4, and 7 would be removed
by direct adsorption on the carbon.  Compounds which fall in box 1  (both
rapidly biodegradable and strongly adsorbed) are few in number.  The only
example that we could find is o-cresol.
      Boxes 4, 5, and 6 represent compounds which are slowly biodegradable.
These compounds probably would not be removed very effectively in a conven-
tional activated sludge system.  In a carton-biological system, compounds in

                                     370

-------
box 4  are removed by direct adsorption and  are  held very tightly by the
carbon.   Compounds in boxes 5 and 6 are  retained in the system by moderate or
weak adsorption.   Because carbon is adsorbing these compounds, their concen-
tration in the liquid stream is reduced.  The microorganisms  are able to
degrade the organics in reduced concentrations,  and as they do the equilli-
brium  between the carbon and these organics in  the waste is disturbed.
Because these compounds are not held very tightly by the carbon, they are
readily desorbed back into the system and a new  equilibrium is established.
In this fashion,  the carbon is acting as a  storage area keeping the concen-
tration of slowly biodegradable organics at a level where they can be handled
by the microorganisms.  Compounds which  fall into the categories represented
by boxes 4, 5, and 6 are effectively handled in a carbon-biological system
because of synergistic effects.  It is primarily these compounds that are
removed more effectively in a carbon-biological system as compared to either
process operating singularly.

      Compounds represented by boxes 7,  8,  and  9 are not biodegradable.  And
in some cases, these compounds are actually toxic to microorganisms.  In our
opinion, carbon performs its most beneficial action in this area.  Compounds
which fall into the category represented by box 7 are removed by direct
adsorption and are held very tightly by  the carbon.  Compounds in box 8 are
removed by direct adsorption, and even though they are held very loosely by
the carbon, it is difficult to disturb the  equillibrium.  This is because
the concentration of these organics remaining in the waste stream are not
being degraded by the microorganisms.  The  compounds represented by box 9
are the only ones that cannot be handled very effectively in  a carbon-biolo-
gical system.  Fortunately, there are very  few  organics which are both non-
biodegradable and weakly adsorbed by carbon.

      Examples of the different compounds are shown in the various boxes in
Exhibit 1.

      One of the important effects of carbon in the system (not related dir-
ectly to adsorption) is the higher levels of biomass that can be used because
of the density and "weighting effect" of the carbon.  (Both the use of great-
er sludge mass and the temporary retention  of slowly degraded compounds by
the carbon gives more time for the compounds to be consumed biologically).

      Carbon adsorbs the pollutants and  oxygen, localizing them for bacterial
attack.  Because the aerobic action is dependent upon the concentration of
the reactants, this localizing effect serves to drive the reaction further
towards completion resulting in improved BOD removal (Reference 2).

      Many pollutants that are not biologically degraded in a conventional
activated sludge system would be if they were in contact with the biomass
for a longer period of time.  When absorbed by  the carbon, these molecules
settle into the sludge.  Contact time is thereby, extended from hours to days.
This results in lower effluent COD's and TOC's.  High density powdered carbons
Improve solids settling in the secondary clarifiers.  This results in lower
effluent suspended solids and also a reduction  in BOD.  Under high organic
load conditions which normally would lead to sludge bulking,  the dense carbon
                                      371

-------
will act as a weighting agent keeping the  sludge  in  the system.   When dis-
persed biofloc results due to low organic  loads,  carbon serves as a seed for
floe formation preventing loss of solids.  Under  these conditions, phosphorous
and nitrogen removal are generally enhanced.

      Powdered carbon improves treatment in activated  sludge process because
of its adsorptive and physical properties.  Powdered carbon can be added to
any convenient point in the activated sludge process to get it into the
aerator.  Direct addition to the aerator,  sludge  return lines, influent
channels, or through the secondary clarifier are  all possibilities.  It is
not necessary to add carbon continuously in most  cases.   A dense,  easily
wetted carbon can be added dry or in slurry form  with  water.

RESULTS

      14 refineries have evaluated HYDRODARCO powdered activated carbons in
full scale activated sludge systems during the past  three years.   The first
treats a 2.2 MGD flow with an average BOD  of 400  ppm in a 1.2 million gallon
aerator.  Mixed liquor solids are maintained at 3600 ppm (2880 ppm volatile).
Waste activated sludge is digested aerobically, centrifuged,  and hauled to
landfill.  Despite a secondary clarifier overflow rate of only 423 gallons/
ft.^ and use of 22 ppm cationic polymer for secondary  solids capture, efflaent
solids averaged in excess of 100 ppm.  Toxic loads caused periodic loss of
aerator biosolids.  Defoamer costs averaged $200/day for aerator foam control.

      HYDRODARCO C, a high density, lignite based powdered carbon, was added
to the aerator over a four and one-half month period.   Eventually, the equil-
librium carbon level reached 1800-2000 ppm.  At the  sludge solids concentra-
tion obtained and wasting rates employed,  it was  possible to maintain this
level with a daily average carbon dose of  only 20 ppm.

      Over the entire carbon test period,  average BOD  reduction equaled 82%
versus 23% during the post test control period  (Figure 1).  As carbon built
up in the system, BOD removals reached the 90-95% range, and the plant was
able to meet their 30 ppm BOD effluent standard.  Effluent COD was reduced
from an average of 1180 ppm without carbon to 350 ppm  with (Figure 2).  Aver-
age effluent TOC decreased from 420 ppm to 100 ppm  (Figure 3), and total
carbon decreased from 520 ppm to 180 ppm  (Figure  4 ).   The lower slope of the
carbon plots also indicate the decreased variability in effluent quality with
carbon present.

      HYDRODARCO C had a dramatic effect on the reduction of oil through the
system (Figure  5 ) •  The effluent concentration was  reduced by 75% (average),
and the range was narrowed as well.

      Both removal of the oil by the powdered carbon and the weighting effect
of carbon resulted in lower effluent solids  (Figure  6).  Prior to carbon
treatment, the plant used polymer at a dosage of  20  ppm, but still experienced
poor solids settling.  When carbon was added to the  system, solids settling
improved, and the polymer dosage was cut  in half. Since effective solids
settling could not be achieved with the use of  carbon  or polymer alone, it
appears that the combination of the two was required to attain the desired
results.  This, of course, represents an  operating cost savings for the plant.

                                     372

-------
Improved  solids  settling increased sludge  thickening which allowed a 65%
reduction in  sludge wasting.  Again, savings  on  the operation of the centri-
fuges,  including power and labor, occurred.

     Use of  HYDRODARCO C eliminated the need for  aerator  defoamer.  Removing
the foaming agents from the wastewater by  adsorbing them eliminated foam
problems  in the  receiving stream.  Defoamers  only  suppress foam in the aera-
tor and do not prevent its reappearance in the effluent.   Carbon can reduce
operating costs  by allowing surface aerators  to  aerate and mix the activated
sludge  rather than expand energy generating foam.

     Both nitrogen (Figure 7 ) and phosphorous  (Figure 8  ) removals were
improved  with powdered carbon.  Reason:  carbon  adsorbs compounds toxic to
nitrifiers and allows them to operate at normal  levels.  In this waste,
neither nitrogen nor phosphorous were limiting for bacteria growth.  Increased
nitrogen  removals are attributed to the fact  that  the  dense carbon settled
the nitrifying organisms which normally would float out of the system.  The
result  is a longer solids retention time which is  more favorable for nitrifi-
cation  to occur.  By the same token, improved solids settling is probably
the reason for decreased phosphorous levels.   We suspect that the phosphorous
is precipitated  with the carbon-biosolids  floe and removed in the sludge
rather  than degraded biologically.

     While the  exact reason for the bug kills prior to carbon was not known,
upsets  were greatly reduced with carbon in the aerator.  A possible explana-
tion is the effect carbon had on the removals of heavy metals such as zinc
(Figure  9).   The ability of activated carbon to adsorb heavy metals from
wastewater has been established elsewhere  (References  3, 4, and 5).

      The second evaluation was conducted  at  a 12  MGD  plant treating an
average 12 MGD flow.  The TOG of the raw waste ranged  from 100-1000 ppm,
averaging about  200 ppm.  Major treatment  problems included aerator foaming
caused  by alkanolamines in the waste; high effluent TOG; oily, difficult-to-
handle sludge; and high effluent solids.

      Effluent TOC's were maintained below 20 ppm  during shock load periods.
This was well within the standard of 50 ppm.   In a post test control phase,
a deterioration in effluent quality was observed as carbon was lost from the
system through sludge wasting.

      A third evaluation was conducted at  a 2.5  MGD plant  treating a 550 ppm
COD refinery waste in a two stage, conventional  activated  sludge system.
Carbon was added to the second stage aerator  over  a six week period.  A con-
stant daily carbon dose was maintained for each  week and increased in
succeeding weeks.  No sludge was wasted intentionally  during this time.

      Optimum treatment was found at relatively  high influent dose of 200 ppm.
Effluent  solids  and COD removals increased 40%,  and BOD removals, already
high, increased  10%.
                                     373

-------
      The major finding of this study was  the  increased removals of cyanide
with carbon in the aerator.  In conjunction with  CuSCfy  treatment, the average
cyanide levels decreased from 1 ppm before carbon to  0.05 ppm.   The precise
nature of this removal is not known and bears  further investigation.

      Results from a fourth study are summarized  in Exhibit R for three
separate carbon addition periods.  This plant  is  a current user of activated
carbon to treat 2.2 MGD flow.  They have reported the following results
from carbon addition  (Reference  6):

                56% reduction of suspended solids.
                36% reduction of COD.
                76% reduction of BOD.
                Foam problem elimination.

      This improved plant performance is achieved at  a  carbon cost of
1.7-4.3C/1000 gallons treated.

      A fifth study was made at a refinery which  had  a  6 MGD (8 MGD design)
activated sludge plant.  Carbon dosage reached approximately 500 ppm in the
aerator before the study was terminated due to loss of  biosolids.  The loss
of biomass resulted from an inadvertent increase  in sludge wasting.   This
plant continued to use the same volumetric wasting rate when carbon was added
to the system and because of the sludge density increase due to carbon addi-
tion, the MLVSS dropped sharply from 2900  ppm  before  carbon to  1500 ppm at
the conclusion of the study.  After carbon addition was stopped,  the MLVSS
returned to 2500 ppm with no change in the wasting rate.

      Although considerable improvement in BOD removal  was achieved at this
plant (Exhibit S), the full benefit of carbon  was not realized  because of
high influent oil concentrations (100+ ppm).

SUMMARY & CONCLUSIONS

      •  In summary, it has been shown that refinery  wastes can be success-
fully treated with powdered carbon in activated sludge.

      •  Powdered carbon can improve organic removals,  aid solids settling
and sludge handling, and provide protection from  toxic  or shock loadings.
In the face of widely varying influent organic or hydraulic loads, carbon
levels effluent quality.

      •  High density carbons are preferred to minimize carryover from secon-
dary clarifiers and to increase sludge compaction.  Such carbons also require
less makeup to maintain the desired aerator equillibrium level since carbon
is lost from the system only during sludge wasting.

      •  Normally, one would think that the use of carbon would increase
costs.  However, savings on defearners, coagulants, powder and labor can often
decrease operating expenses.
                                     374

-------
BIOGRAPHY

       Mr. Paschal DeJohn is Manager of
Purification Sales and Project Leader in Activated
Carbon at ICI United States Inc.; holding this
position since 1972.  Mr. DeJohn holds a B.S.
degree in chemistry from Westchester State College
and an M.B.A. degree from Widener College,
Pennsylvania.  Previous to his present position Mr.
DeJohn was Water & Wasrewater Treatment District
Engineer with Drew Chemical, Parsippany,  New
Jersey.  He is 1977 Technical  Conference Chairman
for WWEMI and the 1978 General Conference
Chairman for this ame organization.  Mr. DeJohn
has been involved with water and wastewater treat-
ment for the past twelve years.
       James P. Black is Industry Coordinator for
Water Purification at ICI United States Inc. He
 has a B.S. degree in chemical engineering from
 the University of Texas at Austin.  Prior employ-
 ment included ICI's Research & Development
 Laboratory, and Corporate  Planning Staff.
                                      375

-------
                           DEJOHN PAPER DISCUSSION
Ed Sebesta, Brown & Root;  Have you observed  any situations where you
mentioned sot e compounds are loosely adsorbed or difficultly adsorbed and
then quite easy to be desorbed?  Have you ever seen any situations or heard
of situations where because of changing influent situations you may suddenly
desorb an accumulation of adsorbed materials  and effect the system in that
way?

DeJohnt  Potentially that can occur.  I would think something like that would
be more prone to happen in a granular carbon  system.  You can design a
granular carbon system around this however.   In the PACT systems we've been
involved in, I'm not aware of any that have desorbed back an accumulation of
adsorbed material.  But there's always the possibility  that this can happen.
                                REFERENCES
 1.   EPA Contract  68-01-2926, April  1,  1975.

 2.   Adams,  A.D.,  "Improving Activated  Sludge Treatment with Powdered Activa-
     ted Carbon -  Textiles" presented at  the  6th Mid-Atlantic Industrial
     Waste Conference,  University of Delaware,  November'15, 1975.

 3.   Esmond, S.  E.  and  Petrasek,  A.  C., "Removal of Heavy Metals by Waste-
     water Treatment  Plants," presented at the WWEMA Industrial Water and
     Pollution Conference, Chicago,  Illinois, March 14-16, 1973.

 4.   Sigworth, E.  A.  and Smith,  S. B.,  "Adsorption of Inorganic Compounds by
     Activated Carbon," Journal  AWWA, Water Technology/Quality. June, 1972
     (p. 306).

 5.   Linstedt, K.  D.  et. al,  "Trace  Element Removals in Advanced Wastewater
     Treatment Processes," Journal WPCF,  43,  No. 7, 1507 (July, 1971).

 6.   Rizzo,  Joyce  A., "Case History: Use of  Powdered Activated Carbon in an
     Activated Sludge System", presented  at the Open Forum on Management of
     Petroleum Refinery Wastewaters, Tulsa, Oklahoma, 1976.
                                     376

-------
                                 EXHIBIT 1
                            CARBON ADSORPTION
>;

>




5
0
0

0
5
   Slow
 None
(or toxic)
              Strong
                                    Moderate
Weak
                                                       TRIPHENYL

                                                       PHOSPHATE
                               377

-------
                                 EXHIBIT R


                                      No Carbon        First Carbon Period

COD                                    .„                    ...-,
  Influent                             «J ppm                457 ppm
  Effluent                             17° PPm                135 ppm
  % Removed                              63                     70

BOD
  Influent                             152 PPm                2]3 ppm
  Effluent                              15 ppm                 15 ppm
  % Removed                              9°                     y3

SUSPENDED SOLIDS
  Effluent                              115                     50
 COD
   Influent                             343 ppm                444 ppm
   Effluent                             266 ppm                183 ppm
   % Removed                              23                     59

 BOD
   Influent                             152 ppm                227 ppm
   Effluent                              30 ppm                 14 ppm
   % Removed                              80                     94

 SUSPENDED SOLIDS
   Effluent                              162                     72
 COD
   Influent                             367  ppm                 379 ppm
   Effluent                             166  ppm                 112 ppm
   % Removed                               55                     70

 BOD
   Influent                             188  ppm                 207 ppm
   Effluent                              12  ppm                   3 ppm
   % Removed                               94                     99

 SUSPENDED SOLIDS
   Effluent                               79                     42
                                      378

-------
                           EXHIBIT S


Flow:   8 MGD design, 6 MGD actual.

Carbon Dose:  500 ppm in aerator.

MLVSS:  2900 ppm before carbon
        1500 ppm during carbon

(A 50% loss of aerator solids resulting from continuing  same
 volumetric wasting rate of the more dense carbon sludge.)

        2500 ppm after carbon

•  BOD removal - 55% before carbon addition
                 70-80% during carbon addition
                 60% after carbon addition

t  Influent oil concentrations were so high (100+ ppm)  during
   test that effects of carbon were overshadowed.
                                379

-------
§  % BOD
   REMOVED
100

 90

 80

 70

 60

 50

 40

 30

 20

 10
                         EFFECT OF POWDERED CARBON  ON
                                 BOD REMOVAL
                     HYDRODARCO C
    FIG. 1
                          5  10     30   50   70     90 95

                            % OF VALUES LESS THAN
                                              99

-------
co
oo
          1000

           900

           800

           700

EFFLUENT  600
COD, PPM
           500

           400

           300

           200

           100
                         EFFECT OF POWDERED CARBON ON
                                 EFFLUENT COD
                                              HYDRODARCO C
                          I	I
                               III  1111
 I   1
                          5  10
                                 30  50   70
90 95
99
                               OF VALUES LESS THAN

-------
do
          1000

           900

           800

           700

EiFFLUENT  600
TOC, PPM
           500

           400

           300

           200

           100
                         EFFECT OF POWDERED CARBON ON
                        EFFLUENT TOTAL ORGANIC CARBON
                                                   I
                                             BLANK
                                                HYDRODARCO  C
                          I	I
                               I   I	I  »  I   I  I
 I
I
                          5  10
                                 30  50   70
90 95
     99
   FIG. 3
                             % OF VALUES LESS THAN

-------
•
             1000

             900

             800

             700

EFFLUENT TC, 60°
             500

             400

             300

             200

             100
                            EFFECT OF POWDERED  CARBON  ON
                                EFFLUENT TOTAL CARBON
                                              HYDRODARCO C
                             5 10
                                    30  50   70
90 95
99
                                % OF VALUES LESS THAN

-------
                100
                 90
                 80
                 70
                 60
  EFFLUENT OIL,
*• rtr-kn/i            50
PPM
               40
               30
               20
               10
                0
                          EFFECT OF POWDERED CARBON ON
                                   EFFLUENT OIL
                                              BLANK
                             II   I   I  I  I  I  I
 HYDRODARCO C
»    I   I     I
                           5  10     30  50   70     90 95
                              % OF VALUES LESS THAN
                                                              99

-------
00
in
EFFLUENT
SUSPENDED
SOLIDS, PPM  500
                          EFFECT OF POWDERED CARBON ON
                             EFFLUENT SUSPENDED SOLIDS
                                                HYDRODARCO C
                           5  10
                                       50   70
90 95
99
   FIG. 6
                            % OF VALUES LESS THAN

-------
u>
oo
           zo
           1.8
           1.6
           1.4
EFFLUENT  1.2
NITROGEN,
PPM        1-0
           0.8
           0.6
           0.4
           0.2
                         EFFECT OF POWDERED CARBON ON
                               EFFLUENT NITROGEN
                                  BLANK
                          •P
                                   I   I  I  I  I   I
                                                HYDRODARCO C
                                                I   I
  FIG. 7
                          5  10    30   50  70     90 95
                            % OF VALUES LESS THAN
                                                        99

-------
oo
-j
              2.0
              1.8
              1.6
              1.4
EFFLUENT     1.2
PHOSPHOROUS,
PPM           1.0
              0.8
              0.6
              0.4
              0.2
                            EFFECT OF POWDERED CARBON ON
                                 EFFLUENT PHOSPHOROUS
                              I	I
BLANK |
      I
                                                     HYDRODARCO C
                                  \   \   \  I	I  I	I
         I	I
                             5  10
                                    30   50   70
        90 95
99
   FIG. 8
                                % OF VALUES  LESS THAN

-------
00
oo
          1000

           900

           800

           700

EFFLUENT  600
ZINC, PPM
           500

           400

           300

           200

           100
                         EFFECT OF POWDERED CARBON ON
                                  EFFLUENT ZINC
                                             HYDRODARCO C
                          II    »   I  I  I  I   I   I
                                                 I	I
                          5  10
                                 30   50   70
90 95
99
   FIG. 9
                             % OF VALUES LESS THAN

-------
                               CASE HISTORY
                   THE USE OF POWDERED ACTIVATED CARBON
                     WITH A BIODISC-FILTRATION PROCESS
                     FOR TREATMENT OF REFINERY WASTES

                               J.F. Dehnert
            Environmental Director, Avon Refinery, Lion Oil Co.
ABSTRACT
     A description of the development of a supplemental petroleum waste
water treating  plant utilizing a Rotating Biological Surface Unit and
Powdered Activated Carbon followed by clarification and filtration
from laboratory and pilot plant studies through construction, start up
and operation to meet July 1977 NPDES discharge requirements.

     Starting in 1972 Pilot Plant studies were conducted to compare the
performance of  activated sludge, trickling filter, RBS and activated
carbon absorption processes in treating the Avon Refinery Waste Water.
The primary objection was to meet the EPA guideline discharge limits
plus the California State limits on fish toxicity.  After several months
of study the treatment scheme of a RBS Unit plus solids removal facili-
ties was selected to meet the Federal standards and powdered activated
carbon was  selected to meet the toxicity limits.

INTRODUCTION

     With the adoption of the 1972 Amendments to the Clean Water Act,
the Staff at the Avon Refinery near San Francisco, then operated by
Phillips Petroleum Co. embarked on an investigative program to determine
the waste water treatment necessary to meet the limitations which would
eventually  be placed on the refinery discharge through the NPDES Program.
At that time and until January 1975, the refinery discharge was already
subject to  limitations imposed by the California Regional Water Quality
Control Board on 5-day BOD, oil and grease, settleable solids, suspended
solids,  coliform and fish toxicity.  In addition, limitations were in
effect on receiving water quality with respect to pH, dissolved oxygen,
undissociated NH/jOH, chromium, lead, H2S, Fish Toxicity, floating oil,
discoloration or turbidity and odor.  The existing waste water treatment
included sour water stripping, API gravity separation, dissolved air
flotation and pH equalizing surge ponds followed by a 108-acre bio-
oxidation pond.   The company had also segregated the refinery sewers so
that as much as possible of uncontaminated storm run off could bypass
the process water treatment.  This storm run off was combined with the
bio-oxidation pond effluent and discharged in an underwater diffuser in

                                    389

-------
the main channel of an arm of San Francisco Bay which receives the
Central California Valley drainage.  This treating process is illustrated
in Figure 1.

INVESTIGATION

    At the start of the investigation, no specific Federal limitations
had been determined and, therefore, the studies were mainly concerned
with a comparison of generally accepted methods of waste water treatment
to determine which of these would be most effective in removing or
reducing the known pollutants in the Refinery effluent.  A primary
objective was to determine a treatment scheme that would result in a
waste water discharge that would meet the more restrictive California
State fish toxicity limitations that were being proposed.

    In early 1972, working with an engineering contractor, three pilot
plants were installed at the Refinery with a slip stream of the waste
water going to the bio-oxidation pond serving as the raw feed.  These
units were an air agitated activated sludge unit, a 21-foot trickle
filter and a four-column granular activated carbon unit preceded by a
mixed media filter.  These plants were operated from June 1972 through
1973 in parallel or in series under wide variations of operating
conditions such as hydraulic loading, recycle rates and suspended solids
concentrations.

    During this period of operation, different sets of tentative EPA
guidelines were issued for the Petroleum Industry.  In each case, the
effluent from the biological treating pilot plants failed to meet these
guidelines and the proposed California fish toxicity standards.  From
the experimental data it appeared the only way these limitations could
be met was by the use of activated carbon at high regeneration rates as
a final treating step.  It appeared that the waste water contained some
non-biodegradable or at least "refractory" organic material as indicated
by COD and TOC tests.

    In the Spring of 1973, information was received describing the
rotating biological surface units which were being proposed for treating
industrial waste as an improved alternate to the other biological treat-
ment systems.  There were several advantages claimed for this process
such as high biomass concentration; low volume, high density sludge
production and low power requirement.

    Subsequently, arrangements were made to install a four stage pilot
rotating biological surface unit at the refinery and to compare its
performance with the other pilot plants.

    The RBS test program consisted basically of three periods determined
partly by a difference in the quality of raw waste water and partly by  the
type of operation of the unit.  During the first period, a direct comparison
was made between the trickle filter, the activated sludge and  the RBS units
with the same feed going to all three units.  For the second test period,
                                    390

-------
the feed rate to the RBS unit was reduced  to a very  low  figure to
establish nitrifying bacteria in the bio-mass.   During the third period
a series of hydraulic loading tests were performed where the rate was  '
varied from 1/2 gpm to 18 gpm representing hydraulic loading of 0.2 to
6.8 gal/day/sq.ft. of surface area.

     As a result of the pilot plant study, it was concluded that the
removal of organic pollutants by the RBS unit compared favorably with
the trickling filter and the activated  sludge processes  and, for some
waste water parameters, the RBS unit appeared to be  superior.  The pilot
plant operation verified most of the claims made by  the  manufacturer,
particularly with respect to energy requirements and ease of operation.

     The removal of organics by the RBS unit was very similar to the
trickle filter and the activated sludge with each unit able to achieve
about the same percentage removal and final concentrations in the final
effluent at their optimum operation.

     A portion of the test program was  devoted  to establishing nitri-
fication and determining the relationship  between hydraulic loading and
the degree of conversion of ammonia to  nitrate.  This was accomplished
by operating the unit at a very low feed rate and adding sodium
bicarbonate to increase the alkalinity.  At the  low  rate, it was possible
to lower the ammonia concentration from 15 to 20 mg/1 to less than
1 mg/1; however, as the feed rate was increased, nitrification decreased
and eventually stopped altogether.  Contrary to  what was expected,over
50% of the conversion of ammonia to nitrate took place in the first stage.
From the data obtained, it was concluded that,  if nitrification is
desired in a commerical unit, it would  have to be designed for about
one-half the hydraulic loading that would  be required for organics
removal.

     One of the most noticeable differences between  the  RBS effluent and
the effluent from the other bio systems was the  suspended solids content.
Although the suspended solids did increase with  feed rate, even at
relatively high hydraulic loading, the  RBS effluent  had  lower suspended
solids than the best operation of the other processes.   At low feed
rates, the RBS effluent after 30 minutes of settling, exhibited a
sparkling appearance that was achieved  on  the other  processes only by
filtration or activated carbon treatment of the  effluent.

     Static bioassays were conducted weekly on  samples of the various
pilot plant effluents using the APHA standard methods to determine the
96-hr, median toxicity (TLm).  Although the RBS  unit was not the answer
to the toxicity problem at the Avon Refinery, in general, this effluent
was less toxic than the effluents from  either the trickling filter or the
activated sludge.  Activated carbon absorption  remained  as the only waste
water treatment,that would produce a completely  non-toxic water  (100%
survival) from the waste water stream.

     Activated carbon treatment data are presented  in Tables  I through
VIII.

                                    391

-------
    Table I presents early data showing the effect on fish  toxicity
of treating various RBS effluents with powdered activated carbon.

    Tables II and III present data on the effect of powdered carbon on
other parameters and indicate that toxicity is improved although other
parameters are not greatly affected.

    Tables IV and V illustrate the effect of pH changes on  toxicity and
possibly the ability of carbon to absorb the toxicants.

    Tables of VI and VII show comparisons of two different  powdered
activated carbons and indicate that selection of the proper carbon source
can make a very great difference in the ultimate  success of carbon
treatment.

    Table VIII is a part of a very large table of data obtained on a
granular carbon test conducted over a six month period.  It is presented
to further illustrate that long after the carbon was "exhausted" with
respect to COD removal it would continue to produce a non toxic effluent
and that it could be rejuvenated by a hot water backwash.

    By the time these pilot plant studies were complete, the final
guidelines had been issued by EPA and the N.P.D.E.S. permit for the
refinery had been issued by the Regional Water Quality Control Board.
This permit outlined not only the discharge limits but a compliance
schedule for submitting a conceptual plan, completion of Engineering,
start of construction and completion of construction.  At this time a
thorough review was made of all the accumulated pilot plant data and the
conceptual plan developed.  It appeared that of the many parameters of
water quality, COD, suspended solids and fish toxicity would control the
design of the treatment system.  Included in the consideration was the
volume of water which varied considerably with the seasonal storm water
entering the process or oily sewers, since practically all  of the annual
rainfall in this location occurs between the first of November and the
first of April.  Several alternate plans were considered but all were
basically a supplemental biotreatment, solids removal and activated
carbon treatment.

    During the long period of monitoring the raw waste quality it became
evident that treating requirements were also cyclic in that both COD and
toxicity increased during the winter months but during part of the year
the discharge would probably meet the 1977 limits without much, if any,
additional treatment.  In our studies with granular activated carbon we
noted that in several instances long after the carbon was "exhausted"
with respect to removal of COD it would still produce a non toxic water.
From this information it appeared the biotreating system should be capable
of handling wide variation in waste loading and that a carbon system should
be designed to be used only when necessary.  From Capital cost considera-
tions, possible ease of handling and the indication that relatively small
quantities would be required, the decision was made to' use  powdered carbon
on a periodic and throw away basis rather than use a granular bed  system.
                                     392

-------
    In the summer  of  1975 the conceptual plan illustrated in Exhibit
II was put out for bids to Engineering-Construction Firms as a "turn key"
project.  As a result Engineering was completed by December 1975, field
construction started  in February 1976, and completed by January 1, 1977,
all well within  the compliance schedule.

    At this writing the treating facilities are still  in the process of
starting up primarily because of an extraordinary length of time required
for biomass to develop on the RBS units and then delays in correcting
minor difficulties with certain mechanical equipment,  instruments and
electrical control systems.

    Our principle  concern was the difficulty in establishing the biomass.
During the pilot plant phase we had started up three different pilot
plants charging  similar waste water and in all cases a good growth was
established within 3  to 4 weeks.  However, in the case of our commerical
unit after four  weeks there was only a very slight indication of biogrowth
on the first stage.   It was determined that low water  temperature and
relatively low soluble BOD were responsible for the apparent lack of
bioactivity.

    With increased temperature and the addition of higher strength
waste water, supplied with a portable pump for 10 days, we observed an
 increase in the  growth of the biomass extending through all three stages.
 Up to this point only the RBS and the clarifiers were  in operation with
 the plant effluent returning to the feed surge ponds.  However, with the
 establishment of biomass, the filter, polymer injection system and sludge
 digester were all  put into operation.  The carbon system was operated for
 a short period primarily to test it mechanically.

    Only limited data has been obtained at this time however, they
 indicate the plant will perform satisfactorily and the waste water
 discharge will be  in  compliance with the July 1, 1977  limitation.
BIOGRAPHY          James F. Dehnert

        James F. Dehnert is the Environmental
Director for the Avon Refinery of Lion OP Com-
pany at Martinez, California.  He  has a B.S.
degree in Chemical Engineering and a B.S. de-
gree in Chemistry from Washington  State University.
He has been employed at this Refinery for thirty
years with various assignments in Techanical
Service, Economic Planning and Unit Operations.
He served as an Area Operation Supervisor be-
fore becoming involved in  Environmental assign-
ments.
                                     393

-------
                                              EXHIBIT I
u>
VO
                     QOH-EFCS
                     COOUIMS TOV/tf'S.S
                         R.EFIU£R.Y
                         OPERATIOM
                                   Tout. WATER.
                                   STR.IPPER.
                                                                                  OXIDATIONS
                                                                               V       PONJD
                                                                                 ^
V CLEA.M CA-WX^L. 1
' DISCHARGE
m.

-------
                                             EXHIBIT II
Cn

-------
                           TABLE 1

              TREATMENT OF RBS EFFLUENT WITH
                 POWDERED ACTIVATED CARBON
Test

No. 1
Sample

RBS Feed
RBS Eff
RBS Eff +20 ppm PAC
  TLm   or   Survival*
  74
  90
                                                          90%
No. 2
RBS Feed
RBS Eff
RBS Eff + 10 ppm PAC
  80
                                                          90%
                                                         100%
No. 3
RBS Feed                       80
2nd Stage RBS                  92
2nd Stage RBS +10 ppm PAC
                                                          90%
No. 4
RBS Feed
RBS Eff
RBS Eff + 10 ppm PAC
< 35
< 75
                                                          60%
No- 5
RBS Feed
RBS Eff
RBS Eff + 10 ppm PAC
  33
  64
                                                          60%
No. 6
RBS Feed
RBS Eff
RBS Eff + 20 ppm PAC
     "  + 35 ppm PAC
     "  + 50 ppm PAC
  33
 >69
                                                          90%
                                                          90%
                                                         100%
*Survival in undiluted waste
                                396

-------
                     TABLE II


    ACTIVATED SLUDGE EFFLUENT TREATED WITH PAC
Parameter

Toxicity  (% Survival)

COD              mg/1

NH3(N)           mg/1

Oil              mg/1

Naphthenic Acids mg/1

Cr(T)            mg/1

Cu               mg/1

Zn               mg/1
Act Sludge Eff

   0  (24 hr)

   108

   35

   0.2

   1.5

   0.02

   0.20

   0.03
Act Sludge Eff
+100 ppm PAC

  100 (96 hr)

   84

   28

   0.1

   0.6

   0.02

   0.25

   0.02
                                 TABLE III


                ACTIVATED SLUDGE EFFLUENT TREATED WITH PAC
Carbon Dosage (ppm)


Parameter
  Toxicity (% Survival)

  COD    mg/1

  Phenol mg/1

  Oil    mg/1

  Naphthenic Acid mg/1
                                         50
                         100
                                                           150
0
150
4.8
0.1
3.1
10
130
4.9
0.1
4.3
100
120
4.7
0.1
3.5
100
120
5.2
0.1
3.1
                               397

-------
                         TABLE IV

                EFFECT OF pH ON TOXICITY
       RBS Effluent @ 7.2 pH
            40%  Survival
       RBS Eff Lowered to 6.5 pH
             0%  Survival
       RBS Eff Raised to 8.5 pH
            90%  Survival
                          TABLE V

            EFFECT OF pH ON CARBON TREATMENT
       RBS Feed 7.1 pH
                                          36 TLm
       RBS Feed @ 6.5 pH + 30 ppm PAC     65 TLm


       RBS Feed @ 7.0 pH + 30 ppm PAC     61 TLm


       RBS Feed @ 7.5 pH + 30 ppm PAC     80 TLm


                          TABLE VI

                  COMPARISON OF TWO CARBONS
Sample

RBS Effluent (as is)
RBS Eff +60 ppm PAC-A
        +90 ppm PAC-A
Toxicity

 58 TLm
 85 TLm
 93 TLm
TOC (mg/1)

    68
    57
    57
RBS Eff + 15. ppm PAC-B
        +30 ppm PAC-B
        +60 ppm PAC-B
 93 TLm
 90% Survival
100% Survival
 398
    57
    54
    44

-------
                    TABLE VII

            COMPARISON OP TWO CARBONS
PAC-A Added Continuously to RBS Pilot Plant and
Additional PAC-A or PAC-B Added to RBS Effluent
RBS Eff '+  45 ppm PAC-A                  40 TLm
RBS Eff +195 ppm PAC-A                  86 TLm
RBS Eff +45 ppm PAC-A + 50 ppm PAC-B    93 TLm
RBS Eff +45 ppm PAC-A +75 ppm PAC-B   100% Survival
                          399

-------
20
                       TABLE VIII

GRANULAR CARBON TREATMENT OP TRICKLING FILTER EFFLUENT
COD mg/1 Toxi

Day
1
2
3
4
5
6
7
8
9
10

Feed
180
180
180
200
260
300
-
-
270
320

No.l
70
80
100
120
210
260
240
250
270
300
Column
No. 2
60
30
50
50
140
160
170
200
220
250

No. 3
30
30
30
30
80
110
100
140
170
210

No. 4
10
30
40
50
90
100
70
100
140
170

Feed No . 1
100
100
90
70
0




61TLm
       Toxicity (Survival)

             Column
              No. 2   No. 3
                                                               No. 4
220
210
170
160
100
34TLm
                                                 100
                                                 100
                                                        100
30
45
50
55
60
70
75
100
180

Hot
190
150
150
Hot
170
130
160
170
Water
160
140
150
Water
170
130
160
170
Wash
120
140
140
Wash
160
120
130
150
120
130
140
130
110
130
130
120
100
140
130
71TLm
45TLm

53TLm
59TLm
45TLm
100
80
0

100
0
80

0
100
100
100
100
                              400

-------
               SESSION VIII

ADD-ON GRANULAR ACTIVATED CARBON


           Chairman

           Nicholas D. Sylvester

           Professor of Chemical Engineering
           University of Tulsa,  Tulsa, Oklahoma


           Speakers
           Fred M. Pfeffer

           W.  Harrison and L. Raphael ian

           "Pilot-Scale Effect on Specific Organics Reduction
           and Common Wastewater Parameters"

           R.  H. Zanitsch

           R.  T.  Lynch

           "Granular Carbon Reactivation-State of the Art"

           L.  W. Crame
           "Activated Sludge Enhancement:  A Viable
           Alternative to Tertiary Carbon Adsorption?"
                   401

-------
BIOGRAPHY            Nicholas D. Sylvester

       Nicholas D. Sylvester is Professor and Chairman
of the Resources Engineering Division - Chemical and
Petroleum 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 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
investigator 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.
                                       402

-------
    ORGANICS REDUCTION THROUGH ADD-ON ACTIVATED CARBON AT PILOT SCALE

                              Fred M.  Pfeffer
                  U. S. Environmental Protection Agency
            Robert S. Kerr  Environmental Research Laboratory
                            Ada, Oklahoma  74820

                    Wyman Harrison and Leo Raphaelian
                       Argonne National Laboratory
                        Argonne, Illinois  60439

ABSTRACT

    The current wastewater  BATEA model for the petroleum refining industry is
the treatment sequence:  activated sludge, mixed-media filtration, activated
carbon.  In an effort to develop data to assist in evaluating the model for
specific organic compounds,  the EPA (Ada, Oklahoma) entered into an Interagency
Agreement with ERDA (Argonne National Laboratory) in January 1975.  In cooper-
ation with API, a  .25 GPM pilot test was conducted at the SOHIO Refinery in
Toledo, Ohio.  Argonne followed with GC/MS analysis of samples collected across
the treatment system to identify specific organics which are treatable versus
those which pass-through  (refractories).

    The EPA's involvement included:  the mobile pilot plant, refinery selec-
tion, conduct of the field study, sample preparation, and reporting.  Argonne1 s
analytical results showing a small overall reduction in organics by mixed-media
filtration and a large reduction by carbon adsorption are discussed.

INTRODUCTION

    In January 1975, the EPA (Robert S. Kerr Environmental Research Laboratory,
Ada, Oklahoma) entered into  an Interagency Agreement with ERDA (Argonne Nation-
al Laboratory, Chicago) to develop data to assist in evaluating the performance
of the BATEA model in the Development Document of 1974 (1).  Since that time
the BATEA regulations (and hence the BATEA model) have been remanded by a
ruling of the 10th Circuit Court on the petition for revision of the guidelines
by the API (2).  However, the requirements for reconsideration and reissuing
of guidelines as stipulated  in the ruling, together with the mandates in PL-
92-500 (Sec. 301. d.) (3), and the Settlement Agreement between EPA and NRDC
(4), are added incentive to  complete the work funded through this Interagency
Agreement.

    The proposed BATEA model was fixed bed carbon adsorption added onto the
BPT model, which is biological treatment followed by granular media filtra-
tion.  The specific treatment train selected for study was activated sludge,


                                     403

-------
mixed-media filtration, and granular activated carbon.   The performance cri-
terion was to be the reduction of major  organic compounds identified in the
influent to the biological treatment system.   Pursuant  to the agreement,
Argonne would perform qualitative organics  analyses on  samples provided by
EPA.  Pilot-scale filtration and carbon  adsorption would be applied to the final
effluent from a full-scale refinery treatment  system.   The results would serve
as guidance for determining the need for larger-scale study and would not be
used in predicting the performance of a full-scale  add-on  carbon system.

Refinery Selection

     Considerable time was allocated to refinery selection, as  there was  suf-
ficient funding to study only one refinery.  Repeated discussions and meetings
were held with members of the API's W-20 Task Group to arrive at a "represent-
ative" refinery.  It was agreed to acquire permission from a Class B refinery
whose final effluent quality met BPT, with the possible  exception of suspended
solids.  Other criteria would include intake water  quality and  variability,
refinery turnaround plans, and final effluent quality, raw waste loading, and
hydraulic detention times typifying the activated sludge process at a Class B
refinery.

     Agreement was reached in September 1976, to conduct the study at SOHIO's
Toledo refinery.  This is a Class B refinery (crude topping and catalytic
cracking) with coking, having a crude capacity of 120,000  BPSD.   The treatment
train at that time consisted of the API Separator,  dissolved air flotation
(DAF), activated sludge (extended aeration) having  16-18 hours  detention, and
final clarification.  The final effluent quality routinely satisfied BPT re-
quirements with the exception of suspended solids.  The  refinery treatment
system returned to steady state in November 1976, following a 1-month turna-
round period.

The Pilot Study

     A 30' EPA mobile trailer was transported to Toledo  and positioned near
the final clarifier.  Facilities aboard the trailer included 6" I.D. glass
columns for filtration and carbon adsorption (Figure 1), a TOC  analyzer  for
monitoring organic carbon breakthrough, pumping and distribution capability,
and sampling gear.  The sampling equipment, pumps,  and distribution lines were
fabricated and installed such that the only materials in contact with water
moving through the pilot treatment system were  stainless steel,  glass, Teflon,
and polypropylene (Figure 1).  Sampling points  aboard the  trailer were:   1)
SOHIO's final clarifier effluent, 2) pilot mixed-media filter  effluent,  and
3) pilot carbon column effluent (Figure  2).  The two remaining  sample points
were SOHIO's intake water and DAF effluent  (Figures 3 &  4).  These five  points
were sampled and iced on 4-hour intervals for 24-hour compositing over a con-
secutive 4-day period.  During the study, there were no  significant changes
in recorded flows through the full-scale treatment  system, as measured by the
hourly biofeed pumping rates.

     Two parallel down-flow mixed-media  filters were utilized  such that  while
one was operating for 24 hours, the second, having  been  backwashed, was  ready
for use the next day (Figure 5).  Figure 2  shows the configuration of  the


                                     404

-------
filtering bed:   anthrafilt, sand, and  gravel.   The sand used has  an  Effective
Size = 0.2 mm and a Uniformity Coefficient  = 4,5.   Backwashing was accomplished
by alternately pulsing with air and  pumping carbon column effluent.

     Two up-flow carbon columns  (Figure  6)  were packed as shown in Figure  2
and operated in series to achieve a  total bed depth of 6 feet.  A constant
flow rate of 0,25 gpm was maintained,  giving a residence time in  the carbon
bed of  36 minutes.   The  carbon used was Calgon's Adsorption  Service Carbon.
Calgon's analyses of a sample from the lot used at Toledo gave these results:

     Apparent Density (gcc):   0.51
     Molasses Number:          282
     Iodine Number:             821
     Sieve  Result  (mesh):      8x40

     Attention  was  given to decontaminating material coming  in contact with
water samples.   All glassware was cleaned by firing, maintaining 550°C for
1-hour.   Sample bottle caps contained Teflon liners which had been cleaned by
Soxlet extraction with methylene chloride—the solvent  later used in the
laboratory for  extracting the organics from the water  samples.

     Each daily composited sample set was transported  in ice chests to Detroit
for air shipment.   The samples arrived at RSKERL in Ada within 9 hours of
final compositing  in Toledo.

Laboratory Phase

     Performance of the  full-scale biosystem and the add-on  filtration/carbon
train for the  common wastewater parameters is shown in Tables 1 & 2.  Some
values are reported as less-than (<), reflecting lower  limits of detectability
as a function  of the sampling and analytical protocol.

     Following  the  field study, the remaining responsibility of EPA was the
preparation of  the  composited water samples for organics analysis by Argonne.
This involved  a tedious  liquid-liquid extraction sequence using methylene
chloride.   Again, all glassware was fired for organics  decontamination.  A
major problem was emulsion formation, requiring emulsion breaking and phase
separation by various techniques.  Each organic extract was  dried by passing
through anhydrous  sodium sulfate and the solvent was stripped, resulting in
1-ml of concentrated extract which was sealed in a glass ampul.  A period of
9 man-hours was involved in preparing each sample to the ampul stage; there
were 20 samples requiring this preparation.

Gas-Chromatography/Mass-Spectrometry (GC/MS)

     The samples supplied to Argonne for analysis consisted  of 1 mililiter
methylene chloride  solutions of the acid, base, and neutral  fractions com-
posited over the 4-day sampling interval.

     Analysis of the specific organics in these fractions was performed on a
Hewlett-Packard Gas Chromatograph/Mass Spectrometer equipped with a  -unta syp."^:


                                     405

-------
and such  peripheral  equipment  as a Zeta plotter and hard copy unit.  Capillary
columns were  used  in the  gas chromatograph.   These columns allow considerably
greater separation and  resolution of the organic components in a sample than
do standard packed columns.  Capillary columns also provide increased sensi-
tivity and drastically  reduced background from column bleed in the mass spec-
tra.  Also as opposed to  typical GC/MS operation, no separator was used to
remove the carrier gas.   The outlet of the capillary column was connected
directly  to the  source  of the.mass spectrometer and, therefore, there could
be no discrimination in the amount of each component reaching  the mass spec-
trometer.   That is, assuming that the individual  components in the mixture are
not lost in the column,  the effluent of the column and the amount of  these
components reaching the source of the mass spectrometer  is a true representa-
tion of the quantities of compounds injected on the  column.  Finally, a Grob-
type injection system was used in place of the inlet splitters typically used
with capillary columns.   The Grob system avoids the  loss of large amounts of
samples and the discrimination, typically found in split systems, of  compon-
ents of the mixture.   It permits the analysis of  minute  concentrations of the
specific organics present.

     Figure 7 is a capillary column GC/MS total ion  chromatogram of the neutral
fraction of the dissolved-air-flotation effluent.  It can be seen that there
are over one hundred peaks or components in this  fraction and  that many of the
components are present in minute quantities; that is, of the order of 200 ppt
of the original water sample,  assuming 100% extraction efficiencies.  It was
found that the organics in this neutral fraction  of  the DAF effluent were
predominantly n-alkanes, alkyl benzenes, alkyl naphthalenes and polynuclear-
aromatic hydrocarbons.

     The activated sludge treatment system reduced the concentration of the
organics in the DAF effluent by nearly 98%, as shown in Figure 8.  It can be
seen from the graphs that the peak height of several of  the peaks in  the FC
effluent is approximately one-twentieth those in  the DAF effluent, indicating
there is approximately a twenty-fold reduction in pollutants by the activated
sludge process.

     Further reductions in organics were accomplished by the multi-media fil-
ter arid the activated carbon column as shown in Figure 9.  The concentration
of the largest peaks of the compounds refractory  to  the  add-on treatment
system is of the order of 10 ppb.  The percent reduction of the major classes
of organics by the multi-media filter and the activated  carbon column is as
follows:

           	Compound                  . % Reduction
            Alkanes                                    70-98
            Alkyl Benzenes                             35-90
            Indenes                                    50-60
            Indanes                                    76-96
            Naphthalene                                 66
            Alkyl Naphthalenes                         65-90
            Anthracene/Phenanthrene                    86-93
            Alkyl Anthracenes/Phenanthrenes            89-98
            Ot-her PNAs                                 96-98

                                      4Q6

-------
     It can be seen that  there is generally greater than fifty percent reduc-
tion in these classes of  organic compounds.

     Work has not yet been  completed on the acid and base  fractions.  Results
will be available in a  few  months.  The results of this study will be published
as EPA and Argonne reports  and as such will be available to  the public.

     It  is  expected  that sufficient  funds  will be forthcoming in the next
fiscal year to  search and manipulate the data  stored  on disc for those consent-
decree organics that may be present  in  these samples.

REFERENCES

(1)   Development Document for Effluent  Limitations, Guidelines and New Source
     Performance Standards for the Petroleum Refining Point Source Category,
     EPA-440/l-74-014-a, April 1974.

(2)   Ruling of  the 10th Circuit  Court,  Denver, on the suit of EPA by API,
     handed down August 11, 1976.

(3)>  Public Law 92-500, 92nd Congress,  S.2770, October 18,  1972.

(4)   Settlement Agreement in the U.S. District Court  for the District of
     Columbia between the Natural Resources Defense Council and the U.S."
     EPA,  Civil Action No. 2153-73,  June  7, 1976.

ACKNOWLEDGEMENT

     The authors wish to acknowledge the assistance of the Calgon Corporation
relating to activated carbon and the efforts of the API's Water Quality Com-
mittee and W-20 Task Group in  arriving  at  a suitable  refinery.  We wish to
thank Messrs. C. Tome, L.S. Van  Loon, and  J.H. Walters for assistance in the
wastewater sampling program and  Mrs. C.S.  Chow for help with the GC/MS
analyses.   Most important, the study would not have been possible without the
cooperation of SOHIO personnel at the refinery in Toledo and in the Department
of Environmental Affairs in Cleveland.

DISCUSSION

Peter J. Foley, Mobil Oil;  Would you comment  on the contributions of the
filter and the carbon columns  in reducing  organics?

 Raphael fan;  Although I do not have the exact numbers at hand, it appears to me that the
 multi-media filter had little or no effect  on the concentration of organics whereas the
 activated carbon removed appreciable amounts of organics.

 L_AjV\cConomy, Calgon Corp.; The naphthalene  removal was only 66% as compared
 to more  than 90% for other PNA's, why?

 Ra^haeliaru  These numbers are approximate  figures based on the average of peak areas of
 individual  components.  Generally,  one  can say that those organics that have a long

                                      407

-------
alkyl chain such as an alkane or an alkylated PNA are adsorbed on the carbon whereas
parent PNA's are not.  This is, of course, a crude approximation. We are still trying to
get better gas chromatograms because all of these results are dependent upon how well you
separate the  compounds.

N. F. Seppi, Marathon Oil;  Please comment on methylene chloride purification - also
what about decomposition products from methylene chloride under basic conditions?

Pfeffer;  Regarding  the purity  of  methylene chloride,  we relied upon the glass
distillation procedures of the  manufacturer (Burdick and Jackson).   In addition,
Argonne  received a blank  extract obtained by taking a high purity water  and
performing  the  acids, neutrals  and basic extractions.   This blank would  also
account  for laboratory contaminations.   I cannot offer any information about
alkaline decomposition products.

Anonymous:  Is  there additional data  from industry on the study at Toledo?

Pfeffer;  I do  not know.   Both  Exxon  and SOHIO conducted parallel work to our
own, presumably into the  realm  of  GC  mass-spec.  We would entertain  comparing
notes with  Exxon and SOHIO at some later date in order to validate what  actu-
ally took place in Toledo.

Judith Thatcher. API:  I  noticed that  the TOG of the influent water  is very
close to the TOG of  the final effluent.   Have you done any identification in
the organics in the  influent  water to the refinery?

Pfeffer;  We are looking  at it,  but haven't identified all the components yet.

Arthur J. Raymond. Sun Oil Co.;  What  phase was used on the capillary columns,
and did  you notice that your  highly branched-chains were degraded much faster
than the less branched?   Also,  was benzene degraded much faster than toluene
and xylene?

Raphaelian; I  don't understand what  you mean by degraded.

Raymond:  Did they decompose  faster or disappear or reduce?  Not in  the  column
but in your system as you went  from the influent water to the final  clarifier.
If you had  percent reductions,  which  compounds went faster than others?

Raphaelian;  I  am still putting all this data together.  However, I  can  say
that it  appears that the  branched-chained alkanes, which were present in smaller
quantities  than the  straight-chained  alkanes, were not removed as well by the
treatment system as  the straight-chained alkanes.  Because of the minute
quantities  of pollutants  present,  I am presently doing single-ion monitoring
to try to get a better idea of  the percent reduction across the treatment
system.

Raymond;  What  phase did  you  use on a capillary column?  What coating?

Raphaelian;  For the work presented in this talk, OV-101 was the liquid  phase
and the  columns were wall coated open tubular  (WCOT) and not support coated


                                      408

-------
open tubular  (SCOT)  capillary  columns.  It  is  difficult with OV-101  to  get
symmetrical peaks  with polar compounds, that is,  peaks without tailing.   I
used FFAP capillary  columns for the acid and base fractions.  By the way  we
see a variety of alkylated phenols in the acid fraction.                 '

Ed Sebesta. Brown  &  Root!  I noticed that for  TSS there is no decrease  across
the filter.  Do you  have any explanation for that?

Pfeffer:  My only  explanation  is that considering the flow rate and  sand spec-
ifications, the TSS  coming from the final clarifier were such that the  filter
was ineffective. Also, at the 10 mg/l  level, differences are probably within the experi-
mental error of the test procedure.

BIOGRAPHIES
 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.
 Wyman Harrison holds the SB, SM,  and a PhD degree in
 Geophysics from the University of Chicago.  He is
 currently the Assistant Director of Applied  Geoscience
 and Engineering, Energy and Environmental Systems
 Division at the Argonne National Laboratory, Argonne,
 Illinois.
 Leo Raphaelian holds the AB degree from Harvard
 University and the MA and a PhD degree in Chemistry
 from Yale University.  He is currently Manager of
 Environmental  Sciences, Energy and Environmental
 Systems Division at the Argonne National Laboratory,
 Argonne, Illinois.
                                       409

-------
Table 1.  DAILY PERFORMANCE FOR COMMON WASTEWATER PARAMETERS


Oil & Grease
Cyanide
Phenol
COD
BOD
TOC
TSS


Oil & Grease
Cyanide
Phenol
COD
BOD
TOC
TSS
MG/L INTAKE MG/L DAF EFFLUENT MG/L FC EFFLUENT
Day 1 Day 2 Day 3 Day 4
<10 <10 <10 <10
<0.02 <0.02 <0.02 <0.02
0.03 <0.01 0.03 0.01
<15 18 <15 <15
<10 <10 14 <10
19 19 17 15
35 29 11 <10
Day 1 Day 2 Day 3 Day 4
22 33 21 22
0.19 0.25 0.31
320 260 520 450
122 172 154 154
82 127 108 96
39 56 72 60
31 56 37 30
Day 1 Day 2 Day 3 Day 4
<10 <10 <10 <10
0.16 0.12 0.20 0.10
0.02 0.01 0.04 0.02
49 50 51 44
<10 15 21 24
22 29 27 17
12 <10 <10 <10
MG/L FILTER EFFLUENT MG/L CARBON EFFLUENT
Day 1 Day 2 Day 3 Day 4
<10 <10 <10 <10
0.16 0.15 0.20 0.10
0.02 0.01 0.02 0.02
42 38 51 44
<10 11 22 27
19 26 23 18
<10 <10 12 12
Day 1 Day 2 Day 3 Day 4
<10 <10 <10 <10
<0.02 <0.02 <0.02 <0.02
<0.01 <0.01 <0.01 <0.01
<15 <15 <15 <15
<10 <10 <10 <10
10 12 11 <5
<10 <10 <10 <10

-------
           Table 2.  AVERAGE  PERFORMANCE OVER 4-DAY STUDY PERIOD
                     FOR  COMMON WASTEWATER PARAMETERS
Oil  & Grease

Cyanide

Phenol

COD

BOD

TOC

TSS
                MG/L  INTAKE
<0.02

 0.02
  18

  21
MG/L DAF

   24

  0.25

   390

   150

   103

    57

    38
                      MG/L FC   MG/L FILTER   MG/L CARBON
0.14

0.02

 48

 17

 24
0.15

0.02

 44

 17

 22
<0.02

<0.01
                                     411

-------
Is)
             Figure 1.  CARBON COLUMNS
Figure 3.  SAMPLING POINT:  PLANT  INTAKE

-------
                                                   WASTE
00
                                                                   .*'•'* '/«'«
                                                                   ;« «t-«'A;

                                                                   •T-SAND
                                                                   J*' - 4 ' «'
                                                                   A-:4 *«*>.'
FINAL CLARIFIER
                                 FIGURE  2 - PILOT TREATMENT FACILITY

-------
Figure 4.  SAMPLING POINT:  DAF EFFLUENT

-------
Figure 5.   MIXED MEDIA FILTERS
Figure 6.  CARBON COLUMNS

-------
     Ill
                                                DAF  EFFLUENT
20
40
                                '   '   '
                                     eo
80
100           120
     TIME  (min)
Figure  7. Total Ion Chromatogram of DAF Effluent (Neutral Fraction, Four-Day Composite)

-------
                                                  OAF EFFLUENT
                                          jii^^
                                                       i—i—i—i—i
T—i	1	1	1	r
   T	1	1	1	1	r
     V
                                                  FC  EFFLUENT
                                                  (IOX)
0
        20
40
   60
TIME  (min.)
80
                                               i   n
100         120
Figure 8.  Total Ion Chromatograms of DAF Effluent and FC Effluent (Neutral Fraction
         Four-Day Composite)                                         '

-------
                                                                        ACTIVATED-CARBON
                                                                        EFFLUENT  (IOX)
                                     ^^^^^
oo
                                                                        FINAL CLARIFIER
                                                                        EFFLUENT
                                                     MAA'
 o
    I    I
20
                       I    I
40
  60
TIME (mm.)
                                                             III
80
100
120
Figure  9.  Total Ion Chromatograms of the Activated-Carbon and Final-Clarifier Effluents (Neutral Fraction
          Four-Day Composite)                                                                '

-------
              "GRANULAR CARBON REACTIVATION:  STATE-OF-THE-ART"


                               R. H. Zanitsch
        Engineering Director, Calgon Environmental Systems Division


                                 R. T. Lynch
                    Process Engineer, Calgon Corporation


     Use of granular activated carbon for treatment of industrial waste-
water is receiving widespread acceptance.  In the past several years, 100
adsorption systems have been installed in industrial plants.  Applications
range from dye plant wastewater reuse to removal of toxic materials.  Gran-
ular carbon is being used to treat flows as low as 1,000 gallons per day
to as high as 20,000,000 gallons per day in industrial waste applications.
It is being employed as a pretreatment step to remove toxic materials prior
to biological treatment, as the main treatment process and for tertiary
treatment  of biological plant effluents.

     In most industrial wastewater applications, cost of virgin carbon pro-
hibits using it on a throw-away basis.  Chemical regeneration is feasible
in only a  limited number of applications and regenerant disposal remains a
problem.  Thermal reactivation is in most cases complete, efficient, and
economical whether it is performed on-site or on a contract basis at a
central reactivation facility.

     The technology of reactivation with industrial waste carbons has de-
veloped in only the last ten years. There are now approximately twenty re-
activation systems installed in the United States which are reactivating
industrial wastewater carbons.

     New thermal reactivation processes (such as fluidized beds and elect-
ric furnaces) are now being developed but no commercial experience with
industrial wastewater carbons has been developed in the United States.
For the purposes of this presentation, we will discuss our experience with
the design and operation of multiple hearth furnaces and rotary kilns as
they relate to industrial wastewater applications.
THE THERMAL REACTIVATION PROCESS

     Granular carbon is usually wet when fed to the reactivation furnace.
Water concentration is a function of carbon size, water temperature dur-
ing the dewatering step, and the amount of adsorbate on the carbon.  In
practice,  moisture content varies from 40 to 50 percent on a wet spent
basis.

     The reactivation process can be divided into three steps:


     1.  Evaporation of moisture on the carbon (Drying).

                                    419

-------
     2.   Destructive distillation of organics resulting in pyrolysis of
         a portion of the carbon from the organic materials  (Baking).

     3.   Activation of the carbon by selectively burning carbon deposited
         during the organic removal step (Activation).

     During the drying step, carbon temperature is increased to approxi-
mately 212°F (100°C) and moisture evaporates into the gas phase.  As moist-
ure evaporates it is also possible for highly volatile organics to be steam
distilled.

     The second step is termed baking or pyrolysis of the adsorbate.  Dur-
ing this step, carbon temperature increases to approximately 1200-1400°F
(649-760°C).  A portion of the organic molecules are thermally cracked to
produce gaseous hydrocarbons which are driven off.  The remaining lower
molecular weight organics are distilled.  During this process, a carbon char
is deposited in the pore structure of the original activated carbon.

     The final step is activation of the carbon - a chemical reactivation
whereby carbon char deposited during the baking step is combusted along
with a small amount of the original carbon.  By this time, temperatures
are in the range of 1600-1800°F (871-982°C).

     Since the fixed carbon and the granules are both carbon, the process
requires that fixed carbon be selectively gasified with minimum gasifi-
cation of the granular carbon.  Steam is added to the furnace and oxygen
concentration is controlled to promote gasification of the fixed carbon
while minimizing burning of the original granular carbon.
REACTIVATION SYSTEM DESCRIPTION


     The basic sequence for thermal reactivation is as follows:   (See Ex-
hibit 1).

     Spent carbon is removed from the adsorbers and transferred as a slurry
to a spent carbon storage tank.  Spent carbon is then transferred to an
elevated furnace feed tank from which it is metered, at a controlled rate,
to a dewatering screw.  The dewatering screw is an inclined screw conveyor
which serves the dual purpose of gravity draining slurry water from the
granular carbon and providing a water seal for the top of the furnace.  A
timer operated valve is used to meter carbon to the dewatering screw.
Drained, but wet, spent carbon then gravity flows into the furnace where
it is dried, baked, and reactivated as discussed earlier.  Reactivated
carbon exits the furnace by gravity and enters a quench tank.  The quench
tank serves the dual purpose of wetting the reactivated carbon and provid-
ing a bottom seal for the furnace.  The carbon is then transferred to a
reactivated carbon storage tank from which it is then returned to the ad-
sorbers as needed.  In most industrial waste applications, an afterburner
and scrubber are provided for destroying organics and removing residual
particulates from the furnace off-gases.


                                    420

-------
    The  spent  carbon storage tank should be designed for five to ten days
storage of  carbon in order to allow for routine furnace maintenance and un-
scheduled shut-downs.  This is usually a lined carbon steel tank with a
cone bottom to  facilitate carbon flow.  The reactivated carbon storage tank
is usually  sized in the same manner with the same materials of construction.
In the case of  the reactivated carbon storage tank, facilities should be
provided  for adding virgin makeup carbon to the system as required.

    The  furnace feed tank is usually sized for at least one shift of op-
eration.  The feed tank insures a constant carbon feed to the furnace in-
dependent of the large storage system.  This tank is usually a cone-bottom,
lined  carbon steel tank.

    Spent  and reactivated carbon are transferred in slurry form using
either eductors, blowcases or slurry pumps.  In the case of eductors and
pumps, dilution water must be provided in order to reduce slurry concen-
tration  to  less than one pound per gallon to minimize carbon abrasion and
line erosion.  Eductors are generally applicable in non-corrosive services
where  static head is not great.  Pumps can be used satisfactorily in high-
head applications, but are subject to erosion and plugging.  Eductors and
pumps  require use of dilution water which necessitates installation of
water  recycle systems.  The blowcase is an efficient method of transferring
carbon.   Both air and water have been used to pressurize blowcases.  In the
case of  a blowcase, carbon is transferred in a much denser slurry (three
pounds per  gallon) and, therefore, care must be taken to maintain control
over line velocities to minimize abrasion and wear.  Material for carbon
slurry lines should be compatible with the wastewater.  As long as slurry
lines  are flushed free of carbon after each transfer, galvanic corrosion
of carbon steel lines will not be a problem; however, if the wastewater is
corrosive,  more exotic materials of construction should be used.  All car-
bon slurry  lines should be equipped with flush connections to facilitate
flushing and unplugging.
CAPITAL AND OPERATING COST ESTIMATES

     Based on our experience with the design, installation, and operation
of multiple hearth furnaces and rotary kilns for reactivating industrial
waste carbons, we have estimated the installed cost of reactivation systems
to reactivate 5,000, 10,000, 30,000, and 60,000 pounds per day.  (See Ex-
hibit 2);   The capital cost curve shown in Exhibit 3 represents a total
installed cost including all equipment, site preparation, foundations, in-
stallation, startup, and indirects.  We have assumed that necessary util-
ities and off-site facilities are available at the battery limits.  As you
can see, we estimate the total installed cost of a 10,000 pound per day
reactivation system to be approximately $1.25 million plus or minus 20 per-
cent.  The time required to design, procure, install, and startup a re-
activation system is usually estimated to be two years assuming a twelve-
Jnonth delivery time on the furnace and associated equipment.

     We have also estimated direct operating costs for reactivating 5,000,


                                    421

-------
10,000, 30,000, and 60,000 pounds per day of industrial wastewater carbons as
shown in Exhibit 4.

     In order to develop these costs, the following elements were consider-
ed:

     1.  Labor was estimated to be one operator per shift at a rate of
         $10/hour.  An allowance of 25 percent of the labor cost for
         supervision was also included.

     2.  Fuel was estimated at 8,000 BTU's per pound at a cost of $3/million
         BTU's.  This estimate includes afterburner operation and an allow-
         ance for inefficiencies due to interruptions and reduced feed
         rates.  Approximately half of the fuel consumption is required for
         the afterburner and idling.

     3.  Power costs for the reactivation system are minimal and were as-
         sumed to cost $0.03/KWH.

     4.  Steam costs were based on an average demand of one pound of steam
         per pound of carbon for reactivation at a cost of $4/1,000 pounds.

     5.  Maintenance costs for an industrial wastewater application can
         range from 8 to 15 percent of the reactivation system cost per
         year.  For this estimate, we assumed a maintenance cost of 8 per-
         cent per year.

     6.  Makeup carbon cbsts were based on an average carbon loss rate of
         7 percent and a virgin carbon cost of $0.57/pound delivered.
         Carbon losses can range from as low as 3 percent to greater
         than 10 percent depending on design and operation of the system.
         Most industrial waste systems operate in the 5 to 7 percent loss
         range.  Makeup carbon costs represent the highest individual cost
         element in the direct operating cost estimate and, therefore, all
         efforts should be made to minimize carbon losses through good de-
         sign and operation.

     7.  A general plant overhead of 10 percent of the above cost was al-
         lowed to cover such items as insurance, taxes, monitoring, ac-
         counting, and administration.

     As can be seen from Exhibit 4, the direct operating cost for a re-
activation system handling industrial wastewater carbons ranges from
$0.11  to $0.19/pound over the range investigated.  This does not include
depreciation or amortization of investment.  The economies of scale are
obvious.  We feel these costs can r,£nge plus or minus 20 percent, but in
general, reflect the cost to operate a reactivation system on industrial
waste  applications.
                                    422

-------
MULTIPLE HEARTH FURNACE


     Exhibit 5 is a cross-sectional view of a multiple hearth furnace.  The
furnace consists of a cylindrical refractory-lined  steel  shell containing
several refractory hearths and a central rotating shaft to which rabble arms
are attached.   From four to eight hearths  are used  in carbon reactivation
furnaces.  The center shaft and rabble arms are  cooled by air supplied by
a centrifugal blower discharging air  through a housing into the bottom of
the shaft.   A sand seal at the top of the  furnace and a sand or water seal
at the bottom are used to seal the furnace against  introduction of extran-
eous air.

     In operation, wet spent carbon is introduced through a chute into the
outside of  the top hearth of the furnace.  The rabble arms are equipped
with solid  alloy rabble teeth which rake the carbon towards the center where
it drops to the hearth below.  The teeth on the  rabble arms are arranged to
move the carbon in a spiral path.  The action is gentle to minimize at-
trition.  The top hearth is termed an "in" hearth since carbon flow is in-
ward.

     The second hearth is consequently an  "out"  hearth where the carbon is
moved outward by the rabble teeth.  Out hearths  have a series of holes
around the periphery of the hearth through which the carbon drops to the
next lower hearth.

     In this manner, carbon passes through the furnace until it is finally
discharged through a chute in the bottom hearth  into the  water filled
quench tank.  The chute extends under the  water  level in  the quench tank
to provide a seal.

     Drying is accomplished in the upper one-third  of the furnace.  Dis-
tillation and pyrolysis of the adsorbate occurs  in  the next one-third.
Activation of the carbon is completed in the bottom one-third of the furn-
ace.

     Burners are mounted tangentially on the furnace shell in burner boxes.
Usually burners are placed on the bottom two or  three hearths and on one
upper hearth below the lowest drying  hearth.  However, if desired, burners
can be mounted on any hearth including the drying hearths.

     On small furnaces, two burners per fired hearth are  used.  On larger
furnaces, three burners are installed.  The burners are of the nozzle-mix
type burning fuel oil or natural gas. Dual fuel burners  are commonly
employed to burn gas when it is available  and fuel  oil at other times.

     Steam addition ports are provided on  the bottom two  or three hearths
to add steam for control of the reactivation process.

     The center shaft is driven through a  variable  speed  drive at 0.5 to
2.5 rpm.
                                     423

-------
     A number of furnaces have been installed with integral  or  "0" hearth
afterburners.  This is less costly from a capital  cost  standpoint than the
separate afterburner.
DIRECT FIRED ROTARY KILN

     Exhibit 6 is a simplified sketch showing a direct-fired  rotary kiln.
The kiln is a refractory lined steel shell enclosed  on  each end with re-
fractory lined stationary hoods.  This sketch depicts a counter-current
operation where gas flow is opposite the carbon flow.   Co-current  operation
is also possible and one carbon reactivation kiln  is currently operating
in this manner.

     The kiln is mounted on two or three sets of trunions  depending on the
length of the unit.  The kiln is sloped from the feed to the  discharge end
and one set of thrust rolls are used to maintain the kiln  in  position on
the trunions.  Proper training and alignment of trunions is important to
minimize excessive wear of the trunions and tires.

     The kiln is driven through a variable speed drive  coupled to  a speed
reducer and pinion gear which meshes with a bull or  girt gear mounted on
the kiln shell.  The kiln is equipped at each end  with  hoods.  Rotary seals
are used to seal between the rotating kiln shell and the stationary hoods.
The hoods are refractory lined.

     A feed screw or chute is used to feed wet carbon into the kiln.  Flights
are usually employed to advance the damp carbon and  to  shower the  carbon in
the feed end to obtain high heat transfer rates during  the evaporation step.
Flights are also used in the first portion of the  baking step up to a point
where the temperature reaches approximately 1200-1400°F (649-760°C).  Ma-
terial of construction for the flights is a function of carbon corrosive-
ness and reactivation conditions in the kiln.

     The hot reactivated carbon, at a temperature  of 1600-1800°F  (871-982°C),
discharges from the kiln and falls down the discharge chute into a water-
filled quench tank.  The discharge chute extends under  the water level in
the quench tank to form a seal to eliminate air leakage into  the kiln.

     A burner is mounted in the discharge hood to  provide  heat for the re-
activation process.  Either fuel oil or gas may be burned. The burner
air-to-gas ratio is adjusted to minimize oxygen concentration in  the  kiln.
A steam addition port also is provided in the discharge hood  to admit steam
into the kiln for control of the reactivation process.

     The exhaust gases, at a temperature of 500-800°F  (260-427°C), leave
the kiln through a duct connected to the feed hood.  In most  installations,
gases are passed through an afterburner for complete combustion of organics
and burning of carbon fines swept out of the kiln.  New installations,  as
is the case with the multiple hearth furnace, will probably require in-
stallation of a wet scrubber to meet air pollution codes in most  areas of
the country.

                                    424

-------
     In a rotary kiln, reactivation process is controlled by varying the
kiln speed to provide adequate retention time and by adjusting the steam
rate and burner temperature at the settings required for the particular car-
bon to be reactivated.  A steam rate of 0.6-1.2 pounds per pound of carbon
and a temperature of 1600-2000°F  (871-1093°C) are ranges encountered in
practice.


OPERATING AND MAINTENANCE PROBLEMS


     A number of unique operating and maintenance problems have been ex-
perienced in reactivating industrial wastewater carbons.  These problems
include:

                      Corrosion
                      Slagging
                      Poor Reactivated Carbon Quality
                      High Carbon Losses
                      Feed Interruptions
                      Hearth Failures
                      Slurry Line Erosion and Corrosion
Corrosion

     Selection of proper construction materials for carbon storage and
handling systems is very important.  We have found lined carbon steel
tanks to be satisfactory, but proper selection and application of lining
material is extremely important.  Lining material should be corrosion
and abrasion resistant.  We recommend thorough corrosion coupon testing
prior to making a final selection.  Erosion of lining material at carbon
outlet nozzles, followed by corrosion of the metal, has been a problem.
We have installed sacrificial wear plates or stainless steel cones on tanks
in order to minimize this problem.  Dewatering screws and quench tanks are
generally constructed of 304 or 316L stainless steel.  In general, these
materials are satisfactory for most applications.  However, the dewatering
screw is exposed to the spent carbon slurry and, therefore, its material
must be compatible with the wastewater.  Corrosion of rabble arms and teeth
in multiple hearth furnaces and lifting and drying flights in rotary kilns
can be a problem when handling chlorinated hydrocarbons and organic sulfur
compounds.   Special attention must be given to material selection to mini-
mize this problem.


flagging

     Formation of clinkers and slag in the furnace is generally a function
of sodium and/or organic phosphate content of the spent carbon.  Slag
formation can be minimized by pretreatment of carbon and maintenance of
proper furnace conditions.  Formation of slag can generally be attributed
to constituents in the water contained in the pores of the carbon as it
                                    425

-------
enters the furnace.  These chemicals react with alumina and  silica  in the
furnace refractory resulting in slag formation.
Hearth Failures

     Hearth failures in multiple hearth furnaces can generally be attributed
to cyclic operation.  Frequent feed interruptions, resulting  in  temperature
excursions on the upper hearths, will weaken hearths and ultimately lead to
failure.  By minimizing the number of feed interruptions and  maintaining
continuous furnace operation, upper hearth life can be maintained for three
to five years.  Another problem leading to hearth failure is  brick attack
by sodium compounds; which leads to the slagging problem discussed earlier.
Also, improper dewatering or improper operation of the dewatering screw,
which would result in excessive amounts of water entering the top hearth,
can result in thermal shock which leads to failure.  In general, hearth life
is a function of the operating philosophy of the furnace.  If frequent feed
interruptions due to improper carbon feed system design or cyclic operation
are encountered, poor hearth life can be expected.


Carbon Losses

     As mentioned earlier, makeup carbon cost is the single most important
cost element for a reactivation facility.  By properly designing the ad-
sorbers, the carbon transfer and handling systems and the carbon storage
and reactivation systems, losses can be controlled at the 5-7 percent level.
Within the reactivation furnace itself, carbon losses should  not exceed
1-3 percent.  Carbon which is lost is due to oxidation during the activation
step.  This can be controlled by maintaining oxygen levels in the activation
zone at 0-2 percent, or roughly that required for destruction of organics
without sacrificing carbon.  Most carbon losses in a granular carbon system
occur due to backwashing of carbon in adsorbers, abrasion in  slurry lines,
spillage, and carryover in overflow lines.  These losses can  all be mini-
mized by proper design of the basic system.  Care must be given  to overflow
rates, backwash rates, and slurry line velocities, and frequent  checks must
be made to see that good housekeeping and operating techniques are being
followed.
Slurry Line Erosion and Corrosion

     As mentioned earlier, construction material of spent  carbon  slurry
lines should be compatible with the wastewater in order  to minimize cor-
rosion.  If spent carbon or reactivated carbon is allowed  to  accumulate in
a carbon steel slurry line, galvanic corrosion can be expected.   Therefore,
flushing of all slurry lines after each transfer is recommended.   If the
wastewater is extremely corrosive, we recommend lined steel or  stainless
steel slurry piping be considered.

     Erosion of slurry lines can be attributed to excessive transfer


                                    426

-------
velocities.   We recommend a slurry line velocity of 3-5 feet per second
which is sufficient to prevent settling and minimize abrasion.  Also
slurry lines should be as direct as possible with a minimum number of
bends.  We recommend that long-radius bends be used to minimize abrasion.
Also, we recommend that all bends be accessible for periodic inspection
and replacement.  Flush connections should be provided at frequent inter-
vals on all slurry lines in case a line becomes plugged.


FURNACE SELECTION CRITERIA


     Choice of a reactivation furnace depends on many factors.  A thorough
analysis of each type of equipment plus reactivation characteristics of
the carbon are necessary to make a final decision on which piece of equip-
ment to use.

    .Both multiple hearth furnaces and kilns are being employed to react-
ivate granular activated carbon used in industrial wastewater treatment.
The quality of reactivated carbon that can be achieved is the same for both
units.

     Major parameters influencing the selection of a reactivation furnace
are as follows:

     1.  Capital Cost - Total installed costs for either a multiple hearth
         furnace or a rotary kiln are approximately the same.  The purchase
         price is generally higher for a multiple hearth furnace.  However,
         the installation costs are lower which tends to make installed
         costs the same.  Site preparation, foundations, and structural
         costs are higher for a rotary kiln because of the greater area re-
         quired to install a kiln.

     2.  Area Requirements - Area requirements are much greater for a rotary
         kiln than for a multiple hearth furnace.  The kiln also requires
         more foundations and structural steel for walkways than a multiple
         hearth furnace.  The multiple hearth furnace is higher than a kiln
         which means more structural steel is required to support the car-
         bon feed equipment.

     3.  Fuel Consumption - Fuel consumption is higher in a rotary kiln be-
         cause of higher heat losses.  In a multiple hearth furnace, in-
         sulation is used behind the wall brick to minimize heat loss.
         This is not possible in a rotary kiln.  Surface area is also higher
         in a rotary kiln than a multiple hearth furnace of equivalent ca-
         pacity.  Fuel consumption for each will be in the following ranges
         depending on capacity and operating rate as a fraction of rated
         capacity.
                                     427

-------
                                      BTU/LB Carbon*

          Multiple Hearth Furnace       2500-4500
          Rotary Kiln                   3500-8000

          *Does not include afterburner fuel requirements.

4.  Capacity Turndown - Capacity turndown ratio is defined as the per-
    cent of rated capacity at which the furnace can be operated while
    producing good reactivated carbon with reasonable carbon loss.
    Capacity turndown for the equipment being evaluated in this paper
    are as follows:

              Multiple Hearth Furnace  -  33 Percent
              Rotary Kiln              -  50 Percent

    The multiple hearth furnace can be operated at a lower fractional
    capacity because of the greater degree of control that can be ob-
    tained in various zones of the furnace.  In a rotary kiln, with
    only one burner and one steam addition point, kiln speed is the
    major parameter that can be varied to operate at lower capacities.

5.  Degree of Control - Better reactivation process control can be
    achieved in a multiple hearth furnace because the furnace is di-
    vided into distinct zones according to the number of hearths in
    the furnace.  Each hearth can be equipped with burners, steam
    addition, and air addition which can be controlled independently.
    Thus, it is possible to control temperature and vary the atmosphere
    in each hearth to optimize carbon reactivation.

    In a rotary kiln, the steam port, and burner can only be mounted in
    the firing end of the kiln.  With this arrangement, the degree of
    control that can be achieved is less than in a multiple hearth
    furnace.  In a properly sized kiln, this is not a distinct dis-
    advantage and good carbon reactivation can be achieved.  However,
    as discussed previously, capacity turndown is not as great in a
    kiln.

6.  Corrosion and Slag - Many industrial waste streams contain in-
    organic impurities which can cause corrosion and slag formation in
    the reactivation furnace.  These impurities are mostly chloride
    and sulfur salts of calcium and sodium.  The multiple hearth furn-
    ace has more exposed alloy parts than a kiln and is, therefore,
    more susceptible to corrosion.  Rabble teeth and arms are expen-
    sive, long delivery castings as opposed to the alloy flights in a
    kiln which are fabricated from readily available plates.  Also,
    considerable corrosion of flights can occur in a kiln before re-
    placement is required.

    Slag buildup in a multiple hearth furnace will require periodic


                               428

-------
    shutdowns to remove accumulated material.  Slag in a rotary
    kiln will be discharged, with the reactivated carbon, into the
    quench tank where it can be removed without shutting down the
    process.

7-   Maintenance - Experience with reactivating industrial wastewater
    carbons indicates higher maintenance costs in a multiple hearth
    furnace.  The factors responsible are:

         a.  Corrosion and slag formation resulting in shutdowns for
             repairs.

         b.  Rabble teeth and arms are more expensive to replace than
             alloy flights used in a rotary kiln.

         c.  Multiple hearth furnaces are more difficult to work on.
             It takes more man-hours to rebuild a hearth than to re-
             place brick in a kiln.  Because of these factors, down-
             time to affect repairs is longer in the multiple hearth
             furnace.

         d.  More instrument components are required with a multiple
             hearth furnace.

8.   Effect of Feed Outages - The upper hearths in a multiple hearth
    furnace can be damaged from temperature cycling caused by inter-
    ruptions in furnace feed.  Periodic planned shutdowns can be con-
    ducted without hearth damage.

    Feed outages are usually not a major problem in a rotary kiln. The
    refractory is much less effected by temperature cycling in a kiln
    than the hearth refractory in a multiple hearth furnace.

9.   Operating Factors - Operating factors for kilns and multiple hearth
    furnaces are as follows:

              Rotary Kiln              85-95 Percent
              Multiple Hearth Furnace  75-90 Percent

    Multiple hearth furnaces must be shutdown more often to clean slag
    and replace rabble teeth.  When a furnace is down for repairs, the
    work requires more man-hours to complete than similar work on a
    rotary kiln.

    Based on the above parameters, the multiple hearth furnace offers
    the following advantages over a rotary kiln:

       • Better control of temperature and atmosphere

       • Lower fuel consumption

       • Greater capacity turndown
                               429

-------
            •  Less area required
            •  Lower carbon losses from carryover and attrition

         The rotary kiln advantages are:

            •  Less corrosion and slag formation
            •  Less downtime
            •  Lower maintenance costs and easier maintenance
            •  Less effect from feed outages
            •  Easier to operate
SUMMARY AND CONCLUSIONS


     Granular activated carbon has been demonstrated to be effective in
treatment of a wide variety of industrial wastewaters.  Both multiple hearth
furnace and rotary kilns can satisfactorily reactivate spent carbons used in
industrial wastewaters provided adequate consideration is given to selection
of materials, sizing of equipment, and operating philosophy.  Experience
gained over the last ten years indicates that corrosion, slagging, poor re-
activation quality, carbon losses and line erosion can all be minimized
through good design.  Although the same types of problems exist in in-
dustrial purification and municipal water and waste treatment applications
using granular activated carbon, they are magnified in industrial waste-
water applications where wastewater quality, and thus carbon exhaustion
rates, are more variable and substantially more corrosive.  However, our
experience with reactivating over 100,000,000 pounds of spent carbon for
more than 75 different industrial wastewater applications, indicates that a
high quality product can be produced on a reliable, economical basis.
                                    430

-------
BIOGRAPHIES


     Roger H. Zanitsch is Engineering Director
of the Calgon Environmental Systems Division
for Calgon Corporation.  He joined Calgon as a
Project Engineer  in 1969 and was  later named
Project Manager of  the Environmental  Engineer-
ing Department.   Zanitsch received his BS degree
in Civil Engineering from the University of Cin-
cinnati and  an MS degree in Environmental Engin-
eering from  the same school.  Zanitsch is a mem-
ber of the Water  Pollution Control Federation.
      Richard T. Lynch is a senior engineer in the
 Process Engineering Group of Calgon Corporation's
 Engineering Department'.  He has a B.S. degree in
 Chemical Engineering from the  University of Florida.
 He has been a project manager  for the design of
 several carbon adsorption reaction systems treating
 industrial waste streams.  He is a member of the
 American Institute of Chemical  Engineers and a
 registered professional  engineer in Florida.
                                      431

-------
                                   EXHIBIT 1
                      TYPICAL REACTIVATION SYSTEM
                               FLOW DIAGRAM
                                                   Exhaust
                                                    Gas
  Spent Carbon
 From Adsorbers
         Spent
         Carbon
         Storage
         5-10
         Days.
Blowcase
 React
Carbon
Storage
 5-10
 Days
                                                                      Reactivated Carbon
                                                                        To Adsorbers
                                      432

-------
                                  EXHIBIT 2

                   INSTALLED COSTS OF REACTIVATION SYSTEMS

                                CAPITAL COSTS
Capital Cost Estimate
($ Million)

   Purchased Equipment
   Installation*
                                     Reactivation Rate - 1,000 Lbs/Day
                                5,000
            10.000
            30.000
           60.000
0.24
0.61

0.85
0.36
0.91

1.27
0.77
1.93

2.70
1.20
3.00
                                                                      4.20
*Installation costs include foundations, structural equipment setting,
 electrical,  instrumentation, site preparation, engineering contractor,
 overhead and profit, and indirects.
Fuel - 8,000 BTU/Lb @ $3/106
BTU
Power @ 30/KWH
Steam - 1.0 Lb/Lb @ $4/1,000
Lbs
Labor @ Supervision
Makeup  Carbon - 7% @ 57c/Lb
Maintenance @ 8% Capital
General Plant Overhead

  Total Operating Cost
  ($l,000/Yr.)
                               OPERATING COSTS


                                     Reactivation Rate -  1,000 Lbs/Day

                                5,000         10.000       30.000     60.000
* -
45
10
10
110
70
70
in
90
20
15
110
145
100
50
265
50
45
110
435
215
110
525
80
90
110
875
335
200
345
  Operating Cost C/Lb Carbon    18.8
 530
              14.5
1,230
                                                            12.2
                                     2,215
                                      11.0
                                    433

-------
                                       EXHIBIT - 3
 10,000
      9
      8
      7
      6
o
o
o
H-
V)
0 1,000
o    9
u
       CAPITAL COST ESTIMATE
       FOR REACTIVATION  SYSTEMS
                                                                         I  I   I I  I
JS
(A
Z
O
I-
8
7
6f-
5 —
     A -mi
      3 -
     2 —
   too
                       I   I  I  I  I I
                                     I
I   I  L  I I  I
                                                                            I  I  I  I
               Z    34-56789        2
     1,000                       10,000
                CARBON  REACTIVATION RATE -
                                                  6789
                                                       100,000
                              S  6 7 B 9
                                        LBS/DAY

-------
                                EXHIBIT - 4
        DIRECT OPERATING  COST
        FOR REACTIVATION  SYSTEMS
I.OOO
        5  6789
              10,000
CARBON  REACTIVATION RATE - LBS/OAY
56789
      100,000
                                                                          5  6 7 8 9

-------
                      EXHIBIT 5
            •Carbon- In
      Hearth
 i

61
   Diameter
                           uu
                                         Rabble Arm
                                         Rabble Teeth
                                         Two Burners and
                                         Steam Inlets at
                                         Hearth 4, 5 and 6,
                                         not shown
                                         Carbon Out
FIGURE 5 - CROSS  SECTIONAL VIEW OF MULTIPLE HEARTH FURNACE
          USED AT POMONA WATER RECLAMATION PLANT
                          436

-------
EXHIBIT 6
c
Dewatered
^
Carbon
Feed
Screw
V
J
Feed
Hood
^^^
—
_

J
_
Gas Discharge
S~~ Duct
_.. Feed End
|K" Seal |
./-Flights /-Refracto
1 _£ - X
\i/iyifiM77*rH*i [til mi unit
7—*' Kiln Shell-^ '
^
ry
k
/
c
1
X,
/•
!•!*
2
•*
Discharge
End Seal
>^-Firing Hood
4-*- Steam
y
p t Fuel
nt
X^Nose
Ring
s^_ Discharge
Chute
   437

-------
DISCUSSION

L. L. Krohn. Union Oil Co.;  Have you made an analysis  of  the  off-gases
from that furnace?

Roger Zanitsch;  Yes, we have.  We've looked at in excess  of 100 different
industrial waste carbons.  We've analyzed the off-gases before and after
afterburning on many of these, to determine what temperature and residence
time is needed for organic destruction.

L. L. Krohn. Union Oil Co.:  Do you have any feel for the  particulate matter
coming out?

Roger Zanitsch;  We have systems operating that are designed with a 2 second
residence time which is primarily there to consume the  particulates.  If
you have a half second residence time which is certainly sufficient to des-
troy most of the organics present, you'll still have some  carbon fines that
will need to be scrubbed.  We feel that the 1-2 second  residence time at
1800eF can destroy the carbon fines as well as the organics.


L. L. Krohn, Union Oil Co.:  Considering the new source review - can we hope
to build this system?

Roger Zanitsch;  Reactivation furnaces are now in operation and many are
being designed for industrial waste applications.  Technology  exists to
handle essentially all air pollution control requirements  at a reasonable
cost.
                 «&
Mac McGinnis. Shirco_t Inc.:  We have made some of the economics that you're
talking about for our electric regeneration furnace and compared them with
similar economics as you have presented here from multiple hearth and other
approaches and just a couple of comments - a couple of  factors that we have
included that you haven't mentioned are in the area of  utilities, scrubber
water which on small capacity units may be a fairly significant contribution
of operating costs; and the other factor youfmentioned  quality of the pro-
duct, laboratory labor, lab time to confirm that the product is indeed of
the desired quality, can be a fairly significant contribution.


Roger Zanitsch;  I'm glad you brought this up.  In the  analysis that I show-
ed, the operating cost included a 10% general plant service allowance on
the total operating cost to cover overhead items such as accounting, qual-
ity control, etc.  As far as scrubber water cost and disposal, it can be a
factor.  In those installations where we have scrubbers, we've recycled
water through the pretreatment system to remove the carbon fines.  Frankly,
we haven't found this to be a significant cost factor.
                                    438

-------
DISCUSSION


Mac McGinnis.  Shirco. .Inc.:  Well,  a half cent here and half  cent  there  it
begins to add  up.  The other general comment is you've indicated that there
is considerable data on regeneration costs in multiple hearth furnaces in
particular and you've showed us  some trend lines in terms of  direct  operat-
ing costs.  Can you comment on any  specific data, you know, accumulated over
a period of time that indicates  an  actual cost figure for some specific ap-
plication?


Roger Zanitsch;  The numbers which  I presented are based on our experience
in operating both small and large furnaces.


Mac McGinnis,  Shirco, Inc.:  One last comment - would you say then that the
actual data would fall within that  plus  or minus 20% about your nominal
curve?


Roger Zanitsch;  On industrial waste applications, yes.  In process  applic-
ations, such as the decolorization  of sugar solutions, operating costs are
substantially lower since they have a constant feed and a very predictable
product.


Colin Grieves, Amoco Oil Co.;  First, would you care to comment on some of
the new technology which you eluded to?   And second, would you like to say
anything about regeneration of powdered  activated carbon?


Roger Zanitsch;  As far as the new  technologies are concerned, I was person-
ally thinking of the electric furnace and the fluidized bed furnaces.  The
Japanese have several different  types of furnaces.  Most of the experience
with the newer furnaces has been in either pilot-scale or on  the commercial
scale, but in considerably less  corrosive application than you have in in-
dustrial wastes.  In industrial  waste applications, the big awakening has
been in the areas of corrosion,  maintenance costs, and feed interruptions.
The new technologies have not been  demonstrated in this type  of service.
As  the new technology develops,  it's going to take some time  to gain the
 experience necessary to apply these new  furnaces in the industrial waste
 effort.  As far as powdered carbon  activation, I don't really feel qualified
 to  discuss it on the basis that  I would  only be expressing my opinions
 since no commercial experience has  been  developed.
                                     439

-------
                  "ACTIVATED SLUDGE ENHANCEMENT:  A VIABLE
                 ALTERNATIVE TO TERTIARY CARBON ADSORPTION"
                               Leonard W.  Crame

                    Senior Chemical Engineer, Texaco Inc.
INTRODUCTION
          In view of the possibility of more stringent 1983 BATEA (Best
Available Technology Economically Achievable) effluent guidelines,1>2»3.4
petroleum refiners are faced with the dilemma of an insufficient data base
to determine the proper approach for making cost-effective improvements.
The EPA previously proposed granular activated carbon adsorption after acti-
vated sludge treatment as BATEA technology; however, the current emphasis is
to consider both effluent quality and the cost effectiveness of attaining the
desired results.  Two proposed approaches to BATEA technology are (1) in-
creasing the sludge age (or mean cell residence time) of the activated sludge
biomass to develop a more diverse population capable of assimilating biore-
fractory organics or (2) adding powdered carbon directly to activated sludge
aeration basins.  Both alternatives to tertiary carbon adsorption would re-
quire little capital investment and would lower operating costs.

          Grutsch and Mallatt5'6'7'8'9'10'11 have proposed that the best
refinery end-of-pipe treatment for soluble organic removal should include pH
control, equalization, optimized dissolved air flotation (DAF), and high
sludge age (20-50 days) activated sludge treatment.  High sludge ages (SA)
require mixed liquor solids levels above conventional levels (5-10 days SA).
These higher levels increase solids flux and must be considered in secondary
clarifier solids loadings.  Also high effluent TSS, despite less frequent
sludge wasting, can result in a loss of mixed liquor solids.

          Grutsch and Mallatt emphasize that optimized chemically-assisted
DAF pretreatment (or comparable pretreatment) reduces the colloid charge
(zeta potential) to maximize particle agglomeration for efficient flotation,
and reduces the organic load on the activated sludge unit (ASU).  Removing
colloids normally present in raw refinery wastewater allows better biofloc-
culation and lower effluent total suspended solids  (TSS) since most refinery
colloids and biosolids have repelling negative charges.  The microbial popu-
lation could then acclimate to the biorefractory organics by producing
enzymes which reduce these to simpler biodegradable substrates.  Current
reports from within the petroleum industry seem to indicate some benefits
for increasing SA.  Other investigators^ have reported that high SA  (low
food/microorganism ratio) produces poor sludge settleability.

                                     440

-------
         As  a  result of pilot studies at  the Du Pont  Chambers Works, Button
and Robertaccio-U were issued a U.S. patent!* for  the  Du  Pont PACT process 15
The PACT process basically involves the addition of  powdered carbon  (or
fuller's earth,  etc.) to an ASU, usually in  a range  of 50-400 mg/1 based on
influent flow.   Du Pont has reported 16, 17, 18 a  number  of  advantages  of the
PACT process  which include:

          (1) color removal,
          (2) stability against shock loadings,
          (3) improved BOD removal,
          (4) improved refractory organic  removal,
          (5) resistance to toxic substances,
          (6) improvement in hydraulic capacity,
          (7) improved nitrification  (mainly in municipal wastes) ,
          (8) foam suppression, and
          (9) improved sludge settling and increased clarifier capacity.

A disadvantage of the PACT process is that the  system  can become  very expen-
sive if powdered carbon addition rates become high (hundreds of mg/1) , even
though powdered carbon is cheaper than granular carbon.

          DeJohn and Adamsl9»20,21 have developed  a  considerable  amount of
pilot study data on activated sludge-powdered carbon systems.  They  report
significant enhancement in studies involving refinery  and petrochemical
wastewaters.   DeJohn and Adams explain the powdered  carbon  enhancement
mechanism as localization and concentration  of  oxygen  and pollutant  as the
result of  adsorption on carbon surfaces, resulting in  a more complete bio-
oxidation.   The adsorption of biorefractory  organics allows a longer resi-
dence time for these components in the system.  Other  researchers", 23 have
found similar improvements using activated sludge-powdered  carbon systems
and propose analogous enhancement mechanisms.
                  has reported a case history of  a full-scale  activated
 sludge-powdered carbon demonstration run  at  the Corpus  Christi, Texas, Sun
 Oil refinery.  Results included better  system stability,  reduction of foaming,
 resistance to upset conditions, lower effluent suspended  solids and clearer
 effluent, and improved organic removal.   These improvements were achieved
 by maintaining only a 450-mg/l powdered carbon reactor  concentration with a
 10rttg/l powdered carbon dosing requirement.   The  shortcoming of this inves-
 tigation was that a parallel control could not be run simultaneously and most
 improvements reported could possibly have been attributed to better
 clarification.

          The merits of powdered carbon enhancement have  been  further
 confused with the more recent development of several types of  powdered
 carbons with significantly different properties.
                                      441

-------
SCOPE OF WORK

1.  Objective

          The objective of this study was to determine if the relatively
simple process changes of increased sludge age or  the  addition of powdered
activated carbon in conventional refinery activated  sludge systems can sig-
nificantly enhance the removal of organic wastewater contaminants to achieve
or approach the level of proposed BATEA  (1983) technology more cost effec-
tively than the addition of granular activated carbon  contactors to BPCTCA
(1977) technology.

2.  Procedures

          In Part I of this study, five completely-mixed  (15  gal)  ASU's  were
operated in parallel with identical 18-hr retention  times and 300-gpd/sq ft
clarifier rise rates.  A sixth ASU was run as a second-stage  unit with the
same 18-hr retention time (Figure 1).  All biological  reactors were located
in a temperature controlled room in an attempt to  dampen  influent wastewater
temperature variations and control biological reactions at about 85 F.

          ASU's A and F served as controls, simulating conventional refinery
units with a 0.3 Ib TOC/lb MLVSS-day organic loading.  Separate controls were
run to determine the effect of optimized pretreatment  on  activated sludge
treatment and tertiary carbon adsorption.  Equalized (24-hr)«*and pH-control-
led refinery wastewater was pretreated by dual-media (sand-anthracite)  fil-
tration  (4.6 gpm/sq ft) and a chemically assisted  DAF  unit (1.5 gpm/sq  ft)
prior to control ASU's A and F, respectively.  The optimized  DAF pretreatment
neutralized the negatively charged colloids, thus  facilitating their removal
and producing a bio-unit feed that contained essentially  only soluble
organics.  Sodium phosphate (monobasic) was added  to the  filter and DAF  unit
effluents for a minimum TOG:phosphorus ratio of 100:2  to  assure a proper
nutrient level.  Effluents from ASU's A and F were continuously filtered
through a dual-media (sand-anthracite) tertiary filter for TSS removal before
passing through a series of four granular activated  carbon contactors to
simulate proposed 1983 BATEA technology.

          ASU's B, C, and D treated optimized DAF  effluent with sludge wast-
ing calculated for a 50-day biological SA.  A commercially available,  conven-
tional-surf ace-area, powdered carbon25 (designated PC-C)  was  added daily to
ASU C to maintain a 500-mg/l reactor operating level.  Similarly,  PC-H,  a
high-surface-area powdered carbon,26 was added to  ASU  D to investigate  its
enhancement capabilities.

          ASU E was also operated at a 50-day SA treating ASU B effluent to
determine if there was any benefit to ASU staging.

          In Part II of this study, the second-stage ASU  E was placed in
parallel with other ASU's treating the DAF unit effluent  as shown in
Figure 1 and redesignated as ASU G.  PC-H was added  to ASU G  to maintain
a 2500-mg/l operating level while powdered carbon  levels  were increased in
ASU C and D to 1000 mg/1.


                                     442

-------
         Effluents were collected daily, Monday through Friday, as grab
samples.  Grab samples  were taken in lieu of composites for convenience since
the pilot unit treatment scheme contained significant equalization.  Samples
from  the equalization basin, biological reactors, and carbon columns were
filtered with Gelman  type A/E glass fiber filters to give the soluble con-
taminant (TOG, COD, S=) level.  Glass fiber filters were used instead of
0.45-micron filters because the solids retained on glass fiber filters
define  the TSS measurement.  Samples were analyzed immediately after collec-
tion  or were preserved  until analyzed using accepted preservation methods.27

         All effluent  data were compared after plotting values on log-normal
probability papers.   Single straight line data fits were determined by calcu-
lating  50th and  90th  percentile values.  The 50th percentile values equaled
the antilog of the mean of the log values of data sets.  The 90th percentile
values  were calculated  assuming a single-tailed log-normal data distribution
                                                   -f~.

                        In N90 = In NSO + 1.282 In 84

where
          S(j  is  the  standard deviation.

Engineering judgment  was used to determine which data sets being compared
appeared different and  required additional statistical analyses to confirm
significance  of  median differences.  Median data values were compared to
determine  if  they were  from the same population using a paired t-test2°
assuming a log-normal distribution as follows:
                       compute t =    ,               -=
                                      /£d2 -  (Id)2/n
                                     V     n(n-l)

                                        n
              3"= mean of differences =]Tdi/n where d^ = In x^-ln y±
                                        1

              for i - l,2,3,.,.n of  n data pairs

              S
-------
H20) and 20-40 mg/1 Dearborn 431 cationic polymer.  Chemical  dosages  were
initially optimized using zeta potential titrations in  conjunction with  jar
tests; however, only jar tests were continued since zeta potential calcula-
tions became tedious and inconsistent due to the high wastewater  specific
conductivity (usually 4000-8000 micromhos/cm).  This salinity was primarily
due to the brackish intake waters to the refinery and may  have  impeded
chemical coagulation at lower chemical dosages as reported by Grutsch and
Mallatt.8

          The superiority of the optimized DAF unit (operating  at 1.5 gpm/sq
ft) as a pretreatment system over simple sand filtration is clearly evident
in the three weeks of run data represented in Figure 2.

          DAF unit effluent 50th percentile TOG (158 mg/1) was  55 percent
less  than the 50th percentile TOG (352 mg/1) in the equalization  basin
influent to the DAF unit, while the sand filter gave only  a 18-percent reduc-
tion.  Since, at best, only a 31-percent TOG reduction  could  be achieved by
vacuum filtration of equalization basin samples with glass fiber  filters
(which define TSS), the true effectiveness (55 percent  reduction)  of  colloid
and oil coagulation and removal in the chemically assisted DAF  unit can be
seen.  The DAF unit effluent contained essentially only soluble organic
contaminants.

          The continuous dual-media filter (operating at 4.6  gpm/sq ft) could
only  manage a TOG reduction of about one-third of the DAF  unit.   There was
no indication that a shorter run time would improve the filter  effluent
significantly.  It appeared that due to the nature of the  solids,  chemical
addition to the filter feed would have been required for an improved  system.
The purpose of the filter, however, was to produce a biological reactor feed
with  characteristics comparable to DAF treatment without chemicals.   It was
observed, on occasions, that DAF pretreatment was very  poor when  chemical
feed  pumps failed.

          Figure 2 illustrates COD removal by both pretreatment systems
employed and again demonstrates the effectiveness of optimized  DAF treatment.
As in subsequent graphs, the data points are not shown  to  avoid congestion of
data.

          Fiftieth percentile oil and grease values during Part I of  this
study were 101, 70, and 16 mg/1 for the equalization basin, filter effluent
and DAF unit effluent, respectively.  The equalization  basin  50th percentile
TSS level of 78.0 mg/1 was reduced to 57.5 mg/1 by the  filter and to  19.0
mg/1  by the DAF unit.  A portion of the TSS in the DAF  unit effluent  was due
to biological growth rather than influent solids.  It is possible that part
of the organics reduction through the DAF unit was the  result of  biological
activity which could also occur in full-scale systems.

2.  Part I - Activated Sludge Performance

          TOG and COD (filtered) effluent variability plots for pilot-scale
18-hr retention ASU's are compared in Figure 3.  These  results  are from  the
initial 3-week data run.


                                     444

-------
         Control ASU F produced a better effluent  than  control ASU A  both
operating at equal  0.3 Ib TOC/lb MLVSS-day  (F/M)  loadings.  MLVSS levels in
ASU A averaged 1,148  mg/1, about twice that  of ASU  F  due to a  twofold
increase in feed TOC.  Considering both TOG  and COD removal, ASU B, C, E,
and F did not show  any significant overall difference in performance.'
Fiftieth percentile TOC values ranged from 53-58  mg/1 while COD values were
97-116 mg/1 as shown  in Figure 3.  Sludge age was not a  controlling perfor-
mance variable as ASU B (50-day SA) and ASU  F  (about  10-day SA) differed
greatly in solids retention time with average MLSS  levels of 1,621 mg/1 and
816 mg/1, respectively.  Chemically assisted pretreatment for  removal of
colloids and oil had  the most significant effect  on organics removal.  The
high-surface-area powdered carbon (designated PC-H) significantly enhanced
organic removal in  ASU D, with a 50-day SA and a  500-mg/l PC-H operating
level.  Enhancement was not evident in ASU C containing  the conventional-
surface-area powdered activated carbon  (designated  PC-C). Powdered carbon
addition increased  the average ASU C MLSS level to  1,885 mg/1  with ASU D
averaging 1,976 mg/1.

         Since a marginal enhancement occurred with  the addition of PC-H at
a  500-mg/l  level,  the scope of this investigation was expanded to evaluate
powdered activated  carbon addition at a 1000-mg/l level  and only PC-H at
approximately  2500  mg/1.  This would give a  greater overview of the enhance-
ment capabilities  of  powdered carbon, especially  the  highly active PC-H.  ASU
E  was taken out of  service since it was only succeeding  in lysing biological
cells as a  second  stage following ASU B.  The  reactor was placed in parallel
with other  units being fed by the DAF unit and redesignated ASU G.  The SA
was maintained at  50 days and PC-H was built up to  a  reactor level of about
2500 mg/1 for  Part  II of this study.

3.  Part  I  - Granular Carbon Adsorption

         Granular  carbon Series A, treating ASU  A  effluent, exhausted two
130-gram carbon beds during 17 days of a 3.4-gpm/sq ft hydraulic loading in
Part I.  A  20-mg/l  soluble (filtered) TOC and  a 44-mg/l  soluble COD effluent
 (50th percentile,  Figure 4) was produced with  0.10  and 0.09-g  TOC/g carbon
accumulative loadings at exhaustion.  Carbon Series F, treating ASU F efflu-
ent, reduced the  50th percentile effluent soluble TOC and COD  to 23 mg/1 and
40 mg/1,  respectively.  Because of the relatively few data points used to
establish Figure  4, there is little significance  in the  difference between
carbon series  A and F 50th percentile values.  A  single  carbon bed was ex-
hausted to  a 0.12-g TOC/g carbon loading.  TOC loadings  of carbon columns in
Series A were  comparable to an average of 0.11 g  TOC/g carbon  reported for
the granular carbon during the four previous exhaustions prior to each
regeneration.

         Granular  carbon effluents were of  substantially better quality than
all biological unit effluents.  The 50th percentile soluble TOC and COD re-
ductions  in carbon Series A were 44 mg/1 arid 84 mg/1, respectively, whereas
carbon Series  F accounted for a 50th percentile 35-mg/l  soluble TOC and 65-
""8/1 soluble COD reduction.
                                      445

-------
4.  Part II - Pretreatment

          Pretreatment by filtration and chemically assisted  DAF treatment
continued as in Part I of this study.  Again, using dosages of  40-mg/l  filter
alum and 20-40 mg/1 Dearborn 431, optimized DAF pretreatment  reduced  the
equalization basin TOG and COD by more than 50 percent as  shown in  Figure 5.
Filtration could only remove gross quantities of oil and solids without pre-
liminary chemical coagulation.  Although equalization basin,  filter,  and DAF
unit effluent 50th percentile TOC and COD concentrations were approximately
equal to those experienced in Part I of this study, there  existed a greater
degree of variability in Part II.  A contributing variability factor  was the
rainfall dilution of refinery wastewater streams as an average  of 0.21  in./
day of rain fell during Part II compared with 0.06 in./day during Part  I.

5.  Part II - Activated Sludge Performance

          The effluent quality for Part II, basis filtered TOC  and  COD, is
given in effluent frequency distributions, Figure 6, for the  six-week run
period.  Control ASU A (F/M = 0.3), without optimized pretreatment, continued
producing the most inferior effluent and experienced three upsets due to the
development of a filamentous bulking sludge.  The unit was restarted  on each
upset occasion with new seed and allowed'to acclimate for  a few days  before
effluent data were used for comparison with parallel systems.   ASU  B, C, and
F, as in Part I, produced nearly equivalent effluents in terms  of filtered
TOC and COD with neither high SA (50 days) nor 1000 mg/1 PC-C enhancing bio-
logical treatment.  PC-H added to ASU D and G at levels of 1000 and 2500
mg/1, respectively, reduced TOC and COD substantially.  Compared with high SA
control ASU B, 50th percentile TOC was reduced an additional  10 mg/1  and 22
mg/1 in reactors D and G, respectively.  COD 50th percentile  reductions below
reactor B were 22 mg/1 for ASU D (1000 mg/1 PC-H) and 39 mg/1 for ASU G
(2500 mg/1 PC-H).  The ASU G run time was abbreviated, however,  due to  the
time required for acclimation at the higher PC-H level.  As in  Part I of this
study, it was observed that as powdered carbon levels were suddenly increased
in ASU C, D, and G, performance was exceptionally good for a  short period of
time.

          Phenols feed concentrations were higher in Part  II  of this  investi-
gation as 90th percentile values reached 18 mg/1, compared with 8.5 mg/1 in
Part I.  ASU D (see Table 1) and G, containing PC-H, provided the best
phenols removal with 50th percentile phenols levels of 0.05 mg/1 and  0.04
mg/1, respectively.  This was slightly lower than the high SA control ASU B
(0.06 mg/1) and low SA control ASU F (0.07 mg/1).  Although the lack  of opti-
mized pretreatment produced higher 50th percentile phenols levels (0.11 mg/1)
in ASU A, even poorer reductions were experienced with ASU C  as in  Part I.
Similar results were obtained in Part I.  An occasional high  phenols  value
was measured in the effluents of ASU D and G but not with  the consistency
or magnitude of ASU C.

          Effluent oil and grease values,  included in Table 1,'illustrate the
significance of removing most of the oil and grease before biological treat-
ment.  The 18 mg/1 50th percentile oil and grease effluent level of ASU A
greatly exceeded the concentration of 5 mg/1, or less, discharged from

                                     446

-------
reactors receiving optimized pretreatment.   The addition of  PC-H  to ASU D and
G gave  slight  oil and grease improvement with SA alone not being  an
enhancement  factor.

          Effluent TSS levels were high in ASU A at a 50th percentile level
of 86 mg/1.  TSS increased to more than 150 mg/1 when filamentous sludge
bulking occurred.  However, ASU's with  DAF pretreatment produced  a more
settleable sludge.  The high SA control ASU B had a significantly higher
effluent TSS level than the lower SA control ASU F, but the  high  SA could
be maintained.   ASU C and D experienced effluent TSS levels  less  than control
ASU B despite  higher reactor solids  due to powdered carbon.   No effluent TSS
increase was observed in ASU G due to the  higher (2500 mg/1)  PC-H level.

          Ammonia nitrogen removal in system A was less than the  80 percent
achieved in the systems with optimized  pretreatment.   The organic loading was
higher  and the sludge age was less than in other systems.  The factors con-
trolling the degree of nitrogen removal were not investigated.  Nitrification
during  Part II was not as complete as that obtained in Part  I.

          Sludge Characteristics and production rates are summarized in
Table 2 for all ASU's.  As expected,  ASU A had the highest measured oxygen
uptake  averaging 0.16 mg oxygen/1-min due  to a higher influent organic con-
centration.  Oxygen consumption averaged 0.10-0.12 mg oxygen/1-min in other
ASU mixed  liquors, but a relationship of increased oxygen demand  and enhanced
biological treatment did not exist.   The sludge volume index (SVI), a measure
of sludge  compactability, significantly improved with SA and  powdered carbon
addition.   Sludge settling velocities were exceptionally high with the worst
rate (ASU  A) being 0.17 ft/min corresponding to a 1830-gpd/sq ft  clarifier
rise rate.  Other mixed liquors settled with zone settling velocities of
0.34-0.39  ft/min.  The average MLSS  concentration of  745 mg/1 in  ASU F was
too low for zone settling to occur.   One of the most  surprising results of
powdered carbon addition was that less  biomass was produced  than  in control
systems.   ASU  G produced an average  of  0.08 Ib biomass/lb COD removed com-
pared with control rates of 0.22 for ASU A and 0.19 for ASU  F.  PC-H was more
effective  than PC-C at reducing biomass production rates at  the same SA.  The
total sludge production of activated sludge powdered  carbon  systems was not
much higher than controls, due to lower biological sludge production rates.

          Powdered carbon inventories and  makeup requirements for ASU's are
summarized in  Table 3 for Parts I and II of this study.   PC-C losses were
slightly higher than PC-H but still  reasonably close  to 2 percent per day.
Since biological sludge was wasted at a rate of 2 percent per day in high SA
reactors,  it is a fairly good assumption that powdered carbon lost In efflu-
ents was in the same proportion to biological sludge  as in the mixed liquor.
Thus both  biological and powdered carbon SA may be assumed to be  equal for
simplification of powdered carbon daily makeup requirements.  The powdered
carbons must be wetted to prevent loss  of  floating carbon in  the  clarifier.
This was accomplished by boiling the carbon slurry in this study.  Vacuum
degassing  could also be used.

          Another observation made during  Part II was that activated sludge-^
powdered carbon systems significantly reduced aeration basin foamir.0	.c	


                                     447

-------
with control systems.  Foaming in ASU aeration basins  was not a problem but
did occur occasionally.

6.  Part II - Granular Carbon Adsorption

          A single granular carbon bed was exhausted from carbon Series A
with an accumulative organic loading of 0.15 g TOC/g carbon during Part II.
The data include three short runs.  The first two carbon  beds required  back-
washing almost daily due to high TSS levels which could not be continuously
removed by dual-media filtration.  Fiftieth percentile effluent soluble TOC
was 30 mg/1 (see Figure 7) for a reduction of 38 mg/1  from ASU A.   ASU  A 50th
percentile soluble COD was reduced by 84 mg/1 to 79 mg/1.   Although phenols
levels were generally low (50th percentile of 0.04 mg/1)  a few very high
effluent phenols levels were detected in carbon Series A  giving a  90th
percentile phenols value of 4.8 mg/1 (Table 1).  Phenols  must have been
adsorbed, concentrated, and then eluted in slugs from  the carbon beds to
achieve such a high level.  Effluent oil and grease levels remained low with
50th percentile values less than 3 mg/1.

          Carbon Series F exhausted a single carbon bed to an accumulative
organic loading of 0.13 g TOC/g carbon while surpassing the performance of
carbon Series A.  The 50th percentile soluble TOC was  significantly lower at
18 mg/1 for a 28 mg/1 reduction (Figure 7) from ASU F.  Fiftieth percentile
soluble effluent COD was 64 mg/1 for a 44 mg/1 reduction.   Phenols levels
were extremely low at 0.02 mg/1 (50th percentile) and  no  sudden loss of
adsorbed phenols was detected during most of the Part  II  data run  (Table 1).
Oil and grease effluent levels (50th percentile) were  again less than 3 mg/1.

          The lower dashed lines in Figure 7 represent the performance  of ASU
G, the best of the activated sludge-powdered carbon reactors.   ASU G produced
an effluent superior to carbon Series A and approached the quality of carbon
Series F.  The powdered carbon enhancement removed about  85 percent of  the
soluble TOC adsorbed on carbon Series F and about 60 percent of the COD based
on 50th percentile effluent values.  The 2,500 mg/1 PC-H  operating level in
ASU G significantly reduced effluent color to a level  comparable with
granular carbon effluent color.

7.  Economics

          Although unequal in overall performance, a high SA activated
sludge-powdered carbon system (ASU G, 72 mg/1 COD) approached the  level of
granular carbon adsorption (carbon Series F, 64 mg/1 COD)  to within 8 mg/1
COD at the 50th percentile point.  Both systems would  require extensive pre-
treatment and tertiary suspended solids removal.  All  other process compon-
ents being essentially equal, daily carbon usage costs were estimated for
theoretical plant flows of 1-5 MM gpd.

          The cost of virgin powdered carbon (PC-H or  PC-C) was estimated at
$0.30/lb and it was assumed that wasted carbon would be thrown away. To
calculate the equivalent powdered carbon dosage required  for an 18-hr reten-
tion ASU it was assumed a 50-day SA would be maintained,  giving an average
2 percent powdered carbon makeup.  This was the equivalent of a 37.1 mg/1

                                     448

-------
powdered carbon addition rate based on influent  flow.   A powdered carbon
feeder and storage  facilities were included  in powdered carbon costs using
Du Pont economicsJ-o and applying the 0.6 rule.   It  was  assumed that the
powdered carbon feeder could handle a 50 mg/1 addition  rate.

         Regenerated granular carbon adsorption costs  were estimated, using
Brown and Root, Inc.  economics,29 and converted  to  1977 dollars.  Daily gran-
ular carbon  costs were estimated using 17.2  percent of  the  fixed investment
for operating and maintenance cost and 17.7  percent for depreciation.  The
total daily  costs for powdered carbon were estimated using  the same per-
centage allowances.

         Daily estimated carbon costs are shown in Figure  8 for theoretical
flows of 1-5 MM gpd by scaling up ASU G and  carbon  Series F carbon require-
ments.  The  cost effectiveness of the relatively simple process change of
adding powdered activated carbon to the activated sludge process can be
clearly seen.  Estimated daily cost savings  would range from $987/day at 1 MM
gpd flow to  $2750/day at 5 MM gpd using high-surface-area powdered carbon
(PC-H) addition rather than granular carbon  adsorption.  The incremental cost
would be about $14.73 per pound of COD at 1  MM pgd  (see Figure 9).

DISCUSSION

1.  Increasing Sludge Age (SA)

         Contrary  to conventional activated sludge design  techniques, the
increased SA did not result in sludge deflocculation, higher SVI, and high
effluent TSS.  With the exception of a few days, the high SA control ASU B
easily  achieved a high SA as a result of good pretreatment  as proposed by
Grutsch and  Mallatt.10,11  However, no enhanced  performance was measured at
the high  SA  nor  in  the two-stage system with both reactors  at a high SA in
Part I.  Possibly more emphasis should be placed on the benefits of optimized
pretreatment than on increased SA as parallel activated sludge systems at
about the same SA   (F/M = 0.3) were vastly different in  performance at
bontrasting  degrees of pretreatment.

         Increasing SA gave a reduction in  biological  solids production as
the conventional F/M control ASU F produced  0.19 Ib VSS/lb  COD removed com-
pared with 0.16 for high SA control ASU B.   This sludge production was not
quite as  significant as it would have been if ASU F had operated at 5-10 days
SA where many conventional ASU's operate instead of about 14 days.  Any re-
duction of biological solids production would help  lower sludge treatment and
disposal costs.

         The high  effluent 50th percentile  ammonia level of ASU A  (11.5
mg/1) in Part II was probably the result of  upset conditions which resulted
from filamentous sludge bulking causing the  loss of biomass. The calculated
10-day  SA of ASU A  was only slightly less than ASU  F (14 days) which Produced
a 50th percentile ammonia level of 2.3 mg/1  in Part II.  ^hougV    ** to
SA generally does  improve nitrification, no  conclusions could be drawn as to
its effect in this  study.  The conventional  ASU  F had already produced an
effluent that was about 90 percent nitrified.

                                     449

-------
2.  Activated Carbon Enhancement Mechanism

          A 0.59-g TOC/g carbon loading was calculated  in ASU G while  operat-
ing with a 2500 mg/1 PC-H level.  This extremely high "apparent" TOG loading,
as explained by Flynn,16 may be the result of continuous  adsorption of slowly
biodegraded organics which are "biologically regenerated" from the  carbon
many times over the biological and carbon SA.  "Apparent" TOG loadings there-
fore increase with higher SA, optimizing the use of powdered  carbon, until
the carbon becomes loaded with completely biorefractory organics.   This
explanation of an "apparent" loading or enhancement mechanism appears  logi-
cal; however, oxygen uptake and biological sludge production  data presented
here negate biological regeneration in this study.  ASU G not only  had com-
parable oxygen uptake measurements with control ASU B,  but it produced about
50 percent less biological solids.  This implies that the actual enhancement
may have been predominantly due to adsorption on the high PC-H surface area.
Considering that PC-H had approximately five times the  surface area of con-
ventional powdered carbons, such as PC-C, the expression  of the TOG loading
as 0.12 g TOG/500 sq m of surface area would be more reasonable.

          DeJohn-'-^ > 30 explains that granular carbon columns are sometimes
undersized because the designer uses virgin carbon and  assumes that regen-
erated carbon will have the same activity.  The thermal regeneration process
will enlarge some carbon pores reducing the surface area  and  decreasing the
adsorption of small molecules which are not so strongly adsorbed on larger
pores.  Assuming that many small molecules require small  powdered carbon
pores for moderately strong adsorption, PC-H may have been more effective
than PC-C because of pore size distribution, provided that the normally
biorefractory refinery organics were small molecules.

          The mechanism of powdered carbon enhancement  of the activaged
sludge process was not defined in this study and needs  further investigation
in Phase II.  Target SA's of the activated sludge-powdered carbon systems
were 50 days.  Ideally, systems should be operated for  periods of several
SA's to insure that equilibrium conditions have been reached  and that  the low
(2 percent) daily powdered carbon makeup rate will continue to give consis-
tent results.

          The selection of the best powdered carbon for a particular acti-
vated sludge enhancement is not a simple task since powdered  carbons vary in
their adsorptivity.   Carbon isotherms performed on a refinery wastewater
would exhibit a wide variability and require a statistical analysis to select
the best powdered carbon.  Isotherms would have to be performed on  the acti-
vated sludge effluent (as in Phase II of this study) to determine enhancement
strictly due to adsorption.

          The powdered activated carbon (PC-H) utilized with  very good
enhancement results is not, as of yet, commercially available.   Because of
the relatively high cost of granular carbon adsorption, other powdered
carbons at similar and higher operating levels would probably offer a  signi-
ficant improvement in activated sludge performance and  remain more  cost
effective than granular carbon.
                                     450

-------
3.  Granular Carbon Adsorption

         Granular  carbon adsorption data indicate  that  the quality of end-
of-pipe refinery wastewater treatment depends on optimization of each treat-
ment  step from primary to tertiary treatment.  The  use of  equilibrium or re-
generated granular  carbon in pilot studies will provide  a  more realistic data
base, recognizing that economics would favor regeneration  for many potential
users.

         The classical approach31'32'33'34 for handling carbon adsorption
data  to establish breakthrough curves was virtually useless in this study
because it assumes  the carbon column influent has a single adsorbate.  In the
calculation of accumulated TOG loadings at apparent breakthrough of carbon
columns there were  several instances where organics were eluted in slugs from
carbons in both series.  At times, phenols were two orders of magnitude high-
er than normal.  This phenomenon is a very real problem  and must be consid-
ered  when establishing stringent effluent discharge guidelines for industry.
Even  the best available technology, disregarding economics, has its
limitations.

SUMMARY

         The EPA 1983 guidelines for the petroleum refining industry have
assumed that 1977-type technology must be upgraded  by the  addition of costly
systems, such as granular activated carbon adsorption.   The results of this
API study indicate  that, should the EPA adhere to the granular carbon techno-
logy  originally proposed, it may be possible to achieve  this level of treat-
ment  technology by  the much more cost-effective method of  adding powdered
activated carbon to the 1977 activated sludge system.

         Process modifications including optimized pretreatment and the
addition of a high-surface-area powdered activated  carbon  can be used to pro-
duce  an effluent which is comparable in quality to  that  obtained by granular
carbon adsorption.   Increasing activated sludge age from the conventional
mode  of operation  (about 10 days) to about 50 days  did not give a significant
system improvement; however, in conjunction with powdered  carbon addition,
high  sludge age allowed higher equilibrium reactor  concentrations (2500 mg/1)
at low (2 percent)  carbon makeup rates.  This benefit has  been demonstrated
with  the high-surface-area carbon and it is possible that  it can also be
obtained with increased levels of conventional powdered  carbon.  The cost-
effectiveness of any powdered carbon will depend on the  wastewater charac-
teristics and powdered carbon adsorptivity, which was greater for the high-
surface-area carbon (2462 sq m/g) than for the conventional-surface-area
carbon (550 sq m/g) investigated here.  Even granular carbon adsorption was
found to have limitations as slugs of phenols were  eluted, on occasion,
into  the effluent.

ACKNOWLED6B86STS"

         This study was funded in part by the American  Petroleum Institute,
Divison of Refining, CREC Liquid Waste Subcommittee.


                                     451

-------
 REFERENCES

 1.   Environmental  Protection Agency,  "Petroleum Refining Point Source
     Category Effluent Guidelines  and  Standards," Federal Register, Vol 38
     (240)  34542 (December  14, 1973).

 2.   Environmental  Protection Agency,  "Petroleum Refinery Point Source
     Category Effluent Guidelines  and  Standards," Federal Register, Vol 39
     (91)  16560 (May 9,  1975).

 3.   Environmental  Protection Agency,  "Petroleum Refining Point Source
     Category Effluent Guidelines  and  Standards," Federal Register, Vol 40
     (98)  21951 (May 20,  1975).

 4.   Jones Associates, "Effluent Limitations in the Petroleum Refining
     Industry," Vol IB,  Prepared for the Office of General Counsel, American
     Petroleum Institute (January, 1976)

 5.   J. F. Grutsch  and R. C.  Mallatt,  "Optimize the Effluent System - Part 1:
     Activated Sludge Process," Hydrocarbon Processing, Vol 55 (3) 105 (1976).

 6.   J. F. Grutsch  and R. C.  Mallatt,  "Optimize the Effluent System - Part 2:
     Intermediate Treatment," Hydrocarbon Processing, Vol 55 (4) 213 (1976).

 7.   J. F. Grutsch  and R. C.  Mallatt,  "Optimize the Effluent System - Part 3:
     Electrochemistry of Destabilization," Hydrocarbon Processing, Vol 55 (5)
     221 (1976).

 8.   J. F. Grutsch  and R. C.  Mallatt,  "Optimize the Effluent System - Part 4:
     Approach to Chemical Treatment,"  Hydrocarbon Processing, Vol 55 (6) 115
     (1976).

 9.   J. F. Grutsch  and R. C.  Mallatt,  "Optimize the Effluent System - Part 5:
     Multimedia Filters," Hydrocarbon  Processing. Vol 55 (7) 113 (1976).

10.   J. F. Grutsch  and R. C.  Mallatt,  "Optimize the Effluent System - Part 6:
     Biochemistry of Activated Sludge  Process," Hydrocarbon Processing,
     Vol 55 (8) 137 (1976).

11.   J. F. Grutsch  and R. C.  Mallatt,  "A New Perspective on the Role of the
     Activated Sludge Process and  Ancillary Facilities," Presented at Joint
     EPA-API-NPRA-TU Open Forum on Management of Petroleum Refinery Waste-
     waters, Tulsa, Oklahoma (January  26-29, 1976).
                                      452

-------
12.  D. L. Ford and W.  W. Eckenfelder, Jr., "Effect of Process Variables on
    Sludge  Floe Formation and Settling Characteristics," Journal Water
    Pollution Control Federation, Vol 39  (11) 1850 (1969)T~~	

13.  G. Grulich, D. G.  Button, F. L. Robertaccio, and H. L. Glotzer, "Treat-
    ment  of Organic Chemicals Plant Wastewater with the Du Pont PACT
    Process," Presented at AIChE National Meeting (February, 1972).

14.  D. G. Button and F. L. Robertaccio, U. S. Patent 3,904,518 (September 9,
15.  E.  I.  Du Pont DeNemours and Company, "Du Pont PACT Process," Technical
     Bulletin.

16.  B.  P.  Flynn and L. T. Barry, "Finding a Home for the Carbon:  Aerator
     (Powdered)  or Column (Granular)," Proceedings of the 31st Annual Purdue
     Waste Conference (May 5, 1976).

17.  B.  P.  Flynn "A Methodology for Comparing Powdered Activated Carbons for
     Activated Sludge," Presented at 168th National Meeting, ACS, Div. of
     Petroleum Chemistry, Symposium on Disposal of Wastes from Petroleum and
     Petrochemical Refineries (September 13, 1974).

18.  B.  P.  Flynn, F. L. Robertaccio, and L. T. Barry, "Truth or Consequences:
     Biological Fouling and Other Considerations in the Powdered Activated
     Carbon - Activated Sludge System," Presented at 31st Annual Purdue Waste
     Conference (May 5, 1976).

19.  P.  B.  DeJohn and A. D. Adams, "Treatment of Oil Refining Wastewaters
     with Granular and Powdered Activated Carbon," Proceedings of 30th Annual
     Purdue Industrial Waste Conference  (May 6, 1975).

20.  A.  D.  Adams, "Powdered Carbon:  Is It Really That Good?," Water and
     Wastes Engineering, Vol 11 (3) B-8  (1974).

21.  P.  B.  DeJohn and A. D. Adams, "Activated Carbon Improves Wastewater
     Treatment," Hydrocarbon Processing, Vol 54 (10) 104 (1975).

22.  A.  B.  Scaramelli and F. A. DiGiano, "Upgrading the Activated Sludge
     System by Addition of Powdered Carbon," Water and Sewage Works, Vol 120
     (9) 90 (1973).

23.  A.  E.  Perrotti and C. A. Rodman, "Enhancement of Biological Waste Treat-
     ment by Activated Carbon," Chemical Engineering Progress, Vol 69 (11)
     63  (1973).

24.  J.  A.  Rizzo, "Case History:  Use of Powdered Activated Carbon in an
     Activated Sludge System," Presented at Joint EPA-API-NPRA-TU Open Forum
     on  Management of Petroleum Refinery Wastewaters (January 26-29, 1976).

25.  ICI United States Inc., "Powdered Hydrodarco Activated Carbons Improve
     Activated Sludge Treatment," Product Bulletin PC-4 (October, 1974).

                                     453

-------
26.  Amoco Research Corporation, Amoco Active Carbon Grade PX-21 Product
     Information Sheet (May, 1976).

27.  APHA, Standard Methods for the Examination of Water and Wastewater.
     13th Ed.,  New York, New York (1971).

28.  H.  E. Klugh,  Statistics;  The Essentials for Research, 2nd Ed., John
     Wiley &  Sons, Inc., New York, New York  (1974).

29.  Brown and  Root, Inc., "Economics of Refinery Wastewater Treatment,"
     American Petroleum Institute Publication No.  4199 (1973).

30.  P.  B. DeJohn, "Carbon from Lignite or Coal:   Which is Better?,"
     Chemical Engineering, Vol 82 (9) 113 (1975).

31.  Metcalf  and Eddy, Inc., Wastewater Engineering,  McGraw-Hill, New York
     New York (1972).

32.  W.  W. Eckenfelder, Jr., Industrial Water Pollution Control, McGraw-
     Hill, New  York, New York (1966).

33.  H.  J. Fornwalt and R. A. Hutchins, "Purifying Liquids with Activated
     Carbon," Chemical Engineering,  Vol 73 (8)  1976 (1966).

34.  J.  L. Rizzo and A. R. Shepherd, "Treating  Industrial Wastewater With
     Activated  Carbon," Chemical Engineering. Vol  84  (1)  95 (1977).
 BIOGRAPHY        Leonard W. Crame

         Leonard W. Crame is a Senior Chemical
 Engineer  in the Air and Water Conservation
 Section of Texaco's Pror Arthur, Texas, Research
 Laboratories. He has a B.S. degree  in Engineer-
 ing Technology (Chemical) and an M.S. in Thermal
and Environmental Engineering from Southern
 Illinois University at Carbondale.  Len has been
 involved in  several refinery wastewater treat-
ment pilot studies and conceptual designs since
 Joining Texaco in 1973.  He is a member of
the Texas \Afater Pollution Control Association
and has recently authored several papers on
wastewater treatment.
                                     454

-------
DISCUSSION

E. A.  Buckley. Lion Oil Co.;  What were the levels of alum and  polyelectro-
lyte used in the DAF pretreatment to  the system?

Len Crame;  In the dissolved air flotation unit we were using,  we  found
by the use of the zeta meter and jar  tests that it required 40  milligrams
per liter of filter alum, 20 milligrams to 40 milligrams per liter of catonic
polymer Dearborn 431.  It was not the intent of this study to try  to zero in
on the best and most economical chemical dosage but mainly to get  the soluble
feed for the bio units.  We found from experience that wide fluctuations in
the feed characteristics did not affect these two chemical doses.

Ed Sebesta, Brown & Root;  In the slide (Figure 2) comparing effluent and
COD concentrations from the various pretreatment systems, were  the samples
filtered or unfiltered?

Len Crame;  They were filtered COD's  for our bio effluents and  carbon
effluents.  In Figure 2,  I did not identify them, but it is total COD.  It
does include suspended solids because we were looking for the contribution
of solids in this case.

          I also would like to make the comment that I do pretty much agree
with everyone else's presentations as far as what work has been done with
carbon on the enhancement mechanism and we will continue to look at this
throughout the second phase of our pilot study.  I think that you  have to be
very careful in using powdered activated carbon.  In a short term  study I
agree with the other gentlemen (Amoco) (DuPont), that when you first put in
activated carbon you have to allow time for this matrix to form, which we
did.  And you don't get the same settling effect as when you initially add
carbon.  When you allow the system to come to an equilibrium and the bio-mass
starts adhering to the powdered carbon, it does greatly improve the sludge
settling characteristics, but it takes a little time.  I believe that when
you initially add powdered carbon your results are going to be  very good
because you're going to get a tremendous amount of adsorption.   We followed
this and have seen it.  I'm very hesitant about including data  right after
you start running an enhanced bio-system.  You will see a sharp decrease in
the effluent organic levels.  You have got to wait until an equilibrium is
reached.

J. Dewell. Phillips Petroleum Co.:  In your cost comparison between enhanced
activated sludge and the granular carbon, I wasn't sure if the  enhanced
activated sludge assumed that the conventional activated sludge was already
in place or not.  Would these comparisons still be valid on a grass-roots
.treating system?

Len Crame;  We were assuming that activated sludge was already  in  place and
actually we were only comparing the cost of carbon contactors and  regenera-
tion equipment against the additional equipment you have to put in to add
powdered carbon.  We were not including filters.  We would believe the filter
would have to be a part of both treatment systems and would have no ettect
on this comparison.

                                     455

-------
J. Dewell:  Do you have any feel for what would be most  cost effective on a
grass-roots basis, assuming no treatment system at all exists?

Len Crame:  I think it is a safe argument that activated sludge  is  probably
the best system for driving the effluent organic levels  down for 1977  and
1983 but not necessarily the most cost effective.  At this  time  I do not see
any other system that perhaps can be enhanced cost effectively with powdered
carbon for 1983.

J.  Dewell:  I was referring to a situation where one does  not have activated
sludge at this time and is meeting 1977 standards so one is going to come to
1983 without any activated sludge.

Len Crame;  I think this must be determined on an individual basis.  As you
know when you calculate out guidelines for '77 or '83 you will find that in
some cases you are stuck with very tight guidelines for  a certain parameter
and I don't think it is appropriate to say which type of treatment  would be
best.  We would definitely not put in any powdered-carbon enhanced  system
until we piloted it and you would be taking a risk if you did.   All treat-
ment systems are unique, including activated sludge systems and  enhanced
biological systems.  We think that powdered carbon addition has  a lot  of
merit, but still you should determine it on a case-by-case  basis.

F. L. Robertaccio, DuPont:  I think that the easiest way to look at it is
that the activated sludge system in this case is common  to  both  the powdered-
carbon addition and the granular-carbon addition system  so  the difference in
cost here would have added to it the cost of the activated  sludge system if
you were starting out with a brand new plant.  You can use  that  as  a first
estimate, but what we have found at the plant I talked about yesterday is
that with a grass-roots plant you have additional savings that you  can accrue
to take full benefit of the system.  We talked about having smaller secondary
clarifiers, higher upflow rates through the clarifiers,  smaller  dewatering
equipment; and having no secondary solids disposal if you go through regenera-
tion.  So our experiences have been that with the grass-roots system you can
put in powdered carbon systems with regeneration for the same capital  costs
and essentially the same operating costs as a conventional  secondary waste-
water treatment system.  If you want further reference on this there was a
paper I referred to yesterday that had details of those  cost estimates.

J. E. Rucker, API;  Please comment on why your COD values were greater than
those we looked at earlier this morning from the Argonne work?

Len Crame;  The refinery where we were located is a very complex refinery
and I am quite sure that the refractory COD that remains is going to differ
from plant to plant.  We did try to exclude everything from the  chemical
plant, but I am not surprised really that we have a different refractory
COD level and I don't think you can compare the refractory  COD's out of
these carbon systems from plant to plant and find a great consistency  as far
as concentration goes.
                                     456

-------
Jeffrey Chen.  Dravo;  What would you propose  to  use  to  treat  the sludge
generated from the pretreatment unit?  Will the  cost associated with the
treatment be cost effective when compared  to  the improvement  of the following
bio system?

Len Crame;  We were assuming  that  for  our  best case  here that when we were
comparing granular carbon with powdered  carbon you would have a primary
sludge treatment and disposal problem  in both cases  so  that it really
doesn't affect our economics  here.  Sludge disposal  is  another problem and
again it does depend upon the availability of land and  other  considerations
and it is just something totally different; but  actually we're comparing the
two systems here and assuming that primary sludge is going to be a problem
in both cases.  You would have to  do a cost effectiveness study on the pre-
treatment and sludge disposal cost vs  the  benefit obtained from it.  But
from an operational standpoint, once you get  the colloids and oil out it is
much easier to operate  the activated sludge process, since the oil and solids
interfere with flocculation and sludge settling.

Tom McConomy, Calgon Corp.;   During the  period you were operating the granu-
lar carbon columns, was the carbon changed or was the same carbon used during
the entire test?

Len Crame;  As you will see in the paper we had  four carbon columns and we
would measure TOC at intermediate  points and whenever we found a breakthrough
on the first carbon column, we would  shift the carbon columns and put a fresh
regenerated column on the  tail-end of  the  system. This is why we are confi-
dent that the final column effluent is representative of what carbon adsorp-
tion can do with the regenerated  carbon.  We did try to determine how much
we had in those columns and we were running about 0.12  to 0.15 pounds TOC
per pound of carbon, which I  think is  fairly typical.  But because of the
activated sludge enhancement  you  get with  powdered carbon, where I won't
necessarily say "biological regeneration"  occurs, you are effectively regen-
erating it somehow by desoprtion,  or whatever, within the process and that
is what makes it economical.
                                      457

-------
                      FILTER
         PH
       CONTROL
    FEED
00
                                    ACTIVATED
                                      SLUDGE
                                      CONV.
                                     0.3 F/M
HIGH SA
                                     HIGH SA
                                        +
                                       PC-C
                                     HIGH SA
                                        +
                                      PC-H
                                      CONV.
                                     0.3 F/M
              FILTER
          CARBON COLUMNS
HIGH SA
                                     HIGH SA
                                        +
                                      PC-H
                                                  FILTER
                           CARBON COLUMNS
                                  FIGURE 1  - TREATMENT SCHEMES

-------
u
o
H
w
3
  600
  500
  400

 .- 300
  200
w
  100

 1000
  900
  800
  700
j 600
§500

 - 400
  300
w
  200
W
  100
                               1—I	I    I     I
                                                           T	T
                                      592^
                   ^DESIGNED FOR LESS THAN OPTIMUM PRETREATMENT
            I	|	I      I    I    I    I    I    I	I      I
            5     10     20    30   40  50  60  70   80     90     95

                 PERCENT OF TIME LESS THAN INDICATED VALUE

          FIGURE 2 - PART  I - TOC AND COD REMOVAL BY PRETREATMENT
                                 459

-------
H  -
a u
wo
Fn Q
fe W
W Od
  W
  H
  hJ
  M
100
 90
 80
 70
 60
 50

 40

 30
    300
             50TH PERCENTILE VALUES, MG/L
                                         TOG   COD
        A,
        B,
        C,
        D,
        E,
        F,
               FILTER - LOW SA
DAF
DAF
DAF
DAF
DAF
HIGH SA
HIGH SA-500 MG/L PC-C
HIGH SA-500 MG/L PC-H
HIGH SA-STAGE 2
LOW SA
64
56
55
48
53
58
128
108
 97
 89
116
105
                   PERCENT OF TIME LESS THAN  INDICATED VALUE

         FIGURE 3 - PART I - ACTIVATED SLUDGE TOG AND COD DISTRIBUTIONS

-------
 200
8100
Q 90
w 80
w 70
B 60
£ 50
£ 30
&

  20
  50

  40

•  30
.J
•—.
§ 20
 I1?
 a  f
 &  «
 w
                                     23
                                  i    i    I     I
                                                         1
                                                               95
                10     20    30  40  50  60  70   80     90

                 PERCENT OF TIME LESS THAN INDICATED VALUE

   FIGURE 4 - PART i - GRANULAR CARBON COLUMN TOC AND COD DISTRIBUTIONS
93
                                   461

-------
   700
   600
   500

   400
>j
g  300-

u
2  200
W
   100-
    90-
    80-
    70-
    60-
                                    334
                   *DESIGNED FOR LESS THAN OPTIMUM
                 j	I    I     II    I    I     I
                 PERCENT OF TIME LESS THAN INDICATED VALUE
          FIGURE  5  - PART II - TOC AND COD REMOVAL BY PRETREATMENT
                                 462

-------
  100
   90
>-)  80
5  70
s  60
o  50
o
w
a
H
3
   30
   20
   10
  300
 cs
 g
  ,200
 §
 o
 £100
 rf 90
 E 80
 £ 70
 H 60
 w
    50TH PERCENTILE VALUES, MG/L
  A, FILTER  -  LOW SA
B,
C,
D,
F,
G,
     DAF
     DAF
     DAF
     DAF
     DAF
HIGH SA
HIGH SA-1000 MG/L  PC-C
HIGH SA-1000 MG/L  PC-H
LOW SA
HIGH SA-2500 MG/L  PC-H
TOG
68"
44
41
34
46
22
COD
T63
111
119
 89
108
 72
	j-^l—sb—30   4o  s'o  b'u  ;i—s*	5*—rf—rt

            PERCENT  OF TIME LESS THAN INDICATED VALUE

 FIGURE  6  - PART II - ACTIVATED SLUDGE TOG AND COD DISTRIBUTIONS
                                   463

-------
    100
     90
     80
     70
  ^  60
  3  50

Bo"  40
w o
     30
w
  w
  S  20
     10
    200
Bo
go 100
.-j    90
fe w  8C
     70
     60
     50
  H
     40

     30
       2
                                    I    I    I
                          I     I     I    I    I
                        T5—JB—SIT

                   PERCENT OF TIME LESS THAN INDICATED VALUE

         FIGURE 7  - PART II - GRANULAR CARBON TOG AND COD DISTRIBUTIONS
                                   464

-------
w
3500


3000



2500



2000
w  1500
o
u
H
O
H
1000



 500



   0
                                 \
                              GRANULAR CARBON
                    POWDERED CARBON
               12345


                          FLOW MM GPD


       FIGURE 8  - COMPARISON  OF  ESTIMATED CARBON COSTS
                        465

-------
    50
    40
    30
o >-<
o w
    20
    10
     0
                    36
0.61
,B cc5b
                                   44
$3.19
LB COD
                  POWDERED
                   CARBON
            GRANULAR
             CARBON
                            $14.73
                            LB COD
           INCREMENTAL
           IMPROVEMENT
  FIGURE 9 - ESTIMATED EFFECTIVE  CARBON COST  AT 1 MM GPD

-------
PHENOLS
TABLE 1- PART II - EFFLUENT SUMMARY




        (ALL VALUES MG/L)




         OIL & GREASE
EFFLUENT
SAMPLE
EQ BASIN
FILTER
DAF UNIT
ASU A
ASU B
ASU C
ASU D
ASU G
ASU F
(HBON COL.
(•KIES A)
ORBON COL.
(PERIES F)
PERCENTILE
50TH 90TH
7.3 18.0
—
—
0.11
0.06
0.15
0.05
0.04
0.07
0.04
0.02
—
—
0.16
0.14
0.38
0.20
0.13
0.17
4.8
0.08
PERCENTILE
50TH 90TH
108
69
14
18
5
3
<3
<3
4
<3
<3
191
130
19
38
7
9
5
3
7
6
4
PERCENTILE
50TH 90TH
64.0 119
34.0
13.0
86.0
27.5
23.0
18.0
22.0
8.4
—
—
74.0
21.0
149
41.0
77.0
57.0
44.0
28.0
—
—
PERCENTILE
50TH 90TH
20.9 27.2
—
—
11.5 20.0
3.9 9.2
3.3 5.4
3.1 4.8
3.1 4.4
2.3 5.4
— —
—

-------
                                          TABLE 2 -PART II - SLUDGE DATA


 AVERAGE VALUE                                                ACTIVATED SLUDGE UNIT

NOMINAL LOADING
ACTUAL LOADING
MLSS, MG/L
MLVSS, MG/L
PC, MG/L
-P» V \JCC
CTi '" V£>::i
OO
OXYGEN UPTAKE, LB 02/LB COD REM
MG/L-MIN
SVI, ML/G
SETTLING VELOCITY, FT/MIN
BIOMASS PRODUCTION RATEd
TOTAL PRODUCTION RATE^
A
F/M =0.3
F/M = 0.3
1,487
1,302
0
88
0.40
0.16
95
0.17
0.22
0.25
B
50-DAY SA
39-DAY SA
1,892
1,562
0
83
0.68
0.12
64
0.34
0.16
0.19
C
50-DAY SA
42-DAY SA
2,728
2,269
l,000a
83
0.71
0.12
41
0.38
0.12
0.17
D
50-DAY SA
44-DAY SA
2,720
2,416
l,000b
89
0.49
0.11
43
0.38
0.11
0.14
F
F/M = 0.3
F/M =0.3
745
689
0
92
0.61
0.11
91
N/AC
0.19
0.21
G
50-DAY SA
56-DAY SA
4,096
3,898
2,500b
95
0.47
0.10
30
0.39
0.08
0.09
aCONVENTIONAL-SURFACE-AREA.
bHIGH-SURFACE-AREA.
CDISCRETE SETTLING
dLB VSS/LB COD REMOVED.
eLB TSS/LB COD REMOVED (INCLUDES CARBON)

-------
TABLE 3-POWDERED CARBON  (PC) REQUIREMENTS
ASU-PART
C-I
D-I
C-II
D-II
G-II
PC
LEVEL
(MG/L)
500
500
1,000
1,000
2,500
PC
TYPE
PC-C
PC-H
PC-C
PC-H
PC-H
PC
INVENTORY
(G)
28.4
28.4
56.7
56.7
141.9
AVG PC
LOSS
(G/DAY)
0.68
0.56
1.50
1.12
2.21
PC
MAKEUP
(%)
2.4
2.0
2.6
2.0
1.6
                469

-------
    SESSION IX

  COSTS/BENEFITS


Chairman

Thomas L. Hurst

Corporate Director of Safety and Environmental Services
Kerr-McGee Corporation, Oklahoma City, Oklahoma


Speakers

W.  Wesley Eckenfelder, Jr.
"Overview of Costs/Benefits"

Lial F. Tischler
"Treatment Cost-Effectiveness as a Function of
Effluent Quality"

Carl E.  Adams, Jr. and John H. Koon
"The Economics of Managing Refinery Sludges"

Melville Gray
"Compliance Monitoring Costs for the Priority
Pollutants"

Robert F.  Babcock, Leo J.  Duffy and Gilbert G. Jones
"Analytical  Costs in the Problem Pollutants"
        470

-------
BIOGRAPHY           Thomas L.  Hurst

        Thomas L. Hurst is Corporate Director
of Safety and  Environmental Services for Kerr-
McGee Corporation in Oklahoma City.  He has
a B.S. degree from North Carolina State Univer-
sity; M.S., University of Washington; and Ph.D.
(Engineering), University of Illinois; and is a
registered professional engineer in  Oklahoma.
He  is a member of AIME, American Petroleum
 Institute, American Mining Congress, American
 Chemical Society, Air Pollution Control Asso-
 ciation, and the Water Pollution Control  Fede-
 ration.
                                          471

-------
                        "OVERVIEW OF COSTS/BENEFITS"

                          W.  Wesley Eckenf elder, Jr.
                Distinguished Professor of Envrionmental and
            Water Resources Engineering, Vanderbilt University

                               Andrew Edwards
       Project Engineer, Associated Water and Air Resources Engineers

INTRODUCTION

     It is difficult to interpret cost/benefits in the classic sense when
considering present water pollution control limitations for industrial
effluents.  Two cases will be considered, the first involving a water
quality limiting discharge, and the second related to present effluent
guideline limitations.

     A water quality limiting discharge will usually involve effluent re-
quirements more stringent than that imposed by effluent guidelines and in
many cases require wastewater treatment beyond conventional secondary
treatment processes.  This discussion will relate to parameters pertinent
to the oxygen balance in the receiving water and the cost effectiveness of
wastewater treatment relative to this impact.

WATER QUALITY

     Several factors should be considered in the water quality limiting case.
A mixed wastewater may contain many organics of varying biodegradability.
The overall rate will be the sum of the individual rates.  For example a
recent study involving treatment in the activated sludge process using a
mixture of glucose, phenol and sulfonilic acid showed individual removal
rates of 0.072 mg/mg/day, 0.049 mg/mg/day and 0.015 mg/mg/day respectively
with an overall removal rate of 0.130 mg/mg/day.  The removal rate coefficient
K can be computed from Equation (1)

                            S  - S      S
                                    _ „ _e                              m
                                    * K  ^                              UJ
                              X t
                               v         o

The biodegradation rate coefficient, K, as defined by Equation (1) reflects
the overall rate for the wastewater in question.  Removal of the more
readily degradable organics through a treatment facility, i.e. the glucose
and the phenol in the above example, will cause a reduction in the de-
oxygenation rate coefficient (K) of the residual organics discharged to the
receiving stream.  For example, a raw wastewater might exhibit a BOD rate


                                     472

-------
coefficient  of 0.25/day, while after treatment and removal of the readily
degradable organics,  the resulting BOD rate coefficient might be 0.08/day.
Hypothetically,  discharging the same pounds of BOD will cause less of an
oxygen deficit if the deoxygenation rate is low as opposed to a high K rate,
as shown in  Figure 1.  This illustrates the fallacy of only considering the'
Ibs.  of BOD  discharged without regard to the resulting deoxygenate rates in
the receiving stream.  Therefore, relative to the impact on the receiving
stream, both the quantity of BOD as well as the biodegradation rate of the
residual BOD should be considered.  The effect of biodegradability on
activated sludge plant costs are shown in Figure 2.   It becomes apparent
that for cost effective design, higher concentrations of effluent soluble
BOD can be discharged for wastewaters of low biodegradability without as
significant  an effect on the oxygen balance in the receiving stream.

     There is an increasing emphasis today on nitrification in a wastewater
treatment facility, particularly as it relates to the oxygen balance in
the receiving water.  Removal of carbonaceous organics may move the nitri-
fication oxidation upstream closer to the wastewater  discharge resulting in
greater depletions of oxygen.  This phenomena is accentuated when the
wastewater treatment plant is nitrifying and thereby  discharging increased
numbers of nitrifying organics to the stream.  Temperature has a major
effect on the nitrification process, both in the treatment plant and the
receiving stream.  A cost effective wastewater treatment plant might be
designed to produce nitrification during the summer months when oxygen
depletion into the receiving stream would be greatest.  Nitrification would
not be significant during the winter months when the  nitrification rate in
 the stream and the resulting numbers of nitrifyers from the plant would be
 minimal.  The effect of temperature on nitrification  design and resulting
 costs  is shown in Table 1.

      TABLE 1 "EFFECT OF TEMPERATURE ON NITRIFICATION DESIGN AND COSTS"

            RAW WASTE LOAD             10 mg/1
                                       150 mg/1 BOD
                                       20 mg/1 NH3-N

            EFFLUENT QUALITY           15 mg/1 BOD
                                       2 mg/1 NH3-N

            TEMPERATURE, °C            10             25

            SLUDGE AGE, DAYS           14              5
            AERATION BASIN VOLUME,     2.0            1-0
              MIL. GAL.
            AERATION, HP               200            170
            CONSTRUCTION COST          $696,000       $481,000

 Significant capital  and operating costs  for  constfu^io\a^P™^ctor **"
 quired to achieve nitrification under cold conditions
 which nee'ds to be considered in the oxygen balance in the receiving stream
 is the fact that many substances that depress the reaeration coetticient,
                                     473

-------
such as surface active agents present in raw wastewaters,  are  removed  in
wastewater treatment.  The net effect is to increase  the reaeration  coefficient
in the receiving water with increasing degrees of wastewater treatment.  This
has the net effect of permitting higher organic  loads without  further  oxygen
depletion.  Because of the nature of the biodegradation process, both  in the
wastewater treatment plant and the receiving stream,  cost  effective  design
and operation should lead to a two-tiered standard in those parts of the
country where cold weather temperatures effect the biological  oxidation
process, both in the treatment facility and in the receiving water.

EFFLUENT GUIDELINES

     The second case considers those plants subject to effluent guideline
limitations.  It should be recognized that most  industrial plants are  or
will be coming into compliance with Best Practicable Control Technology
(BPT) regulations effective July 1, 1977.  This  would imply that most
industrial plants discharging organic wastewaters will have installed
biological wastewater treatment, and that any consideration of additional
reduction in pollutional loads should consider the existence of a bio-
logical wastewater treatment facility at that time.  The original effluent
limitations relating to Best Available Technology Economically Achievable
(BATEA) generally considered some in-plant reductions in wastewater  volume
and principally add-on end-of-pipe treatment units such as filtration  and
carbon adsorption.  Cases that have been evaluated by the writers would
indicate that other approaches to effluent quality improvement may be  con-
siderably more cost effective than indiscriminant add-on treatment facilities.
These approaches are:

     1)  In-plant changes to eliminate or reduce pollutional loads.

     2)  Installation of treatment systems for process modification  at
         specific discharge sources to eliminate, reduce or modify the
         wastewater characteristics to render them more compatible with
         existing wastewater treatment facilities.

     3)  Add-on tertiary*treatment units to the  existing wastewater
         treatment facility.

A detailed study was conducted for the Effluent  Standards  and  Water  Quality
Information Advisory Committee (EPA) in order to define the cost effective-
ness of pollution reduction by in-plant changes  with  existing  treatment
facilities as compared to additional end-of-pipe wastewater treatment  as
defined by the then BATEA criteria.  This study  considered an  activated
sludge plant in place at the time improved effluent quality was to be
considered.  The alternatives considered are coagulation,  filtration,
carbon adsorption and in-plant changes to reduce wastewater flow and strength.
The in-plant changes include equipment revision  and additions, unit  shut
downs, scrubber replacement, segregation, collection  and incineration, raw
material substitutions, reprocessing and miscellaneous small projects.. ,
Table 2 summarizes the results of this study.  Cost relationships  for  COD
removal for the various options are shown in Figure 3.  It is  apparent that
                                     474

-------
little benefit  in effluent quality is gained by  adding  carbon adsorption to the
activated sludge plant effluent over in-plant  changes with biological and
chemical  treatment.    As can be seen, additional end-of-pipe treatment,
including filtration and carbon adsorption, resulted in an increased removal
of 23,930 pounds of COD per day at an annual additional cost of $4.1 million
dollars per year.  In^contrast, in-plant  changes with minimal additional
treatment, i.e., chemical coagulation and filtration, resulted  in a COD
reduction of 19,260 pounds per day at a cost of  $1.3 million dollars per year.
It is readily apparent from Figure 3 that a cost effective analysis would
mitigate  against the application of carbon adsorption for the minimal
improvement in effluent quality achieved.

     In most cases, effluent treatment facilities have  been designed to treat
total wastewater discharges to levels consistent with effluent  guidelines
limitations established by EPA for specific industrial  categories and sub-
categories.  In many cases, removal of particular constituents  which are in-
hibitory  to the biological treatment process or  possess a very  low degradation
rate,by treatment of these constituents at their source can result in a
marked improvement in performance and an  increase in capacity of existing
biological wastewater treatment facilities.  Table 3 illustrates a case in
which one wastewater stream markedly reduced the overall biodegradation rate
in the biological wastewater treatment facility.

              TABLE 3 "REACTION RATE COEFFICIENTS WITH  AND WITHOUT
                      CARBON PRETREATMENT OF A PESTICIDE WASTEWATER"

                        Non-Carbon Treated          Carbon Treated

               T °C          K I/day                    K I/day

               28°C           2.25                       23.1

                8°C           0.81                        6.5

 Removal of this constituent by separate carbon adsorption treatment rendered
 the  total wastewater stream considerably  more  degradable. This would sub-
 stantially reduce the effluent pollutant  levels  from the biological treatment
 facility or permit higher organic  loadings through the  facility with
 resultant reductions in effluent discharges.   An evaluation of  both the bio-
 degradability and the effect of wastewater constituents on biodegradation on
 specific sources within the industrial  facility should  in many  cases lead to
 marked improvement in both wastewater treatment  plant operation and effluent
 variability.  This is particularly true where  new products are  to be intro-
 duced into the plant which may affect the overall biodegradation character-
 istics.

     A few years ago post-tertiary treatment  from the petroleum and chemicals
 industries generally considered filtration followed by  granular carbon
 adsorption columns.  Recent developments  in  the field show promise for  other
 processes and process modifications which are  considerably more cost
 effective for further pollutant reduction. These include the  application ot


                                      475

-------
polymers or coagulants prior to the final clarifier  for  improved suspended
solids reduction.  This approach involves minimal capital  expenditures.  The
application of powdered activated carbon in conjunction  with the biological
treatment process has shown considerable promise in  recent plant and pilot
plant studies.  In fact, several papers at this seminar  have reported on the
results of such studies.  Modification to post filtration  systems using
modified media, such as compacted clay or coal, to encourage and enhance bio-
logical action has shown the capability of reducing  effluent soluble BOD
levels to less than 5 mg/1 and suspended solids to less  than 10 mg/1.  In
this case post filtration serves the dual purpose of removal of suspended
solids and further biological oxidation,  since oxygen is  limiting BOD
reductions in the order of 10-20 mg/1 are feasible.  In  some cases, depending
on oxygen limitation, nitrification might also be achieved in this process.
Most of the data available to the writers at this time involves application
of this process to domestic wastewater.  Further experimental studies would
be required to indicate the feasibility of applying  such processes to
tertiary treatment of industrial wastewaters.
 BIOGRAPHY

        W.  Wesley Eckenfelder holds  a  B. E.  from
 Mahattan College;  a M.S.  from Pennsylvania  State
 University; and a M.E.  from New York University.
 Professor  Eckenfeleder  has  served  on the faculty
 of Manhattan College, the University of Texas  at
 Austin, the University  of Delft in Holland,  and
 is now Distinguished  Professor  of  Environmental
 and Water  Resources Engineering at Vanderbilt
 University.  Wes is also Board  Chairman of
 AWARE, Inc. and has written over 200 publications
 and 11 books on water quality management.   Profes-
 sor Eckenfelder has received many  awards.

        Andrew W.  Edwards is a  Project Engineer
 with the Operatinal Services Division  for Associ-
 ated' Water and Air Resources Engineers,  Inc.  (AWARE)
 of Nashville,  Tennessee.  He holds the following
 degrees: B.E.  Chemical  Engineering,  Vanderbilt
 University and M.S. Chemical Engineering,
 Vanderbilt University.   He  is a professional
 engineer in the State of Tennessee.
                                      476

-------
       POINT OF DISCHARGE
O
(T



S
(fl

H-
o
                                       = 0.25/day
                   MILES DOWNSTREAM



        FIG. 1. EFFECT OF DEOXYGENATION RATE ON OXYGEN SAG CURVE
                            477

-------
   3,000
o 2,500
o
o

o
   2,000
o  1,500
o:
CO

0  1,000
    500
                             BASIS    10 mgd
                             BOD     150 mg/l
                          (ACTIVATED SLUDGE ONLY)
Low K   K= I day-I
            10             20             30
              EFFLUENT  SOLUBLE BOD,  mg/l
                       40
       FIG. 2.  EFFECT OF BIODEGRADABILITY ON ACTIVATED SLUDGE COSTS
                            478

-------
(T)
a:
<4
_J
o
Q
-.  3
          Option
             A
             F
             G
             H
o
o
                             •
Technology

Activated Sludge and Coagulation
A and Inplant Changes
A and Inplant Changes and Filtration
A and Filtration and Carbon Adsorption
                          REDUCTION  IN COD
                            (1000 Ib/day)

         FIG.  3 ECONOMIC ALTERNATIVES FOR COD REDUCTION
                             479

-------
            TABLE  2  "EFFLUENT QUALITY  COST EFFECTIVE ALTERNATIVES "
Graph
Code
BPT
A
F
G
H
Pollution
Reduction
Scheme
Activated
Sludge
Activated
Sludge 5
Coagulation
All in-plant
changes 5
Activated
Sludge 5
Coagulation
All in-plant
changes 6
Activated
Sludge fi
Coagulation
§ Filtration
Activated
Sludge f,
Coagulation
6 Filtration
f, Carbon
Adsorption
Influent to
Treatment
Flow Sol. BOD
(MGD) (1000 Ib/day)
11.1 55.7
11.1 55.7
B.3 37.1
8.3 37.1
11.1 55.7
Effluent from
Treatment
Sol. BODr Tot. BOD COD TSS
> 	 ~ nnnn iTwH-nr^ 	 	 . .->

3.50 6.30 41.78 8,51
3.50 5,00 36,33 4.62
2.35 3.50 25.43 3,46
2.35 2.81 22.52 1.38
2.27 2.57 17.85 0.92
Annual Costs*
Capital Operating Total
Millions of Dollars
	 	 	
0.21 0.22 0.43
0.71 0.22 0.93
0.93 0.40 1.33
1.20 2.90 4.10
Costs presented are those above the cost of the installed BPT facility.  All costs are 197S dollars.

-------
    "TREATMENT COST-EFFECTIVENESS AS A FUNCTION OF EFFLUENT QUALITY"

                             Lial F. Tischler
            Austin Office Manager, Engineering-Science, Inc.


INTRODUCTION

    The cost of required wastewater treatment is a major concern to industry
and the petroleum  refining industry is no exception.  By July 1, 1977, all
petroleum refineries  in the United States are required  to have, as a minimum,
treatment technology  that will provide an effluent of the quality which the
U.S. Environmental Protection Agency (EPA) has defined  as "Best Practicable
Control Technology Currently Available" (BPCTCA).  Five years later, unless
the Congress acts  to  modify Public Law 92-500, the next level of treatment
technology, designated as "Best Available Technology Economically Achievable"
(BATEA), will be required.  In the interim period, effluent limitations for
certain toxic constituents in wastewaters, probably including petroleum
refining effluents, will be promulgated by EPA under the auspices of Section
307 of Public Law  92-500.  Thus, the petroleum refining industry will be
required to implement even more costly methods of wastewater control than
those which have been applied to date.

    This paper addresses the costs associated with the end-of-pipe treatment
processes which are most frequently associated with petroleum refinery secondary
and tertiary wastewater treatment.  The objective is to provide a frame of
reference for evaluating treatment costs as a function  of increasingly stringent
effluent quality limitations.  As such, the information presented herein is
not all-inclusive  in  terms of either the costs of reaching a given effluent
quality or the end-of-pipe treatment processes considered.  Because each
refinery has different wastewater problems, it is impossible to accurately
assess  the effectiveness and cost of in-plant controls  for reducing or elimi-
nating  raw waste load.  Therefore, these costs are not  included in this analysis.
Furthermore, to simplify the evaluation and allow comparison of these costs
with costs in other  industrial categories and between individual petroleum
refineries, it is  assumed that the costs are based on treatment units follow-
ing primary and secondary oil/solids separation.  Finally, a cost comparison
of various treatment  alternatives at a given level of effluent quality is not
the subject of this paper, but it should be recognized  that in a specific case
at a particular refinery, there may be several options  for biological treat-
ment and effluent  polishing which should be considered  in order to select the
least-cost alternative.

     All costs shown  in this paper are adjusted to January 1, 1977.  An annual
inflation rate of  six percent was used to perform this  adjustment and it is
recognized that this  may be too conservative in some cases.  To determine the
annual  cost of capital, an interest rate of 10 percent  amortized over 15 years
is used.  No attempt  was made to adjust the costs presented in this paper for
other considerations  such as geographical location, the availability of existing
!and, and similar  factors.  Reports which have been prepared on the costs of
wastewater treatment  in the petroleum refining industry have addressed these
                                      481

-------
considerations in depth  (Ref. 1, 3) and, while  their importance is recognized,
their exclusion from this analysis will not  affect attainment of the paper's
stated objective which is to provide a framework for evaluating cost as  a
function of effluent quality.  It must be  emphasized,  however,  that the  costs
presented herein are to  be used only to provide order  of magnitude comparisons
of one level of treatment with another and may  vary substantially at given
petroleum refineries.

     The remainder of this paper is organized into two main sections:  the
first presenting limited cost information  from  petroleum refining and petro-
chemical waste treatment facilities which  are considered representative  for a
cost comparison of this  type; and the second being an  analysis  of the costs
associated with a model  refinery which is  used  as an example to illustrate the
increasing costs as the  effluent quality limitations become more stringent.
The cost comparisons presented in this paper are presented on a unitized basis
using both the flow of the wastewater being  subjected  to treatment and the
removal of pollutants associated with each of the unit processes.  This  is a
useful way to evaluate cost data inasmuch  as it demonstrates unit process
cost-effectiveness not only in terms of the  quantity of wastewater treated,
but also in terms of the removal of pollutants  which it is designed to treat.

     The three levels of end-of-pipe control technology which are addressed in
this paper are biological treatment, as represented primarily by the activated
sludge process for petroleum refinery wastewaters, effluent polishing for
suspended solids removal which is represented by granular-media filtration,
and tertiary effluent polishing which is represented by activated carbon
adsorption in columns or beds.  In addition, cost data are presented for the
model refinery for removal of selected potentially toxic pollutants which may
occur in petroleum refinery wastewaters.

INDUSTRIAL CASE HISTORIES

     Industrial case history data for three  levels of  effluent  control are
presented in the following sections:  biological treatment, granular-media
filtration of biologically treated effluents, and activated carbon treatment
of selected industrial wastewaters.  These cost data,  which are primarily for
petroleum refining applications, are presented  to give the reader an idea of
the range of costs which might be encountered for treating petroleum refining
wastewaters to each of the effluent quality  levels which can be assumed  to be
approximated by the end-of-pipe treatment  technology described above.

Biological Treatment

     Table 1 presents cost data for six petroleum refinery case histories
utilizing biological treatment.  As described previously, these costs repre-
sent the biological treatment module only  and in all cases the treatment
process has been preceded by an air flotation unit for secondary oil/solids
removal.  The only solids handling costs included in Table 1 are for aerobic
digestion and thickening of waste activated  sludge.
                                      482

-------
     Two  basic types of biological  treatment  systems are shown;  completely
mixed activated sludge and rotating biological surface (RBS).  Other  bio-
logical processes which might achieve  similar levels of effluent quality,
the most  notable of which for petroleum refining wastewaters  is  the aerated
lagoon, are not included because  either insufficient data are available  or it
would be  difficult to make a proper comparison between the process  and the two
processes shown in Table 1.  The  aerated lagoon falls into the second category
because it is extremely land-intensive and also cannot be assured of  con-
tinually  meeting the level of effluent suspended solids provided by the
activated sludge and RBS systems.   This is not to say that the aerated lagoon
would not provide the necessary effluent quality and be the most cost-effec-
tive system for biological treatment at some  petroleum refineries.

    •The  cost data for the biological  treatment processes shown  in  Table 1 are
largely self-explanatory.  However, several interesting conclusions can  be
drawn from reviewing these data.  First, there is suprisingly little  variation
in the cost, in $/1000 gal, of biological treatment of petroleum refining
wastes considering the rather significant size and raw waste  load differences
between the petroleum refineries.   There is an obvious economy-of-scale  as a
function of flow for both the activated sludge and RBS systems.   The  influence
of organic raw waste load on both the  capital cost and operation and  mainten-
ance  (O&M) costs of biological waste treatment is demonstrated by these  data.

      It must be recognized that  there  are several factors which  influence  the
 cost-effectiveness of biological  treatment for a particular petroleum refining
 or other industrial waste.  One  factor is the size of the facility  in terms of
 both  quantity of wastewater treated and raw waste load.  There is substantial
 economy-of-scale in biological waste treatment systems at both higher flow
 rates and higher organic loadings.  Another important factor  in  determining
 the  costs of biological waste treatment is the biological kinetics  for  the
 particular wastewater being treated.   A wastewater with a high organic removal
 rate  requires a relatively smaller  aeration basin but may also require higher
 aeration capacity than a wastewater with a lower removal rate.  These factors
 have  obvious influences on both  capital and operating costs.   Thus, it must be
 recognized that the unit costs shown in Table 1 are subject to both increases
 and  decreases, depending upon the specific characteristics of a  particular
 petroleum refinery wastewater and the  design of the biological treatment
 facility.

 Effluent Suspended Solids Removal

      The next step in end-of-pipe treatment usually considered following
 biological treatment is enhanced  suspended solids removal.  In some cases,
 biological treatment systems for  petroleum refinery wastewaters  have  con-
 sistently demonstrated effluent  suspended solids concentrations  in  the  range
 of 15 to 10 mg/1.  This, however, is the exception rather than the  rule.
 At most refineries utilizing activated sludge plants, an average effluent
 total suspended solids (TSS) concentration in the range of 30 to 40 mg/1 can
 be expected.  Tertiary effluent  polishing consisting of either polishing ponds
 or granular-media filtration is  common for decreasing effluent suspended
 solids concentrations.  The polishing  ponds cannot be considered as an option
 which will work in all cases; therefore, only the granular-media filter systems


                                       483

-------
are considered in this cost analysis.  The  filters  are designed mainly to
remove TSS, however, organic materials, as  measured by five-day BOD and COD,
representing the organic content of the solids  removed by the filters, are
also removed.  In addition to this improvement  in effluent suspended solids
content and organic content, polishing filters  can  be considered a requirement
prior to further effluent polishing by activated  carbon adsorption or more
exotic processes such as reverse osmosis.

     Table 2 shows case history cost data for two different industrial
installations of biological effluent polishing.   These costs represent com-
parisons of different types of granular-media filtration systems used at the
two industrial plants.  The unitized capital costs  reflect a large economy-of-
scale engendered by the great difference in size  between the two facilities in
terms of both the quantity of wastewater treated  and the pounds of suspended
solids removed.  The annual O&M costs also  reflect  the economy-of-scale,  but
to a lesser extent than the capital costs.  The unitized costs are reasonably
low in terms of quantity of wastewater treated  and  pounds of suspended solids
removed, at least for the larger filtration system.   However the effectiveness
of a filtration system for removing organic material is quite poor,  as is
reflected by the high cost per pound of five-day  BOD removed in the filtration
system.  Thus, it can be concluded that the filters are a good unit process
for removing suspended solids if suspended  solids are considered a serious
pollutant.  On the other hand, the cost-effectiveness of the filters in
removing organic materials is very poor, as might be expected, and an alterna-
tive would be to design the biological system to  remove more of the organics,
if this is feasible, rather than using the  filtration system to remove organics.

Tertiary Organics Removal

     The remanded 1983 guidelines for the petroleum refining industry were
based on the use of activated carbon adsorption to  remove additional organic
constituents from biologically treated petroleum  refining effluents.   The
intention was to remove those organic compounds which pass through the bio-
logical treatment system, as measured by COD and  total organic carbon (TOC),
and which might either exert a long-term oxygen demand or create chronic
toxicity in the receiving waters.  Recently, it has been suggested that the
primary application of activated carbon might be  as pretreatment for selected
wastewater streams within a plant to reduce refractory or potentially toxic
organic compounds at the source.  Notwithstanding this possible future change
in application for activated carbon, there  is still emphasis by the EPA on
the use of carbon adsorption technology as  part of  BATEA effluent treatment.
Unfortunately, no full-scale operating data are currently available for carbon
adsorption technology in the petroleum refining industry.  Table 3 provides
some unitized cost information for the use  of activated carbon treatment
technology in several applications including the  only two full-scale plants in
the petroleum refining industry.  None of the unit  costs presented in Table 3
are directly comparable.  They represent different  types of treatment systems
and different wastewaters.  The two petroleum refining systems include the
continuous carbon columns at the BP petroleum refinery in Marcus Hook, Penn-
sylvania which treat refinery wastewaters that  have been pretreated by gravity
separation and granular-media filtration.   Biological pretreatment is not
included in this system.  The second petroleum  refinery system is at the ARCO

                                      484

-------
refinery in California.  This  system operates on an intermittent basis without
biological pretreatment and  consists of fixed beds which are used for organics
removal from mixed storm and process wastewaters.  The third set of unit costs
shown in Table 3 are estimates prepared for a tertiary carbon system following
biological treatment of mixed  industrial wastewaters, including petroleum
refinery wastes.  The costs  presented for the above three facilities exclude
the costs of pretreatment  systems.   The final two sets of unitized costs shown
in Table 3 are for the use of  powdered activated carbon in activated sludge
aeration basins.  Obviously, the level of effluent quality attainable with
these latter two systems is  not comparable with that of the former three, but
since this application is  receiving considerable attention, it was included in
this Table.

     Given the basis of the  data provided for the activated carbon systems in
Table 3, only general comparisons can be made.  However, it is easy to see
that the cost-effectiveness  in terms of both quantity of watewater treated and
pounds of organic materials  removed is much poorer for the activated carbon
systems than it is for biological treatment.  This is not unexpected consider-
ing the fact that the activated carbon systems are both energy and labor
intensive and the carbon itself is quite expensive.  In the tertiary polishing
application, the quantity  of five-day BOD and COD removed is relatively small
compared to that removed in  the biological treatment phase which tends to make
the unit costs of removal  much higher.  This phenomena is expected and, as
will be demonstrated in the  next section, makes the cost of this level of
effluent treatment quite high  in terms of cost per unit of pollutant removed.

MODEL REFINERY COST ANALYSIS

     The preceding sections  illustrated how the unit cost of waste treatment
 increases substantially as the effluent quality limitations become increas-
 ingly more stringent over  a  rather narrow range in terms of the commonly used
 parameters.  Since these costs reflect a wide range of waste characteristics,
 specific plant characteristics, and other similar factors which result in the
 rather wide ranges demonstrated by these data, it is useful to prepare an
 example analysis for a specific petroleum refinery to observe how treatment
 cost-effectiveness relates to  effluent quality.  This is accomplished by
 creating a hypothetical petroleum refinery, referred to hereinafter as the
 "Model Refinery," of a given size and with specified wastewater characteristics
 and  then preparing cost estimates for increasingly stringent levels of end-of-
 pipe control.  It is reemphasized that the costs presented in this paper
 include only end-of-pipe treatment and exclude costs associated with in-plant
 control and the pretreatment system prior to biological treatment.  The costs
 of solids handling, which  are  always significant, are not included in this
 analysis.

     The Model Refinery selected for the cost-effectiveness analysis is the
 Subcategory B, median refinery which was used in a study for the National
 Commission on Water Quality  (Ref. 3).  The characteristics of this hypothe-
 tical refinery are shown in  Table 4.  It is emphasized that this refinery is
 strictly hypothetical and  does not represent any actual refinery.
                                      485

-------
     The wastewater characteristics at varying levels of treatment for this
Model Refinery are shown in Table  5.  The values indicated in this Table can
be considered annual average concentrations  and flows.   This Table shows the
expected levels of effluent quality which will be obtained by the application
of biological treatment, tertiary  filtration,  and tertiary carbon adsorption.
These concentrations are used as the basis for the cost-effectiveness analysis.
In Table 6, the incremental mass of pollutant  removed by each of the three
end-of-pipe treatment processes being considered is shown.  It is obvious that
biological treatment removes the majority of the organic materials from the
petroleum refinery wastewater and  the additional levels of treatment, includ-
ing tertiary filtration and carbon adsorption, remove relatively small quan-
tities.  This impacts heavily on their cost-effectiveness.

     The capital costs, annual O&M costs, and  annual energy costs for each of
the three incremental levels of end-of-pipe  treatment for the Model Refinery
wastewater were calculated using cost data from a recent study conducted for
the State of Texas.  These costs are shown in  Table 7.   The annual O&M costs
shown in Table 7 do not include the cost of  energy, which is shown separately.
These costs can be considered to be planning-level cost estimates for the Gulf
Coast area.

     The data on the size and performance of the Model Refinery wastewater
treatment plant presented in Tables 5 and 6  can be used with the cost data in
Table 7 to analyze the cost-effectiveness for  each of the three levels of end-
of-pipe treatment.  These unit costs, presented in terms of quantity of waste-
water treated and pounds of specified pollutant removed, are presented in
Table 8.  To obtain the total annual cost for  a given end-of-pipe treatment
system and a selected basis, i.e., flow, BOD,  etc., the two figures shown as
capital and annual O&M costs can be summed.  To facilitate comparison of the
cost data shown in Table 8, several Figures  have been prepared.  Figure 1
shows the total annual cost of end-of-pipe wastewater treatment as a function
of annual average effluent five-day BOD concentration.   The unit costs in
terms of both flow and five-day BOD removal  are shown in this Figure.  In
terms of unitized costs as a function of flow, the cost increase between
activated sludge and granular-media filtration is approximately 40 percent.
The same increment, when viewed in terms of  additional five-day BOD removal,
represents an approximate sixfold  increase in  unit cost.  As might be expected
by the low incremental removals of BOD shown in Table 6 and the high cost of
treatment shown in Table 7, the activated carbon adsorption unit results in an
almost exponential increase in cost for a 10 mg/1 improvement in effluent
five-day BOD concentration.  However, since  the activated"carbon unit is
primarily designed to remove materials which are not easily biologically
degraded, its effectiveness is better viewed in terms of COD removal, as shown
in Figure 2.  Once again, the unit cost of COD removal increases very rapidly
as treatment beyond the biological step is implemented.  A threefold decrease
in effluent COD concentration between biological treatment and activated
carbon levels of effluent control  technology results in an approximate tenfold
increase in total annual cost, expressed as  $/lb COD removed.  Figure 3 ex-
presses the cost data shown in Figures 1 and 2 in a slightly different manner,
showing the cumulative total annual cost expressed as $/lb BOD or COD removed
as a function of percent removal of BOD and  COD from the pretreated  refinery
wastewater.  Once again, this Figure shows  the almost exponential increase in
                                                             /•
                                     486

-------
cost associated with the removal of  a  small increment of BOD going from
biological effluent to the activated carbon system,  whereas it shows  an almost
linear increase in the cost of  COD removal as a function of percent removal.

     Figure 4 presents the total annual costs of end-of-pipe waste treatment
at the Model Refinery from a different perspective than that presented  in the
preceding Figures and Tables.   This  Figure illustrates the actual cumulative
total annual costs for end-of-pipe waste treatment for the Model Refinery and
shows the distribution of these costs  between O&M and the cost of capital for
each type of facility.  This is simply a graphical representation of  the costs
shown in Table 7 but illustrates effectively some interesting relationships
not shown by the unitized costs.  The  addition of granular-media filtration to
an activated sludge system results in  about a 35 percent increase in  total
annual cost for the end-of-pipe treatment system excluding primary and  second-
ary oil/solids removal.  Most of this  cost is associated with the capital
costs of the filter system as the increase in annual operation and maintenance
costs due to the filter is only about  10 percent of the same costs for  the
biological system alone.  When  the  carbon adsorption step is added for
removal of refractory organics, however, the O&M costs actually exceed  the
annual costs for amortizing  the capital investment in the activated carbon
units.  This is brought about principally by the energy costs associated with
carbon regeneration and the  cost of  purchased carbon for makeup of system
losses.  The increase in total  annual  costs due to the addition of the  carbon
system is about 390 percent  of  the cost of the biological system alone  but is
a far smaller percentage increase than if the costs are considered on a unit-
ized basis, as illustrated in Figures  1 through 3.  This serves to reemphasize
 the point made earlier:  the capital costs of implementing the carbon technology
for polishing of biological  effluents  are not significantly greater than the
 cost for biological treatment plus filtration and are less than the overall
 costs of meeting BPCTCA effluent quality.  However, the cost of this  level of
 treatment per unit of pollutant removed is many times that of the biological
 system and the petroleum refineries  will pay dearly for every mg/1 of COD
 removed by this method.  This raises very strongly the question of whether or
not the uniform application  of  carbon  adsorption technology to polishing
 biological effluents is a necessary  and cost-effective requirement for  meeting
 the quality standards set for our nation's waters.

     The preceding cost evaluations  have dealt with the end-of-pipe treatment
 systems which will be used in petroleum refineries to remove organic  substances
 measured as BOD, COD, and suspended  solids.  An extremely important considera-
 tion over the next several years will  be the impact on the petroleum refining
 industry of the toxic effluent  limitations which are currently being  prepared
 by  EPA in accordance with Section  307  of Public Law 92-500.  In a consent
 decree from the Circuit Court of Appeals in Washington, D.C., the EPA agreed
 to  promulgate toxic effluent limitations for selected constituents from a list
 of  some 65 different classes of compounds.  As of the date of this paper, some
 109 different chemical compounds are being studied in detail to determine
whether or not toxic limitations should be promulgated under Section 307.
 Since these limitations have not yet been promulgated, an evaluation of the
 cost-effectiveness of the associated treatment technology is not possible at
 this time.  However, as an example,  several constituents which may be found in
 petroleum refinery wastewaters  can be  used to evaluate cost-effectiveness of
 treatment processes designed for specific constituent removal.
                                      487

-------
     Cadmium, cyanides, and chromium can be found in some petroleum refinery
wastewaters and can be toxic to  certain organisms in the receiving waters if
they are discharged at high enough  concentrations.  There is, however,  con-
siderable controversy as to the  toxic concentrations of many of these materials
and the selection of these for an example cost-effectiveness analysis is not
to be construed as suggesting that  the limitations selected for the performance
of treatment processes to remove these constituents represent their toxic con-
centrations in receiving waters.

     Cadmium is not present in most crude oils;  therefore, the source of
cadmium in petroleum refinery wastewaters is either intake water or cadmium
addition during processing (Ref. 3).   Cadmium in petroleum refinery waste-
waters can generally be traced to intake water,  corrosion products, the
addition of cadmium compounds for distillate desulfurizing, or as a lube oil
additive to prevent oxidation.   Because of the diffuse nature of the cadmium
sources in petroleum refineries, it is not generally practical to attempt to
segregate cadmium-bearing waters for separate treatment.  Thus, for the Model
Refinery used in the cost analysis,  treatment for removal of cadmium involves
treatment of the entire process  wastewater effluent.  Lime precipitation and
filtration is probably the most  dependable method of removing cadmium.   Cad-
mium forms an insoluble and highly  stable hydroxide precipitate at an alkaline
pH and, in the absence of appreciable complexing agents, precipitation  and
filtration provide effective removal.   The completeness of the reaction is a
function of pH and lime addition to pH 10 is required to meet the concentra-
tion limit selected for this evaluation, which is 0.05 mg/1.  The cadmium
removal process selected requires provisions for lime addition, rapid mix,
flocculation, sedimentation, filtration, and reneutralization to pH between  6
and 9.  Table 9 shows the total  annual costs, the amortized capital costs, and
the annual 0&M_costs for removing cadmium from the Model Refinery wastewater
stream of 1.0 MGD.  The total annual cost of $0.40 per 1000 gallons is  approxi-
mately two-thirds due to O&M costs,  most of which are associated with the
purchase of lime, and one-third  due to the amortized capital costs.  No cost
per pound of cadmium removed can be calculated for this process because of
insufficient information on cadmium concentrations in petroleum refining
effluents.  However, it is known that these concentrations are generally quite
low, and thus the cost per pound of cadmium removed is likely to be extremely
high.

     Simple and complex cyanides are generated in cracking and coking
operations as reaction products  and enter the effluent from overhead receivers
and through washing operations  (Ref.  3).  Most or all of the cyanide genera-
tion in a refinery occurs in these  two operations, so the cyanide-bearing
water can generally be isolated  from other waste streanis within the refinery
complex and treated separately.  Alkaline chlorination is the most proven
treatment technology for removing cyanides from wastewaters although its
application in the petroleum refining industry has not been extensively tested
(Ref. 3).  Cyanides can be reduced  to less than 0.025 mg/1 with appropriate  pH
control, chlorine dosage, and residence times.  Biological treatment of re-
finery wastewaters is quite effective in reducing cyanides to concentrations
below 0.5 mg/1, however, for the purposes of this cost-effectiveness analysis
it is assumed that alkaline chlorination will be required to produce a final
effluent concentration of approximately 0.025 mg/1 total cyanides.  In the


                                      488

-------
Model Refinery this will be  accomplished by segregating the sour water streams
after stripping, from the  fluid catalytic cracker and coker and subjecting
them to alkaline chlorinatlon.   The total estimated quantity of the wastewaters
requiring treatment in the model refinery is 150,000 gpd, containing approxi-
mately 100 mg/1 of total cyanides.   The annual costs associated with treatment
of these waste streams in  the Model Refinery to remove cyanides are shown
in Table 9.  The capital costs  of the cyanide removal system are quite low,
since the only equipment items  required are a chlorinator, reaction tank, and
a caustic feed system.  However, the annual O&M costs per 1000 gallons of
waste treated are extremely  high, resulting in a total annual cost of $2.24
per 1000 gallons.  In terms  of  total annual cost per pound of cyanide removed,
the cost-effectiveness of  this  unit operation is even poorer, resulting in a
cost of $2.68 per pound of cyanide removed on an annual basis.  The majority
of the costs for this process are associated with the neutralization chemical,
sodium hydroxide, and the  chlorine.  Another potential drawback of this pro-
cess which should be mentioned  is the possible formation of chlorinated hydro-
carbons which, of themselves, may prove to be toxic materials.

     The final specific constituent considered for removal is chromium.  The
major source of chromium within petroleum refineries is the chromate-based
corrosion inhibitors used  in cooling towers.  These inhibitors find their way
to the wastewater treatment  plant in cooling tower blowdown.  Chromium is
included in the effluent  limitations for the petroleum refining industry which
have been promulgated by  EPA, but it is also being considered under the toxic
effluent limitations being reviewed at this time.  In petroleum refineries,
 the  effluent chromium limitations can usually be met without resorting to
 separate treatment of cooling tower blowdown.  However, in many other types  of
 industrial plants, particularly in the organic chemicals industry, cooling
 blowdown constitutes a very  high volume and is not treated in the biological
waste treatment plant.   In these cases, it is often necessary to remove
 chromium from the cooling tower blowdown before it is discharged to the
 receiving waters and several treatment methods have seen wide application.
 The  types of treatment  technology used for removing chromium from cooling
 tower blowdown include  chemical reduction/precipitation, ion exchange, and
 electrolytic reduction/precipitation.  Each of these treatment processes has
 advantages and disadvantages.  The electrolytic process is finding increasing
 applications for treating cooling tower blowdown and is used in this example
 cost analysis for the Model  Refinery.  It is assumed for the purposes of this
 example that the Model  Refinery generates approximately 190,000 gpd of cooling
 tower blowdown which is  treated by the electrolytic reduction/precipitation
 method.  The costs of applying this treatment procedure are shown in Table 9.
 The  total annual cost for chromium removal by this method for this size
 facility is approximately $0.66 per 1000 gallons, but is $4.06 per pound of
 chrome removed.  This assumes an influent chromate concentration in the
 cooling tower blowdown  of  approximately 20 mg/1 and an effluent concentration
 from the unit of approximately 0.5 mg/1 of chromate.

 SUMMARY
    rjhls  paper  has presented an evaluation of the cost-effectiveness of
 several  important processes in petroleum refinery wastewater treatment using
 both  actual  cost data from full-scale facilities and an example cost analysis

                                      489

-------
for a hypothetical petroleum refinery.  The salient  points  of  this  evaluation
can be enumerated as follows:

     (1)  Cost-effectiveness, expressed on a unit  cost  basis as  a function of
          effluent quality, should be based upon the constituent or con-
          stituents of the wastewater which are to be removed  by the waste
          treatment process being evaluated, if the  most meaningful compari-
          sons are to be made.

     (2)  In terms of removal of organic materials from petroleum refinery
          effluents, biological treatment is by far  the most cost-effective.
          Suprisingly, multimedia filtration is the  second  most  cost-effec-
          tive, and tertiary effluent polishing with carbon adsorption is the
          least cost-effective of the likely alternatives for  this  applica-
          tion.

     (3)  The poor effectiveness of carbon adsorption technology when used on
          a biologically treated effluent for  the  purpose of removing trace
          refractory materials raises a serious question as to whether this
          is the most suitable application for this  treatment  process.  In
          the absence of known specifically toxic  organic constituents,
          questions should be raised about the efficacy of  removing trace
          residual organics, as measured by COD or TOG, simply for  the purpose
          of reducing the effluent concentrations  unless there is a con-
          commitant improvement in receiving water quality.  This particular
          point was made quite effectively by  Mr.  Joe Moore at last year's
          Open Forum.  At that time he indicated that the cost-effectiveness
          of BATEA technology should be carefully  considered before requiring
          nationwide implementation (Ref. 6).

     (4)  The cost-effectiveness of unit processes designed for  the removal of
          specific constituents in petroleum refining waste streams will
          probably be quite poor in terms of unit  cost  per  quantity of waste-
          water treated or constituent removed.  This is not unexpected,
          however, and, if a constituent is truly  toxic at  the effluent
          concentrations being considered, the unit  cost of removing the
          constituent has no real meaning since Public  Law  92-500 precludes
          discharge of toxic materials in toxic quantities.

     (5)  Expressing the performance of various waste treatment  unit processes
          in terms of unit costs per quantity  of flow treated  or mass of
          constituent removed is an effective  basis  for cost comparisons
          between alternative unit processes.  It  is a  convenient method for
          putting costs of treatment processes designed to  perform  a given
          service on a consistent basis and allows a direct comparison of the
          respective costs of each option.
                                      490

-------
                                 REFERENCES


1.   Brown & Root, Inc.  "Economics of Refinery Wastewater Treatment,"
    American Petroleum Institute,  Publication Number 4199, Washington
    D.C.  (August 1973).

2.   Gulp, G.L. and A.J. Schukrow.   "What Lies Ahead for PAG?," Water and
    Wastes Engineering, pages  67-74 (February 1977).

3.   Engineering-Science,  Inc.   "Petroleum Refining Industry - Technology
    and Costs of Wastewater Control," National Commission on Water Quality,
    Washington, D.C.  (June 1975).

4.   Engineering-Science,  Inc.   Selected technical reports (1970-76).

5.   McCrodden, B.S.   "Treatment of Refinery Wastewater Using Filtration
    and Carbon Adsorption," Paper presented at a Technology Transfer
    Seminar jointly  sponsored  by Environment Canada, The Pollution Control
    Association of Ontario, and the Canadian Society of Chemical Engineers
    (October 24, 1974).

6.  Moore, J.G.  "The Role of  the National Commission on Water Quality
    (NCWQ)," Proceedings  of Open Forum on Management of Petroleum Refinery
    Wastewaters, EPA, API, NPRA, and the University of Tulsa, Tulsa,
    pages 39-50  (1976).

7.  Prosche, M.A.  "Activated  Carbon Treatment of Combined Storm and Process
    Waters," Proceedings  of Open Forum on Management of Petroleum Refinery
    Wastewaters, EPA, API, NPRA, and the University of Tulsa, Tulsa,
    pages 399-410  (1976).

8.  Rizzo, J.A.  "Case. History:  Use of Powdered Activated Carbon in an
    Activated Sludge System,"  Proceedings of Open Forum on Management of
    Petroleum Refinery Wastewaters, EPA, API, NPRA, and the University of
    Tulsa, Tulsa,  pages  359-374 (1976).

BIOGRAPHY      Lial F. Tischler

       Lial  F. Tischler is manager of the
Austin, Texas office of Engineering-Science,
Inc. He holds the degrees of B.S. in Civil
Engineering from the  University of  Texas at El
Paso and M.S. and Ph.  D. in Civil Engineering
from the University of Texas at Austin.  He is a
registered  professional engineer in the State
of Tpps.  Prior to joining Engineering-Science,
 > 'tlte was Director of Systems Engineering at
Texas Water Development Board.
                                      491

-------
             CUMULATIVE  TOTAL ANNUAL COST OF END-OF-PIPE TREATMENT
N)
N
s£
*r"^f ^^
m ^

31
m>

-< o
   o
      m
      m
      m
        jp m
          CD
          o
            in
          >




          I
          m
          m o
          m
          O |M
          o o

          r>
            IN)
                     P
                     KI
                     01
                     O
                     O
                         p
                         in
                         o
      FLOW $/IOOO GAL
                                    01
        o
        o
        ro
        01
                                                                  CD
        ro
        b
        o
                                             •ACTIVATED CARBON-
                               GRANULAR MEDIA FILTRATION
                            •ACTIVATED SLUDGE
                                             8
                                             -i
                         ro
                         b
                         o
OJ
b
o
b
o
                                                           o
.
b
o
                                                                       00
                                                                       b
                                                                       o
                                         -Q3AOW3M aoa

-------
     1.75
z
LJ
i
     1.50
     1.25
iL

o
UJ

o
O
  >  1.00

  o:

  «» 0.75
    0.50
    0.25
              ZERO COST IS
              AT39Omg/l
              COD, AFTER
              PRETREATMENT
             25
                     50
75
100
125
150
            ANNUAL AVERAGE EFFLUENT COD CONCENTRATION
                          mq/l

   Fig.2COST OF WASTE WATER TREATMENT FOR
       MODEL  REFINERY AS A FUNCTION OF
       EFFLUENT COD
                        493

-------
     8.00 i—i
ID
o
      7.00
o
O  o
    LU
    LJ
    o:
      6.00
      5.00
    CO

    O
    o

    .n

    •w-
      4.00
LU

>    3.00
      2.00
      1.00
               70      80       90      100
         PERCENT REDUCTION IN DESIGNATED CONSTITUENT-
         ZERO IS AIR FLOTATION UNIT EFFLUENT QUALITY


 Fig.3 UNITIZED  TOTAL ANNUAL COST OF

     WASTE TREATMENT AS A  FUNCTION OF
     REMOVAL EFFICIENCY-MODEL REFINERY
                    494

-------
  400,000
  300,000
H-
(f)
O
O

UJ 200,000
O
  100,000
           ACTIVATED GRANULAR TERTIARY
            SLUDGE   MEDIA   ACTIVATED
                   FILTRATION  CARBON
 Fig.4TOTAL ANNUAL COSTS OF WASTE
     TREATMENT-MODEL  REFINERY
                  495

-------
                                                TABLE  1

                   CASE HISTORIES OF BIOLOGICAL TREATMENT AT  PETROLEUM REFINERIES
System
Type
Rotating
Biological
Surface
Rotating
Biological
Surface
Activated
Sludge
Activated
Sludge
Activated
Sludge
Activated
Sludge
Flow
(MGD)
0.9
4.3

4.6

3.9

1.9

2.3
Organic Removal
BOD5 COD
(Ib/day) (Ib/day)
840 1,200
5,300

9,100 26,200

41,500 83,000

8,600 13,000

3,100
Amortized Capital Cost*
($71000 gal) ($/lb BOD) ($/lb COD)
0.23 0.24 0.16
0.19

0.12 0.06 0.02

0.19 0.02 0.01

0.24 0.05 0.03

0.19 0.14
Annual Operation and
Maintenance Costs
($71000 gal) ($/lb BOD) ($/lb COD)
0.06 0.12 0.05
- - -

0.10 0.05 0.02

0.07 0.006 0.003

0.12 0.03 0.02

— — —
*  Adjusted to 1977 and 10 percent  interest  for  15  years.
(Ref. 4)
METRIC CONVERSIONS _
  I/sec       =    (MGD)  (43.81)
  kg          =    (Ib)(0.454)
  1000 liters =    (1000  gal)(3.785)

-------
                                              TABLE 2

Type
Filter
System
Downflow
Deep Bed
Downflow
Shallow Bed
Downflow
Deep Bed
Gravity
Downflow
Deep Bed
Pressure
Tri-media
Gravity
Tri-media
Pressure
* Adjusted
(Ref. 4)
INDUSTRIAL CASE

Solids
Flow Removal
(MGD) (Ib/day)
1.44 240
1.44 240
48.5 24,269
48.5 24,269
48.5 24,269
48.5 24,269
to 1977 and 10 percent
HISTORIES OF

TERTIARY GRANULAR-MEDIA

Amortized Capital Cost*
($/1000 gal)
0.19
0.14
0.05
0,07
0.04
0.04
interest for
($/lb TSS) ($/lb BOD)
1.15 3.83
0.85 2.85
0.11 0.36
0.13 0.44
0.08 0.28
0.09 0.29
15 years.
FILTRATION

Annual Operation and
Maintenance Costs
($71000 gal) ($/lb TSS) ($/lb BOD)
0.07 0.41 1.37
0.06 0.34 1.14
0.04 0.07 0.22
0.03 0.07 0.22
0.03 0.07 0.22
0.03 0.06 0.19

METRIC CONVERSIONS
I/sec       =    (MGD)(43.81)
kg          =    (Ib)(0.454)
1000 liters =    (1000  gal)(3.785)

-------
                                                                  TABLE 3
00



T? T S\T,T
E J.OW
System (MGD)
Petroleum Refinery - '
Process wastes without 2.16
biological treatment,
continuous columns
Petroleum Refinery -"'
Storm/process water 0.85
without biological
treatment, fixed beds**
(4)
Mixed Industrial - '
Tertiary treatment , AiMt 48.5
continuous columns
(a)
Petroleum Refinery -
Powdered carbon 1.08
addition to
aeration basin
(21
Municipal - '
Tertiary powdered
activated carbon 10.0
addition to
aeration basin
* Adjusted to 1977 and 10 percent
CASE HISTORIES OF INDUSTRIAL ACTIVATED CARBON TREATMENT

Organic Removal Annual ODcration and

BOD5 COD Amortized Capital Cost* Maintenance Costs
(Ib/day) (Ib/day) ($/1000 gal) ($/lb BOD) ($/lb COD) ($/1000 gal) ($/lb BOD) ($/lb COD)

783 2,090 0.31 0.62 0.23 0.134 0.37 0.14



1,311 - - - 0.56 - 0.25



5,600 67,550 0.17 1.43 0.12 0.27 2.31 0.19


11 347 None, Manual addition of 0.03 2.72 0.09
powdered activated carbon



0.08 - - 0.02


interest for 15 years. METRIC CONVERSIONS
       ** Intermittent operation.
       ***Excludes cost of necessary pretreatment by granular-media filtration.
       Note:  Numbers in parentheses following system description indicate references which
              can be found at the end of the text.
I/sec
kg
1000 liters  =
(MGD)(43.81)
(Ib)(0.454)
(1000 gal)(3.785)

-------
                                   TABLE 4
                      CHARACTERISTICS OF MODEL REFINERY
                       FOR COST-EFFECTIVENESS ANALYSIS
Subcategory B, Cracking
Crude throughput:  42,500 BPSD
Refinery characteristics:
          Crude desalting
          Atmospheric distillation
          Vacuum distillation
          Hydrocracking
          Fluid catalytic cracking
          Coking
          Hydrotreating
          Catalytic reforming
          Asphalt
EPA Configuration:  5.91
       METRIC  CONVERSION
                    (0.0159)(BPSD)
   3
  m /day
42,500
42,500
16,000
 9,000
12,000
 2,000
 8,000
10,000
 1,000
BPSD
BPSD
BPSD
BPSD
BPSD
BPSD
BPSD
BPSD
BPSD
 (Ref. 3).
                                   TABLE 5
                  MODEL REFINERY WASTEWATER CHARACTERISTICS
AT VARYING LEVELS OF TREATMENT





Annual Average
Parameter'
Flow, MGD
BOD5, mg/1
COD, mg/1
TSS, mg/1
(Ref. 3)

After
Air
Flotation
1
150
390
30


After After
Biological Tertiary
Treatment Filtration
1
25
125
40
METRIC CONVIISION
499
1
15
97
12
I/sec

After
Tertiary Carbon
Adsorption
1
5
40
6
= (MGD) (43. 81)


-------
                                   TABLE 6
INCREMENTAL MASS
	 	 { ,„„ 1,111,1 	 	 	 	
Parameter
BOD,., Ib/day
COD, Ib/day
TSS, Ib/day
OF POLLUTANT REMOVED
11 11 i — •• ii — 	
Biological
Treatment
1,042
2,210
-
BY TREATMENT
^^^^^^^•^^^^^MP^MMto^^^MM^^^MHI
Tertiary
Filtration
83
233
233
PROCESS
1 "- - i • •• i". I, .
Tertiary
Carbon Adsorption
83
475
50
METRIC CONVERSION
         (Ib)(0.454)
                                   TABLE 7

        INCREMENTAL COSTS OF WASTEWATER TREATMENT FOR MODEL REFINERY
          Treatment Type
                    Annual            Annual
   Capital       Operation and        Energy
    Cost       Maintenance Cost        Cost
Activated Sludge

Tertiary Granular-media
Filtration

Tertiary Activated Carbon
$ 461,000
$ 32,000
$  9,000
$ 220,000
$ 740,000
$ 6,000
$ 92,000
$ 2,000
$ 52,000
Basis:  1977 Gulf Coast Costs.
                                     500

-------
                                  TABLE 8

          INCREMENTAL COST-EFFECTIVENESS OF WASTEWATER TREATMENT
                            FOR MODEL REFINERY
    Cost-effectiveness
         Parameter
Activated
 Sludge
                                         Treatment Method
 Tertiary
Filtration
    Tertiary
Carbon Adsorption
     Columns
Flow, $/1000 gal


BOD5


COD,


TSS,


Capital
Annual O&M
, $/lb removed
Capital
Annual O&M
$/lb removed
Capital
Annual O&M
$/lb removed
Capital
Annual O&M
0.16
0.09

0.16
0.08

0.07
0.04

-
-
0.08
0.02

0.95
0.20

0.34
0.07

0.34
0.07
0.27
0.25

3.20
3.04

0.56
0.53

5.31
5.04
Basis:   1977 Gulf  Coast,  Capital amortized at 10 percent for 15 years.

METRIC  CONVERSIONS
  $71000 liters -  ($/1000 gal)(0.26)
  $/kg          =  ($/lb)(2.2)
                                     501

-------
                                   TABLE 9

              COSTS FOR REMOVAL OF SELECTED WASTE CONSTITUENTS
FROM MODEL REFINERY EFFLUENT

Waste
Constituent
Cadmium*
Cyanides**
Chromium***

Amortized
Capital
Cost
($71000 gal)
0.13
0.22
0.23

Annual
O&M
Costs
($/1000 gal)
0.27
2.02
0.43

Total
Annual
Costs
($/1000 gal) ($/lb removed)
0.40
2.24 2.68
0.66 4.06
*   Cost based on treating entire refinery process  effluent.
**  Assumes only FCC and coker process wastes are treated.
*** Assumes treatment of cooling tower blowdown only; 190,000 gpd.

METRIC CONVERSIONS
  1000 liters  =  (1000 gal)(3.785)
  kg           =  (Ib)(0.454)
                                     502

-------
                  THE ECONOMICS OF HANDLING REFINERY SLUDGES

                                Carl E. Adams, Jr., President
                                            and
                       John H. Koon, Director, Wastewater Management

                      Associated Water and Air Resources Engineers, Inc.
                                    Nashville, Tennessee

      The various types  of sludges generated in refinery operations may require handling using
alternative thickening, stabilization  and dewatering processes  prior to final disposal. Some or all
of these processes may be required depending on the exact  nature of the sludge prior to final
disposal. Final disposal practices might include landfilling, lagoohing, land farming, or incineration
of the sludge prior to disposal  of ash using one of the  above methods. The predominance of each
of these disposal  methods as it existed in 1973  in an estimate for 1983 conditions is presented
in Table 1. These data indicate that landfilling and lagooning  are the predominant methods in use
at the present time, while it is anticipated that the predominance of lagooning as a disposal
method will decrease significantly in  the next several  years and will be accompanied by a signifi-
cant increase in the  popularity of land  farming. Incineration is not expected to be in widespread
use in the industry in the foreseeable future.
      Due  to the large  number of alternative processes  for handling and disposal sludges, the
cost associated with the construction and operation of these processes plays a large role  in the
selection of  optimum sludge handling systems. This paper presents a discussion of the types and
characteristics of  sludges originating from refinery operations, alternative techniques for handling
these sludges, and  the costs associated with various process sequences.

SOURCES AND CHARACTERISTICS OF REFINERY SLUDGES

      Normal refinery operations generate several major types of sludges which can be classi-
fied as follows:

       1.  Storage tank bottoms.
      2.  API separator bottoms.
      3.  Crude desalting sludge.
      4.  Catalytic solids.
      5.  Spent clays and coking fines.
      6.  Solids from utilities operations and biological wastewater treatment systems.

      The basic  source of many of the solids  is the crude oils which  contain materials that are
present in  the oil  as taken from the well and which separate during transportation and storage of
the crude oil. The solids usually associated with the crude oil include iron rusts, iron sulfides, clay,
sand, salt crystals, wax, and paraffin. These solids will generally settle out either in the storage
tank bottoms or  API separators. Solids from biological treatment of wastewaters are generated
when soluble and colloidal biodegradable organics are converted into a biological mass  which is
separated  from the  treated wastewater  by gravity  and  concentrated  for  disposal.  The waste
sludge from  API  separators plus typical secondary treatment facilities will  contain approximately

                                            503

-------
1,500 to 2,000 Ib of dry solids per million gallons of effluent treated. A delineation of types and
characteristics of refinery solids are given in Tables 2 and 3.
       Those sludges which pose the major handling and disposal problems to a refinery are from
the utilities and wastewater treatment systems. 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
relatively  inert and have no heat content, they should be dewatered and landfilled directly. Sludges
gnerated from the wastewater treatment facility 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 organic sludges.
       The sludges  from a refinery which are  readily combustible include the waxy bottoms, oil
sludges, coke fines, and waxy tailings. The non-combustible sludges are sand, rust, silt, tetraethyl
lead sludge, salt, spent catalysts, and lime sludge. The excess biological sludge, although not readily
combustible, can be dewatered  to an autogenous state.

SLUDGE HANDLING METHODOLOGY

        In order to implement  a successful sludge management program in a refinery, both in-plant
recovery and  reclamation of applicable solids combined with final  sludge handling and ultimate
disposal are required.  In-plant management consists of tight operational control, preventive  main-
tenance 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 filtration, pressure filtration, pressure
belt filtration, or centrifugation, and  final disposal by incinceration, landfill, land farming and
barging to sea.  Alternative  sludge handling methods for oily and biological sludges are shown in
Figures 1  and 2, respectively.
       There are  basically three types of sludges  which must be handled by  refinery  wastewater
treatment and sludge handling systems:

        1.  Recoverable oils.
       2.  Oily  sludges.
       3.  Biological sludges.

       Solids concentrations attainable using various dewatering processes is included in  Figure 3.
An in-depth discussion of alternative handling  methods has been presented elsewhere  (2) and, al-
though many refineries handle the oily sludges in combination with the biological  sludges, the
handling will be presented separately for clarification in this paper. Detailed design procedures for
these processes are presented in Reference 4.

Recoverable Oils

       Generally,  recoverable oils are separated from water mixtures by heating with steam to the
range of 150° to  180° F. Chemicals are added and the emulsion  is broken  into three phases:  oil,
water,  and  sludge. The oils are usually recovered, the water sent  to  the wastewater treatment
system, and  the sludges discharged  into  the oily sludge handling system. Separator  skimmings,
which  are generally referred to as slop oils, require treatment before they can be reused, due to the
high content of  solids and water. Solids and water contents in excess of one percent generally inter-
fere with processing. These slop oils are easily treated by heating to 190°F, retained at this tempera-
ture for 4 to 6 hours, then settled for 12 to 24 hours. At the end of the settling period,  three layers
exist:  a top layer of clean oil, a middle layer of secondary emulsions, and a bottom layer of water
containing solubie components, suspended solids and oils. Frequently, it is advantageous or neces-
    to add acid or a specific chemical to  destabilize slop oil emulsions. The water layer resulting

                                             504

-------
from this treatment  contains higher levels of BOD and COD and must  be treated through the
biological system prior to discharge. Slop oils are also successfully treated  by precoat filtration(2).

Oily Sludges


       Oily sludges are derived from oil-water separators, tank bottoms or cleanings, air flotation
treatment of wastewaters,  and cleaning or dredging from  lagoons or oxidation  ponds. A major
consideration  in  treating oily sludges  is that these materials are generated from various  sources
and are discharged at irregular intervals. Thus, the  sludge composition is highly variable from day
to day.  Often  tank  bottoms contain  heavy emulsions which require special treatment, and the
sludge and skimmings from air flotation systems can present special  difficulties in handling also.
In many cases, the skimmings from air flotation are sent to a skimmed oil tank from which they
may be discharged to landfill or pumped to an  oil recovery unit. Chemical treatment is generally
necessary for oil recovery.
       The most common  processes for handling  oily sludges are gravity thickening, vacuum fil-
tration or centrifugation dewatering, and disposal  by landfill or land  farming. A summary of the
sludge handling methods used for oily sludges is given in Table 4.

       Thickening.   Gravity thickening methods  are generally utilized for oily  sludges. Solids
loading rates in the range of 5 to 30  Ib/sq  ft-day are reported with thickened solids concentra-
tions from 3 to  10 percent. Basically, solids and  oils recovery  are low from gravity thickeners.
 Due to the presence of heavy particles in oily sludges, flotation thickening has not been very suc-
cessful for these sludges.  Dissolved  air flotation has been  successful for thickening froth flota-
tion sludges, but other refinery oily sludges are not deemed practical with this method.

       Dewatering.   Dewatering alternatives for oily sludges include centrifugation, vacuum fil-
 tration, and pressure  filtration. Figure 4 presents a schematic of  a system being utilized for a num-
 ber of oily sludges. This two-stage system  utilizes a first-stage horizontal solid bowl centrifuge to
 separate  oil from the centrate. In this process the sludge is heated to 180° to 200° F prior to centri-
 fugation. 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 tank bottoms.
       3.  A  horizontal solid bowl  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
           cake will consist of 1 to 5 percent oil and approximately 50 to 60 percent solids.
       4.  A high-speed nozzle centrifuge separates 95 to  98 percent of the feed oil in the oily
           phase and 2 to  5 percent in the nozzle water. Thirty to 50 percent of the feed solids
           will be  removed with the oily  phase with the remainder  being in two water phases.

       A significant  number of refineries  use vacuum filtration for dewatering oily sludges; and
 if properly implemented, vacuum filtration renders the solids suitable for direct landfill or incinera-
 tion. 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.  The  major conclusions regarding
 vacuum filtration of oily sludges are:

       1.  Increased feed temperature greatly improves vacuum solids performance.
       2.  Addition of spent clay decreases oil recovery and solids filtration rate.
       3.  Measured  filtration  loadings of 0.8  to 3.0 Ib/sq ft-hr  are required for oil recovery.
                                             505

-------
       A fixed plate, high pressure filter press may be used to dewater many types of oily sludges.
Effective breaking of solids-stabilized emulsions is obtained with cake solids in excess of 50 percent
of oil concentrations in the range of 5 to 20 percent being observed. Increased filtration tempera-
ture has a dramatic effect on cake solids and cake  oil concentrations at  fixed cycle times.  Re-
finery  experience indicates that a definite increase  in  filtration efficiency is  obtained  with  lime
additions. A significant decrease in cake  oil  content can be  obtained  by washing  the  cake with
hot water.
       Test  results  indicate that the recovered 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,
       A new method of sludge dewatering in  the United States employs gravity draining of sludges
coupled  with pressure filtration applied  by  mechanical means of rollers and belts.  These belt
pressure filters are capable of achieving solids content  in a range of 12 to  30 percent using poly-
electrolyte additions of approximately 5 to 2-0 Ib of polyelectrolyte per ton of dry solids.

        Ultimate  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 is viewed as a
short-term alternative and is eliminated as an option. The disposal of oily sludges on soil is accep-
table if it can be shown that such disposal will not contaminate groundwater or contaminate 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 decompositon of the oil-containing solids.  Land farming involved
spreading the sludge in 4 to 6-in. layers, allowing the sludge to dry about one week, adding nutri-
ents, and then discing the sludge into the soil. Decomposition rates have found to average approxi-
mately 0.5  Ib/mo-cu  ft  wihtout nutrient addition  and  1.0  Ib/mo-cu ft with nutrient addition.
        Incineration of oily sludges with landfilling of the ash may provide an acceptable means of
final  disposal in land limited situations. The three  most common types of incinerators include
fluidized bed, rotary kiln, and multiple-hearth furnaces. The fluidized bed incinerator is best suited
for feeds that are partially liquid so that the incinerator can be fed by pumps and screw conveyors.
The multiple-hearth incinerator will be more economical if most of the  feed is in the form of cake
or non-pumpable solids.

Biological Sludges

        The sludge handling methodology used for biological  sludges is similar to  that described
for oily sludges.  However, due to the biodegradable  potential of these sludges, stabilization is
required prior to disposal by sea or land.  The basic process sequence for biological sludge handling
consists  of stabilization,  thickening, dewatering, and final disposal. Often the  thickening step will
precede  stabilization in  order  to reduce the stabilization cost which is highly  dependent on flow.
A summary of the sludge handling processes and design basis is shown in Table 5.

        Biological  Sludge Stabilization.  The  most common method of stabilizing waste biological
sludge in the refining industry is by aerobic  digestion. Aerobic digestion is employed to  stabilize
the sludge and render it suitable for land disposal, although  the overall sludge quantity may be
somewhat reduced and the dewatering characteristics slightly improved. Approximately 10 to 20
days detention time 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 considered suitable for land  application.
                                             506

-------
       Thickening of Biological Sludges.  The three most common processes used to thicken bio-
logical  sludges in the refining industry are  flotation, gravity thickening, and centrifugation  The
belt pressure filter is receiving increasing attention as a thickener and dewatering device for refinerv
sludges.
       Results of tests  using  refinery sludges indicate that these sludges are quite suitable for flo-
tation thickening. Data indicate that the optimum pressure is approximately 50 to 55 psig and no
polymer  addition is justified. The optimum air-to-solids ratio appears to be about 0.01 Ib air/lb
solids and that float solids on the order of 3 to 5 percent can be readily achieved.
       With gravity thickening (the most common thickening process in refinery practice), loading
rates in the range of 2 to 15 Ib  per sq/ft  per day are  employed with solids concentration being
achieved in the  range 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 when
solids  loading  do not vary  substantially.  In extremely warm  climates, gravity thickeners  may
generate obnoxious biological odors if they precede the stabilization process.
       Centrifugation has been utilized sparingly for thickening waste biological sludges in the re-
fining  industry.  Basket centrifuges are capable of thickening waste activated sludge to levels of
5 to 6 percent  concentration with an 80 to 95  percent solids capture.  One refinery has reported
using basket centrifuges for thickening to a concentration of 8 percent.  However, excessive main-
tenance was experienced because of vibrational problems.

       Dewatering of  Biological  Sludges.  The most common methods of dewatering biological
sludges are vacuum filtration, centrifugation, and pressure filtration with the belt filtration press
achieving an increasing use in  the refining industry.
       Vacuum filtration is the most common method used to dewater refinery wastewater sludges.
Chemicals are  usually  required  and 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  combination  from  an  economical  and performance  standpoint.  The  optimum ferric
chloride dosage  usually ranges from 200 to 400 Ib/ton of dry solids. Tests and full scale experi-
ences  indicate  that  vacuum filtration of refinery  biological sludges will usually achieve solids
concentration in the range of 10 to 16 percent at filter loading rates ranging from 1 to 5 Ib/sq ft-hr.
 Digested sludge will usually dewater slightly  better than raw, undigested sludge.
       Basket centrifuges have been found to provide the best centrifugal method of  concentrating
 waste  biological  solids,  and  there are a number of  applications utilizing basket centrifuges in the
 refining  industry. Normally,  centrifuges provide a cake solids  ranging from 8 to 18 percent con-
 centration. There have been  some  problems in  the utilization of the  basket centrifuge due to
 mechanical vibrations; however, these problems are being solved by constructing the  baskets so
 that it is driven from the top rather than the bottom.
       Pressure  filtration  will usually achieve a  solids concentration up to 50 percent with bio-
 logical sludges, thus producing the driest cake for disposal. The pressure levels for the pressure fil-
 tration system range from 50 to 225 psig. Normally, it has been found that the pressure filtration
 system required higher chemical  dosages than the vacuum filtration or centrifugation systems.
       The use  of various pressure belt filters for dewatering of waste biological solids and some
 oily sludges is becoming increasingly widespread in refining and other  industries. A summary of
 data which has recently been obtained using these devices is shown in Table 6.  Generally, pressure
 belt filters are capable of dewatering refinery waste activated sludges to keep concentrations be-
 tween 15 to 20 percent.  More limited data processing a combination of waste activated and oily
 sludges indicates cake solids contents of approximately 27 percent are obtainable. The wide range
 of polymer dosages used in these tests are indicative of variation both in the types of polymers and
 sludges processed.  Loading rates  to the belt presses range from 1 to 5 Ib/inch  belt width for bio-
 logical and oily sludges. Figure 5 presents capital costs estimates for pressure belt filters as a func-
 tion of the quantity of biological sludge.  It is important to note from data in Figure 5 that increas-
 ing the solids content applied to the belt filter by a factor of 2 will  reduce the capital cost almost
 in half. Consequently, it appears that thickening ahead of the pressure belt filters is definitely justi-
 fied.
                                             507

-------
       Ultimate Disposal.   The alternative for ultimate disposal  of biological  sludges are similar
to those for a waste oily sludge.  In  many cases, a thick sludge or activated sludge can be placed
in 4 to 8 inch layers and utilized as a limited nutrient and soil conditioner. In this case, additional
waste is removed by natural evaporation and infiltration, into the  underlying soil. Following initial
periods of drying, the sludge layer can be disced to encpurage the activity of aerobic soil bacteria.

ECONOMIC CONSIDERATIONS

       It is extremely difficult to collect representative  and comparable cost data on sludge hand-
ling systems in the refining industry. Different combinations of  oily and biological sludges, dif-
ferent labor costs, and construction  and capital costs which might  be a function of total equip-
ment purchased from a manufacturer (such as a complete wastewater system in which the sludge
handling were only a fraction of  the total equipment purchased), make it extremely difficult to
develop cost as a function of  size of  equipment. Therefore, some  data are gathered which are con-
sidered to be reliable and presented in a form for discussion in this section.
        In  Figure  6, a sludge  handling system is  shown whereby biological and  oil sludges were
combined for dewatering by  pressure filtraton or vacuum filtration prior to landfill (5). The bio-
logical solids (1,070 tn/yr) were aerobically digested and gravity thickened. The oily sludges (1,000
tn/yr) were thickened and stored.  The combined biological and oily sludges (2,070 tn/yr) were de-
watered  by pressure  (Alt.  I) or vacuum filtration (Alt. II) and transported to landfill. The third
alternative consisted  of direct transport by piping of the combined  thickened biological and oily
sludges for land farming.                                   x   ,            ,
           Costs for Alternative I (dewatering using pressure; filtration followed by landffll dis-
posal of the sludge  cake) were based on operation of the pressure filter 140 hr/wk and transport
of the dewatered  cake approximately two miles  to the landfill site. Costs were included for the
use of 1 Ib  conditioner/lb sludge dewatered and it was assumed that a cake having a solids content
somewhat  greater than 50  percent would  be obtained.  The filters had facilities included for pre-
coating and were  designed  for a two-hour cycle time. The total sludge volume transported to the
landfill site  was 18 cu yd/day.
       Costs  for  Alternative II  employing  vacuum filtration for sludge  dewatering  included
operation of the vacuum filter 135 hr/wk. The filter  was designed using a loading rate of 1.4 Ib/sq
ft-hr and it was assumed that a 40  percent cake would be produced. The vacuum  filter was also
designed as a  precoat system and it was assumed that the amount of precoat required would equal
10 percent  of the weight of  sludge  to be dewatered. Transport  distance  to the landfill was also
assumed to be 2 mi  and the quantity of cake requiring disposal was calculated to be approximately
40 cu yd/day.
       Costs for land farming  of the sludge (Alternative III) were based on the application of sludge
in 4 to 6 inch layers at an application rate of 5 Ib/sq ft-yr. It was assumed that following applica-
tion the sludge would be allowed  to  dry for one  week during which nutrients would be added to
the soil, and subsequently followed by discing of the sludge into the soil.
        In this situation, dewatering  by vacuum filtration was 15 percent cheaper than pressure
filtration; however, the transportation costs were  almost twice as  much due to the wetter vacuum
filter sludge. In summary, the vacuum filtration scheme  was approximately 4 percent cheaper than
the pressure filtration scheme. However, Alternative III,  land  farming, was about 50 percent more
economical  than either pressure or  vacuum filtration. The land  farming  assumes discing with a
tractor and disc rather than the more expensive bulldozer used in some other published methods.

Pressure Filtration Costs

       Estimated  costs for  pressure  filtration  dewatering  of oily sludges  is presented  in Table 7.
These costs were  based on  dewatering API separator bottoms, tank bottoms, decant pit sludges,
flotation units solids, spent clays,  oil emulsions, and other miscellaneous oily sludges (6).  Each
press was designed using a 1.5-inch cake thickness, and 64-inch diameter plates. It was also assumed
that the feed sludges would be heated to a temperature  of  150° to  200° F prior to the dewatering
period.  It was assumed that  lime would be used as a conditioning agent with a dose of 0.1 Ib lime/lb
dry solids dewatered.

                                             508

-------
       Dewatering costs presented in Table 7 range between $26 and $40/tn including both capital
and operating costs. Costs for pressure filtration shown  in Figure 6 are substantially higher than
those shown in Table 7 where an in-depth cost analysis of pressure filtration was performed. The
higher costs associated with pressure filtration in  Figure 6 are partially attributable to the presence
of biological solids which make the sludges more difficult to dewater and require larger quantities
of conditioning chemical. In addition, a substantially larger quantity of sludge was used as a basis
of the estimates presented  in Table 7 - 9,200 tn/yr compared to 2,070 tn/yr used in Figure 6.
       The analysis presented in Table 7 indicates that dewatering costs are minimized with three
shifts per day operations and become progressively higher with 2 and 1 shift per day operations.
However, the cost estimates for three shifts per day operation contain no flexibility in the design
since only one machine was used for  24 hr/day operation.  It was estimated that the  use of two
presses, each having the capacity to dewater 60 percent of the total solids throughout, would add
approximately $125,000/yr to the total annual cost. On a unit cost basis, this would increase the
total sludge dewatering costs to approximately $39/tn. In many instances, it might be more eco-
nomical to provide storage facilities for holding sludge when maintenance of the press is required.
In individual cases, the optimum design would require consideration not only of initial construc-
tion and operating costs but also of labor policies  which might interfere with three shift per day
operation, the operational flexibility desired, the  degree to which  alternative designs could accom-
modate plant expansion in the future, etc.

Costs for Centrifugation of Oily Sludges

        In Table 8, the  cost of a centrifuge system for handling oily sludges is presented. The cost
to dewater 660 tn/yr of flotation unit skimmings using a basket centrifuge was estimated to cost
$160/tn (7). A two-stage system consisting of a horizontal  solid bowl and disc-nozzle centrifuge
in series for handling combined oily sludges was  estimated to cost only $45/tn for  treating 1,750
tn/yr. Costs for the basket centrifuge treating flotation skimmings was estimated based on a 17 gpm
flow rate, 90 percent on-stream factor, and 2 percent influent solids concentration. Costs of the
basket centrifuge and disc-nozzle system in series was estimated based on a 7 percent influent solids
concentration at a flow of 13 gpm and a 90 percent on-stream factor. In  using the horizontal
solid bowl and disc-nozzle centrifuge in series, the feed  would initially be processed through the
basket centrifuge with  the centrate and the emulsion  contained  in this stream being broken and
handled with the disc-nozzle centrifuge.

 Incineration Costs

        Costs  for the incineration  of refinery sludges using alternative dewatering methods is pre-
sented in Table 9.  In  each case, the sludge consisted of flotation unit solids, refinery oil sludges,
spent clays and emulsions (Whitewater). Each example was based on dewatering 8,400 tn/yr dry
solids.  If only the flotation solids are  dewatered using centrifugation and the remainder of the
streams are fed to the incinerator without dewatering, the total cost of incineration would amount
to approximately $53/tn. By dewatering both flotation solids and oily sludges using centrifugation,
costs were observed to  decrease approximately $45/tn.  Pressure filtration of all wastestreams prior
to incineration further reduced the total incineration costs to approximately $41 An.

 Land Farming

        For Allowable oil loadings ranging from  2.5 to 25 Ib/sq  ft-yr, the required land  area for
land farming of oily sludges can be determined from Figure 7.  Literature costs for land farming
of oily sludges are shown in Table 10.  Although the total costs per ton of waste are approximately
the same, $22 and $24/tn, there is a significant discrepancy in each case regarding the loading rate
of oil to  the soil, cost  of transportation, and cost of cultivation. The cost for land  farming taken

                                             509

-------
from Figure 6 totaled $35/tn of solids. Based on a reported 50 percent oil content for this sludge,
the cost would be equivalent to approximately $70 of oil applied to the land. Costs for this latter
example were developed based on an application rate of 5 Ib oil/cu ft-yr applied to a depth of ap-
proximately 6  in. which  is equivalent to a surface  loading of approximately 2.5 Ib/sq ft-yr. While
the unit costs for land farming  estimated in  each of these three examples are not extremely diver-
gent, very significant differences exist for cost estimates for individual items. It is felt that this
is partially the result of differences inherent to each individual situation. However, it is also likely
more accurate costs will be available as the practice of land farming continues to be used in the
refining industry.

                                       REFERENCES

1.     Jacobs Engineering Company.  "Assessment of Hazardous Waste Practices in the  Petroleum
       Refining  Industry."  Environmental  Protection Agency, Contract Number  69-01-2288,
       Pasadena, California, June 1976.

2.     Adams, Carl E., Jr., and Stein, Robert M., "Sludge Handling Methodology for Refinery
       Sludges," Proceedings of the Open Forum on Management of Petroleum Refinery Waste-
       waters. The University of Tulsa, 1976.

3.     Cross, F.  L. and Lawson, J. R., "A New Petroleum Refinery." American Institute of Chem-
       cal Engineering Symposium Series, Vol. 70, No. 136, p. 812.

4.     Adams, Carl E., Jr., and Eckenfelder, W. Wesley, Jr., Process Design Techniques for Indus-
       trial  Waste Treatment. Nashville, Tennessee:  Enviro Press, 1974.

5.     Ford, Davis L., "A Preliminary Engineering Study for Wastewater Treatment and Pollution
       Abatement," September, 1971.

6.     Ford,  Davis L., "A Preliminary  Engineering Study for Solid Waste Disposal and Pollution
       Abatement," February,  1972.

7.     Kincannon, C. B.,  "Oily Waste Disposal  by  Soil Cultivation," Proceedings of the Open
       Forum  on  Management of Petroleum  Refinery  Wastewaters. The University of Tulsa,
       1976.

8.     Huddleston, R. L. and Cresswell, L.  W., "The Disposal of Oily Wastes by Land  Farming,"
       Proceedings of the  Open  Forum on  Management  of  Petroleum  Refinery Wastewaters,
       The University of Tulsa, 1976.
                                            510

-------
BIOGRAPHIES

    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 industries.  He holds the
following degrees:  B.E., Civil  Engineering,
Vanderbilt University; M.S.,  Environmental
Engineering, University of Texas.  He is a pro-
fessional engineer, registered  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 professional societies,
has authored and presented more than 50 technical
papers, and edited and  co-authored three books
dealing with industrial wastewater treatment
technology.
    John H. Koon is Director of Wastewater
Management for Associated Water and Air Resources
(AWARE), Inc., Nashville, Tennessee.  He holds a
B.E. degree in Civil Engineering and a M.S. degree
in Sanitary and Water Resources Engineering from
Vanderbilt University and a Ph.D. degree in
Sanitary Engineering from the University of
California at Berkeley.  Dr. Koon is a registered
professional  engineer with eight years of experience
in industrial and municipal wastewater management.
He is also the author of  numerous technical papers
and reports.
                                       511

-------
            TABLE 1.  REFINING INDUSTRY SLUDGE DISPOSAL PRACTICES3

                                                       DISTRIBUTION (%)

     METHOD                                1973                         1983 (est.l
 LANDFILL
 LAGOON ING
 LAND FARMING
 INCINERATION

 ONSITE DISPOSAL
 OFFSITE DISPOSAL
                   50
                   40
                    9
                    1

                   44
                   56
                                     44
                                     19
                                     34
                                      3

                                     73
                                     27
aJacobs Engineering Company. "Assessment of Hazardous Waste Practices in the Petroleum
  Refining Industry." Environmental Protection Agency, Contract Number 68-01-2288,
  Pasadena, California, June 1976.
          TABLE 2. CHARACTERISTICS OF REFINERY SOLIDS WASTES (Ref. 3)
                                Typical Composition, Percent
       Waste Type

API Separator Sludge


Tank Bottoms

Chemical Treatment Sludge

Air Flotation Froth

Precoat Vacuum Filter
 Sludges

Biological Treatment Sludges
   Raw
   Mechanically Thickened
   Centrifuged

   Vacuum Filtered
   Screw Pressed

Water Treatment Sludge
   Oil or
Hydrocarbon  Water
    15
66
48
5
22
22
40
90
75
29
4 8
5
3
49
     0
     0
     0

     0
     0

     0
98
94
85

75
40

95
      Volatile    Inert
       Solids   Solids      Characteristics
 6      13    Fluid slurry of oil,
               water and sand
                                    Oil-water mixture

                                    Slightly viscous fluid

                                    Thick, oily fluid

                                    Temperatures
 1.5     0.5   Water Consistency
 4       2    Thick, but pumpable
10       5    Viscous-peanut butter
               consistency
15      10    Wet crumbly solid
40            Intact, solid cake

         5    Pumpable Fluid, some-
               times gelatinous
Cross, F. L. and J. R. Lawson. "A New Petroleum Refinery." American Institute of Chemical
Engineering Symposium Series, Vol. 70, No. 136, P, 812.
                                        512

-------
        TABLE 3.  ESTIMATES OF REFINERY SOLID WASTE QUANTITIES (Ref. 3)

     Waste Types                                                    Unit Loads

API Separator Sludge                                           200 mg/l Suspended Solids

Chemical Treatment                                            50 mg/l Suspended Solids
 (API Separator Effluent)                                        Removed Only

Biological Sludges                                        '      0.7 Ib Dry Solids per Ib
                                                              BOD Removed

Water Treatment Sludge

 A. Lime Soda Ash                                            2 parts Dry Sludge per
                                                              1 part Hardness Removed

 B. Ion Exchange                                              0.4 Ib Salt per 1,000 Grains
                                                              Hardness

Office Wastes                                                 1 -0 cu yd per Employee
                                                              per Month

Cafeteria                                                      0.6 Ib per Meal


Cross, F. L. and J. R. Lawson, "A New Petroleum Refinery." American Institute of
Chemical Engineering Symposium Series, Vol. 70, No. 136, p. 812
                                          513

-------
                            TABLE 4.  SUMMARY FOR THE THICKENING AND DEWATERING ALTERNATIVE OF OILY SLUDGES
      Type of
       Sludg
     Oily Sludge
  P'rocess

Thickening
Gravity
Flotation
                     Dewatering
                     Centrifu-
                     gation
                    Vacuum
                    Filtration
en
                    Pressure
          Average Performance


3-7 percent solids, 50% recovery
84-93% solids recovery, 3.5 - 5.5%
float solids
                75-90% solids recovery when charged
                with heated oily sludges. Oily phase
                will consist of 1-5% solids and cake will
                contain 50-60% solids. Nozzle ejector
                centrifuge separates 95-98% of oil in
                oily phase and 205% in the nozzle and
                nozzle water. 30-50% of nozzle ejector
                feed solids will be removed with oily phase.
                5-20% oil, 40-70% solids cake
Design Parameters