PB85-207538
                                       EPA/600/2-85/059
                                       May 1985
    THE RUTGERS STRATEGY FOjR COMPOSTING;
         PROCESS DESIGN AND CONTROL
       H. S.  Finstein,  F.  C.  Miller,
    S. T. MacGregor,  and K.  H.  Psan'tmos
Rutgers, The  State University of New  Jersey
      New Brunswick,  New Jersey  08003
            Grant Nc,  R806829010
              Project Officer

               Atal  E. Eralp
        Wastewater Research Division
   Water Engineering Research Laboratory
          Cincinnati, Ohio  45268
   WATER ENGINEERING RESEARCH  LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION  AGENCY
          CINCINNATI, OHIO   45268

-------
                                   TECHigiCAL REPORT DATA
                            (I'lcase read Imtniclions on the reverse before completing)
1. REPORT NO.
 _ EPA/600/2-85/059	
4. TITLE AND OUUTITLE
  The Rutgers  Strategy for Composting:   Process

  Design and  Control
             8. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
  M.S. Finstein,  F.C.  Miller, S.T.  Mac^reqor and

 _K.M. Psarianos
                                                            0. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  The State  University of New Jersey,  Rutgers
  Department of  Environmental Science
  P.O. Box  231 New Brunswick
  New Jersey 08903
             li REPORT DATE
             10. PROGRAM ELEMENT NO

               CAZB1B
             11. CON TO ACT/GRANT NO.
                                                              R806329010
12. SPONSORING AGENCV NAMt AND ADDRESS
  Water Engineering Research Laboratory -  Cin, OH
  Office of  Research and Development
  U.S. Environmental Protection Agency
  Cincinnati, Ohio   45268
                                                            13. TYPE OF REPORT AND PERIOD COVERED
               Final Project  Report
             14. SPONSORING AGENCY CODE
               EPA/600/14
15. SUPPLEMENTARY NOTES
  Project Officer:   Atal  E. Eralp   (513)  684-2621
16. ABSTRACT
       A strategy  for sludge composting  was  developed to ccuriter the tendency of
  other composting systems to operate  at high temperatures  t,l«at  inhibit and slow
  decomposition.   This method, known as  the  Rutgers strategy,  can be implemented
  in a static  pile configuration to retain  structural and operational  simplicity,
  or_in a more  elaborate enclosed or reactor structure system.   The method main-
  tains a temperature ceiling that provides  e high decomposition rate  throunh on-
  demand removal of heat by ventilation  (thermostatic control  of a blower).
       Compared with the approach currently  in widespread use, the Rutgers strat-
  env yields high-rate composting that decomposes four times more waste in half
  the time.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDCN TIFIERS/OPEN ENPED TERMS   C.  COSATI Field/Group
18. DISTRIBUTION STATEMENT
   RELEASE TO PUBLIC
                                               19. SECURITY CLASS (Tins Report/

                                                 UNCLASSIFIED
                           21 . NO. OF PAGES
                                  284
20. SECURITY CLASS (Thispagel

  UNCLASSIFIED
                                                                          22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION i= OBSOLETE

-------
                                 DISCLAIMER


     Although the Information described in this document has been funded
wholly or in part by the United States Environmental Protection Agency
through assistance agreement number R806829-01-0 to Rutgers, The State
Universityof New Jersey, it has not been subjected to the Agency's required
peer and administrative review and therefore does not necessarily reflect
the views of theAgencyand no official endorsement should be Inferred.
                                  ii

-------
                                 FOREWORD
     The U.S. Environmental Protection Agency is charged by Congress with
protecting the Nation's Ian;!, air, and water systems.   Under a mandate of
national environmental laws, the agency strives to formulate and implement
actions leading to a compatible balance between human  activities and the
ability of natural systems to support and nurture life.   The Clean Water
Act, the Safe Drinking Water Act, and the Toxics Substances Control  Act
are three of the major congressional laws that provide the framework for re-
storing and maintaining the integrity of our Nation's  water, for preserving
and enhancing the water we drink, and for protecting the environment from
toxic substances.  These laws direct the EPA to perform  research to
define our environmental problems, measure the impacts,  and search for
solutions.

     The Water Engineering Research Laboratory is that component of EPA's
Research and Development program concerned with preventing, treating,  and
managing municipal and industrial wastewater discharges; establishing  prac-
tices to control and remove contaminants from drinking water and to prevent
its deterioration during storage and distribution; and assessing the nature
and controllability of releases of toxic substances to the air, v/ater, and
land from manufacturing processes and subsequent product uses.  This publica-
tion is one of the iroducts of that research and provides a vital  communica-
tion link between t!>e researcher and the user community.

     The report concerns composting as a sewage sludge treatment technology.
Unlike the familiar, informal, small-scale "backyard"  composting of leaves,
grass clippings, and other plant remains, municipal-scale composting poses
significant, problems in facility design and control.  By following scientific
and technical principles, as developed herein, the public acceptability and
cost-effectiveness of such facilities can be greatly improved.
                                       Francis T.  Mayo, Director
                                       Water Engineering Research Laboratory
                                   iii

-------
                             ABSTRACT


      The  main  determinant  of composting  process  performance  is  de-
 composition  rate.   This  rate is  negatively affected  by temperature
 exceeding 60°G,  owing  to the inactivation  of  the responsible micro-
 bial  community.   Nonetheless,  composting masses  typically  self-heat
 to  80°C,  at  which point  the  rate of  decomposition is  low.  The  de-
 sired temperature ceiling  providing  a  high rate  can  be maintained
 through on-demand ventilative  heat, removal (thermostatic control of
 a blower).   This  constitutes,  in essence,  the  Rutgers  strategy  for
 composting process design  and  control.

      This strategy,  compared to  the  approach  currently in  widespread
 use,  yields  high  rate  composting in  that approximately 4X  more  waste
 is  decomposed  in  half  the  time.   Although  the  strategy can be im-
 plemented in certain enclosed  (in-vessel)  configurations,  the sug-
 gested unenclosed static pile  configuration is advantageous  in  being
 structurally and  operationally simple, and capital non-intensive.

      The  rational  for  this strategy  may  be expressed  symbolically as
 follows :
where:   0  a heat removed through vaporization  (energy/time)

        0.9 a approximate proportion of total heat removed  through
              vaporization

          m B dry air mass flow  (mass dry air/time)

       h    ° enthalpy of outlet air (energy/mass dry air)

        h.  •= enthalpy of inlet air  (energy/aass dry air)


The goal is to maximize Ov , because this is functionally equivalent
to maximizing the rate of waste  (e.g. sewage sludge) decomposition,
Realistically, this can only be approached through manipulation of
m (^ time-variable adjustment of ventilation rate) such that the
value of hp. t corresponds to a temperature of 60°C or less.  This
translates to temperature feedback (thermostatic) control of a
blower.

     This report was submitted in fulfillment of Contract No.
R806829010 by Rutgers University under the sponsorship of the U.S.
Environmental Protection Agency.  This  report covers the period
10/8/79 to 11/7/81, and work completed as of 11/7/81.
                                IV

-------
                               CONTENTS

 foreword	,	iii
 Abstract	„	iv
 Acknowledgments	  .  vii
     1.   Background	    1
     2.   Materials  and Methods		6
     3.   Comparison Between the Rutgers and Beltsville Control
         Strategies:   Observation and Interpretation of System
         Behavior	.	   22
     4.   Sequence of Limi*\tions  Induced by Control  Strategy.  ...  57
     5.   Rutgers  Strategy:   Replacement of ' odchips   with  Re-
         cycled  Compost as  the  Bulking Agent	  65
     6.   Comparison Between the Rutgers Strategy and  Beltsville
         Process:   Materials  Balance	91
     7.   Mathematical  Description o£  Process Control  Dynamics.  .  .   96
     8.   Drying Associated  with Composting,  and  Non-Biological
         Air Drying		101
     9.   Comparison Between the Rutgers and  Beltsville Strategies:
         Effect on  Curing Stage	116
   10.   Rutgers  Control Strategy:  Diagnosis  of Processing
         Failure.  „„,•,...„	„	122
   11.   Pathogen Inactivation.  ....  	  o  .....   132
   12.   Uniform  Provision  of Air Along the  Length of  a  Compost-
         ing Pile.  ........................  148
Conclusions.	155
Recommendations	,	1S6
References.	 157
Appendices
    A-l.  Temperature  Observations for  Piles  7, 8, 9A and 9B. . . 162
    A-2.  Oxygen Observations  for Piles 7 and 8.  ........  204
    A-3.  Moisture Content in  th* "Whole  Sample" and  the "Non-
          Woodchip Fraction".  .................. 209

-------
A-4.  Representativeness of Pile 9B	213
A-S.  Forces Driving Vaporization: Metabolically Generated
      Heat, and Unsaturation of Inlet Air	21S
A-6.  Relative Sensitivity of the Moisture Content and
      Volatile Solids Tests. .  .  .	216
B.    Advantages of Fuel Production Through Composting vs.
      Direct Combustion of Sewage Sludge Cake. .......  217
C.    Temperature Observations  for Piles 11A, 11B, and 11C. . 218
D-l.  Air Needed to Remove Heat and Supply Oxygen. .....  236
D-2   Unsuccessful Attempt by the Beltsville Group to Im-
      prove Drying, in Isolation From Considerations of
      Process Dynamics. ............ 	 . . 236
E.    Temperature Observations  for Pil= 12	237
F.    Observations on Blower Operation, Temperature, 0-f
      C0~, pH, and Moisture Content for Piles 6A,  6B,
      and 6C. ................ 	 241
G.    Temperature Observations  for Pile 13	266

-------
                          ACKNOWLEDGMENTS
     This investigation was funded by  the U.S. Environmental  Pro-
tection Agency under Contract No. RS06829010  (Dr. Atal  E.  Eralp,
Project Officer).  Support was also provided  by  State funds from the
New Jersey Agricultural Experiment Station  (publication No. H-07472-
1-84) .

     Thanks are expressed to the field crew at the Caraden  County
(New Jersey) Municipal Utilities Authority Sewage Sludge Co- posting
Facility (Horace T. Banks, Foreman), for their cooperation in '-Act-
ing up  the field trials.

     We greatefully acknowledge Dr. Peter F.  Strom for  many dis-
cussions on the intricacies of composting dynamics.
                              VI1

-------
                               SECTION 1

                               BACKGROUND
SLUDGE MANAGEMENT GOALS
     Composting can serve as a sludge treatment technology because it
advances the goals of decomposing putrescible  (odor-causing) material,
decreasing volume, weight, and water content,  inactivating pathogenic
organisms, and producing a stabilized process  residue.  This residue
is more easily stored, transported, and disposed of than the sludge.
Thus, composting is a treatment option within  an overall sludge
management plan.

     A minimum accomplishment of composting would be to convert a
sludge which is not acceptable for disposal at a sanitary landfill to
a process residue of lesser amount, which is acceptable.  More desir-
ably, a use would be found for the residue.  The traditional use is
as a compost (organic soil amendment) in gardening and horticulture,
and this might be extended to reclamation of disturbed land and the
preparation of 'j. and fill cover material.   A novel possibility, result-
ing from composting's capacity to remove water, is to use the residue
as a waste-derived, low grade, solid fuel,

COST-EFFECTIVENESS AND PUBLIC ACCEPTABILITY

     Even if a use is found for the process residue its monetary value
will, in all likelihocu, be small relative to  capital and operating
costs.   Consequently, cost-effectiveness does  not reside primarily in
product sales, but rather in economical  construction and operation.
Thus, the f-.cility should not be viewed, for example, as a compost
factory, but rather as a sludge management center.   This leads to a
process, rather than a product, orientation (1).

     In addition to being cost-effective, the  facility must be publicly
acceptable in terms of aesthetics.   This requires an absence of odor
nuisance, insect breeding, rodent harborage, etc..   Such nuisances
often havs public health overtones.

IMPORTANCE OF DECOMPOSITION RATE

     Both public acceptability and cost-effectiveness hinge on the rate
of decomposition.   The basic preventative measure against odor produc-
tion is to speed the decomposition of the putrescible organic waste.
Similarly, a high rate of decomposition  is  consistent with lov; capital
and operating costs.  This is because a  nigh rate results in a need for
less facility time/space and structural  appurtenances to achieve a

-------
 given degree  of stabilisation,  less  material  to  be  handled,  stored,  and
 transported,  and improved bandleability.   It  is  necessary,  therefore,
 to identify the operative rate-limiting factors.

 OVERRIDING RATE LIMITING FACTOR:   INTERACTION BETWEEN  HEAT  GENERATION
 AND TEMPERATURE

       With the  partial  exception  of  pathogen  inactivation (see  Section
 11),  the  aforementioned sludge  management  goals  (decomposition  of
 patrescible material,  decreasing  the volume,  weight, and  i-ater  content,
 producing a stabilized  residue) may  be  equated with  the  generation of
 heat.  _This is  because  the  heat is generated  microbially  through de-
 composition of  the  waste;  hence,  heat generation  is  functionally
 equivalent to waste treatment.  In understanding  the physical,  chemical,
 and biological  dynamics which govern the composting  system,  it  is use-
 ful to focus on heat  generation.   In particular,  the overriding determi-
 nant  of system  behavior is  an interaction  between heat generation and
 temperature (2-3).

       Consider  the  behavior  of  a  non-managed  pile of organic material
 having the following  characteristics:   it  is  large enough to be self-
 insulating; the  material  is  moist and nutritionally supportive  of micro-
 bial  growth; the material  is sufficiently  porous  to allow gas exchange;
 the exchange suffices to  prevent  gross  oxygen depletion in the  inter-
 stitial atmosphere.   Such  a  pile  spontaneously increases  in temperature
 because,  for a  period of  time,  the rate of microbial heat generation
 within the pile  exceeds  the  rate  of  h",at loss  to  the surroundings,  This,
 "self-heating"  phenomenon  is the  basis  of  the  composting process.

       At  the outset of  self-heating  a positive feedback  loop becomes
 established between microbial heat generation  and temperature,  in that
 higher  temperatures favor microbial  growth with its associated metabolic
 generation of heat.  When the temperature  begins to exceed approximately
 38  C the  feedback turns negative, because  higher levels are progress-
 ively  unfavorable to mesophilic*  growth and activity.  This slows the
 temperature ascent  and would, in  the absence  of subsequent events,
 terminate  it at approximately SO°C.

      The  temperature ascent is renewed, however, with the initiation of
thermophilic* growth, starting at approximately 4S°C.  This reestabli-
shes positive feedback between heat  generation and temperature.   Since
the thermophilic community in self-hsating organic masses is most active
at approximately SS°C, the feedback  starts to become negative when the
temperature exceeds this value.   The  temperature ascent again slows,
typically peaking at 80°C.  At this  temperature, heat generation is
slight.


 Mesophile - an organism living in the temperature range  around  that  of
 warm-blooded -animals; thermophile =  an  organism living at high  tempera-
 ture  (T.D. Brock,  D.W.  Smith,  M.T. Madigan.   Biology of  Microorga-
 nisms, 4th Edition, Prentice-Hall,  Inc.,  1984).  A competent  array of
 both types is  reliably present  in organic  wastes.

-------
      To  realize  a  high  rate  of  decomposition  the  system's  tendency to
 self-limit  via  inhibitively  high  temperature  must be  countered.   A means
 of  doing so is  to  match ventilative  heat  removal  to heat  generation in
 reference to a  60°C  ceiling,  as accomplished  through  temperature  feed-
 back  control of  a  blower.  This development,  which paved  the  way  for rhe
 present  work, was  described  earlier  (2-7).  It  constitutes  the  approach
 to  composting process control known  as  the  Rutgers strategy.

 CONFIGURATION AND  STRATEGY:   TWO  LEVELS OF  COMPOSTING PROCESS ORGANIZATION

      Two levels  of composting process organization can be  recognized.
 Strategy lies at the conceptual level,  in that  it represents  a  plan for
 guiding  the interacting physical,  chemical, and biological  events.   Im-
 plementation of  a  strategy is through some  set  of physical  and  mechanical
 elements, including  geometry  of the  composting  mass,  equipment  for venti-
 lation,  ar:d machinery for material handling.  The elements  at this level
 of  process  organization are  collectively  termed configuration.  The dis-
 tinction between strategy and configuration is  basic  to this  investiga-
 tion.

      The configuration  used was that of unenclosed static  pile  —  meaning
 that  the material  was in the open, was  ver.f.Hated by  blower,  rind  wa:i not
 agitated mechanically during  composting.

 SIGNIFICANCE OF  THE RUTGERS-BELTSVILLE  COMPARISON

      An  essential  feature of this  report  is a comparison between  tht
 Rutgers  strategy and the strategy  embodied  in the Beltsville  Static Pile
 Process.  These  represent fundamentally different approaches  to the man-
 agement  of  the composting microbial  ecosystem (TABLE  1).

      The  Beltsville Process was developed specifically for  the  treatment
 of  sewage sludge (8-11), and is in widespread use (12-13).  This  process
 is  advantageous  in its  structural  and operational  simplicity, owing  to
 the static  pile  configuration.  Like most composting  systems, however,
 it  suffers  from  slow decomposition as a result  of inhibitively high tem-
 perature.   By employing  the Rutgers  strategy, however, in static  pile
 configuration the  structural and operational simplicity can  be retained,
 while benefitting  from  rapid decomposition.

      The difference in  behavior induced by the  two strategies originates
 in  the management  of ventilation.   The  Rutgers  strategy focuses on  heat
 removal  for  temperature  control, whereas  the Eeltsville Process focuses
 on  the maintenance of an oxygenated condition.  These  operational  ob-
 jectives are met through the respective approaches taken to blower  con-
 trol,  sizing, and operation mode.   The  physical,  chemical,  and biological
 dynamics governing the  composting system,  however, dictate  certain  con-
 sequences that might not be immediately apparent.   By  focusing on  tem-
perature in this manner  an abundance of  oxygen is  automatically provided,
whereas by  focusing on  oxygen inhibitively high temperatures  result.
 These  principles pertain to composting  in general, regardless of  the
particular  type  of waste being treated,  the configuration employed, or
 the name or proprietary  status of the process.

-------
TABLE 1.
FUNDAMENTALLY DIFFERENT COMPOSTING PROCESS CONTROL STRATEGIES
                      	Process control strategy	
Process control operational
     objective
                                  Rutgers
                      Maintain 60°C
                      temp ceiling
                                            Beltsville
Maintain 0, at
S% to 15%
Blower control
Blower sizing
                      Fixed schedule
                      initially,  fol-
                      lowed by temp
                      feedback

                      Must meet peak
                      demand for heat
                      removal
Fixed schedule
throughout
Prescribed as 1/3 hp
per 50 ton pile
Blower o elation mode
Consequences o£ strategy
                      Forced-pressure

                      System oxygena-
                      ted;  a high rate
                      of  heat genera-
                      tion   and  vapo-
                      riza.tion;  dry-
                      ness  may come
                      to  inhibit acti-
                      vity,  tailess pre-
                      vented; good
                      pathogen kill.
Vacuum- induced

System oxygenated;
temp peaks, by de-
fault, at an inhi-
bitively high level
                                                         CT ;  a low rate
                                                    of heat  generation
                                                    and vaporization;
                                                    good pathogen kill.
 Both strategies were implemented in unenclosed static pile
 configuration.
 Heat generation is equivalent to decomposition.

-------
POSITION OF THE PRESENT REPORT

     Most of the essential findings reported herein have already been
published (1-7, 14-26) or are in an advanced stage of publication (27-
30,.  These papers represent different facets of a coherent line of re-
search, but are scattered.  The present Report permits a more integra-
tive treatment in a single volume, and provides an opportunity to docu-
ment the data in full detail.

     Finally, our basic view on composting process design and control,
first enunciated in detail in 1980 (4), has received independent con-
firmation by at least four groups.   Two of the studies were sponsored
by EPA (31-33) and two were conducted in European countries (34-36).

-------
                              SECTION 7.

                        MATERIALS  AND METHODS


      Materials  and methods  generally employed  in the  investigation  are
described  in  the present  section.   Modifications  and  additional pro-
cedures  pertaining  to  particular  studies  are  described in the  indivi-
dual  sections.

      Site.   Our pilot-scale composting research  frcility was  located
at  the~~CTamden County Municipal  Utilities  Authority  sewage treatment
plant at Second  Avenue and Jackson  Street,  Camden,  N.J. ,  where  the
Authority  routinely composted the sludge  from their Jackson Street
and Baldwin Run  treatment plants  (total 18-23 dry tonnes/day)  (20-
25  tons/day)  by  means  of  the Beltsville static  pile process (8-11).
Although sludge,  woodchips,  and supportive  services were  provided by
the Authority, t'?.s  investigation  reported herein was  independently
designed and  executed.

      At the  pilot research  facility (Figure  1) a shed housed  the in-
strumentation and control systems (Figure 2).   The  shed was heated and
ventilated to prevent  temperature extremest

      SJLudgjs.  The sludge consisted of a  mixture from  two sources.
Approximately 90% of the material came from the Jackson Street  plant,
where sewage  treatment  consists solely of primary settling.  The re-
mainder  came  from Baldwin Run,  where "partial digestion"  was accom-
plished  in an Imhoff tank.   Thus, the experimental  material was
essentially raw  fresh  sludge.

      The Jackson Street plant  served a mixed residential-industrial
area, where a food processing plant  was the largest industrial  contri-
butor.   The sludge was  preconditioned with  the  addition of approxi-
mately 1 kg (2.2 pounds) of  polymer  (Allied Colloid Percal  728, a
chloride based cationic polyeletrolyte) per metric  tonne  (1.1 ton) dry
solids,  followed by dewatering  through a  belt filter press.  This
generally yielded a cake of  approximately 25% solids  (oven dry  weight
expressed on  a wet weight basis).   Analyses of  the  sludge are avail-
able  (37).

      Woudchips.   Virgin woodchips  from hard  and soft  woods having
nominaT~3T5ieiTsTons of  2.5 x  2. S x 0.6 cm  (1 x 1 x 1/4  in.)  were used
to form  a base for the  composting pile, as  a  bulking agent  for  mixing
with the sludge,  and to form  a  cover  over the composting  pile.

      Mixture.  Sludge  cake  (by wer  weight)  and woodchips  (by volume)
were mTxed in a ratio-of approximately 1  tonne  sludge  per 1.7 m-5 wood-
chips (1 ton  per 2 yd  ).  Mixing was .done in  an industrial  pug  mill

-------
                               0 LOWERS
    VAN
    WITH
           f
          •63
          '&',
            *
C C M U A
OPi R ATIONS
AREA
Figure I.    Layout of  the  Department  of  Environmental  Science,
    Rutgers University,  research  facility  at  the  Camden County
    Municipal Utilities  Authority treatment plant.   A  typical
    layout for three 6-  metric  tonne  piles is illustrated.

-------
   i
Figure 2.    Shed  interior  showing  instrumentation.   Photo  by
    F.C. Miller,                                            z

-------
 (McLanahan  Corp.,  Hollidaysburg,  PA).   The  mixture  was discharged to
 a dump  truck  for  transfer  to  the  pilot  facility,  300  meters distant
 (980  feet).

      Mechanical  ventilation.   Corrugated  f'lexhose  (Agway Inc. ,
 EnglisFtoT7nT~N77.J~wTtTrT~HTameter  of  10.2  cm (4  in,)  was used  for
 ventilation ductwork.   This was nonperforated,  or perforated,  as
 required,  A  radial  blade  blower  (Dayton,  Inc.,  Chicago,  2C820) with
 a 1/3 hp motor  (Dayton  #SK586a) provided ventilation.   Unless noted
 otherwise,  the  ductwork was fitted  to  the  outlet  end  of the bloi-er,
 thus  providing  ventilation in  the forced pressure direction (7).
                          .       f^rst:  steP  in  pile  construction was
to form~a~EaTe~bf woodcETps,  in  which  was buried  one  cr  more  lengths
of perforated flexhose  (Figure 3],   This was connected to  the blower
via a length of non-perforated flexhose.  Over the  base  was placed
the sludge-woodchip mixture,  or  the  sludge-recycled compost mixture,
which was covered with  woodchips (Figure 4).   The piles  were  parabolic
in cross section, a slight  longitudinal axis parallel to the  flexhose
in the smaller piles, and a distinct axis in the  larger  piles (see
Figures 3 and 4).

      Blower coiitrg_l -  Rutgers strategy.  The  Rutgers strategy (2-7)
Uitiliz"e~d~aT temp^Fature-actuated~~5Towercontrol system consisting of  a
temperature controller  with an adjustable set  point (Fenws.ll  Inc.,
Ashland, MA., Series 551),  and a thermistor  (Fenwall  28-232306-304)  in
the pile (Figures 5 and  6).   The  controller continuously received and
interpreted a signal from the thermistor.  To  dampen response,  the
thermistor was shielded  by  threading it into a short length of I  in.
NPT iron pipe.  When the signal  indicated a temperature  less  than the
set point the controller actuated  the  blower through the timer,  as
scheduled.  When t-he temperature was above the set  point the  controller
directly actuated the blower  until ventilation decreased the  pile
temperature (in the vicinity  of  the  thermistor) to  less  than  the
set ]  'int.  Control is  thus "on  demand," based on the feedback  of
a temperature signal.

      Monitoring - ^utc^atj^c.  Temperature was determined  with  thermo-
coupJ eT~macle o± 20 gauge copper-constantan wire (Thermoelectric  Corp.,
Saddle Brook, N.J.).  Thermocouples were taped to wooden dowels  at
0.3m (1 foot) intervals, and  the dowel  vere inserted horizontally into
the pile to establish a vertical 0.3 x J.3m grid  (Figure 7).   The grid
was perpendicular to the long ax^s of  the flexhose.  The grid was in
the center of the pile, unless noted otherwise.   (For photographic
examples see Figures 44, 45 and  46.)

      Thermocouple Isads were; organized through the use  of subjunction
boxes  at each pile, through a main junction box? and finally  were wired
to a recording monitor  (Doric Scientific, San  Diego, CA.  ,  Digitrend
220).   The temperature reading was logged every hour.

      Gas  sampling probes (Figure  8) were inserted  into  the pile.  The
probes led,  via 0.64 cm (j in.) I.D.  Tygon tubing, to  a condensate trap

-------
Photo by F.C.  Miller,
                                         f woodohip
                  10

-------
             l^c _r    I T
             U>k f*. ./, ' •
                                   :A
                                   r--.
                        P    \
                            if
                            ,*' .
^•.if   r^    ^-
1 it-i fj-  !  -    IE , 3f I \' ,1 .

     W
'* ! ((/ I '
<• \Vv->}//
               i^
• ii: -^
                                     cd
                     ^  \ '  i«ix
                          -x   '  •  4 i  ^
,[\l_'   t>
                      ,^i-      • ; '   /
                   ,. -  J , T"'«"-M''-v?^ ^^
                    "k^yr •'./,    'v%; V*"
                     ^(-f   * J^^_ . -  ^ ^T ' >^ , -
    '< „ •-  -  i ,.   . , - ;    •,„ vo-  ---'^'
  •rf * "-  r , * ^- •" •& . -; '^n:Kr' ,„ • -•"li-
   ' ^/% r'  ' ^ _ « ^  ^  - ,, *,„ -  >      ^
  •     '•" ,f    ^ < '•>   ', >.:->•*'  ,. ^i
                       "              "
Figure 4    Typical pile construction,  placement of sludge-
   P?C Miller   " °VSr thS W°°dchiP  base"  Ph°to by
                         11

-------
                                                     «•	THERMtS TORI-—*
                T <  T(SET)
Figure 5.   Blower  control logic flow diagram.

-------
                           . O
                            u
                                     ORMAL
                                                          EVENT RECORDER
        r
       HOI
       IN
            r
          NlUT
           IN
GRNO
 IN
           FHERMISTOR
TEMPERATURE

CONTROLLER



  H   N   G
                HIOH
                HOI
      ,   r
                                                       -*•-  TO BLOWER
                                             10
                                                             WIRING
                                                             TERMINAL
                                                             STRIP
                                                          -*»-TO PILE
                                               TIMER
                                              NO.
                                                 N  H N
Figure  6.   Blower control wiring schematic,
                                  13

-------
                                         j:r^
                                              -*l
                     \ '-
               »•-
Figure 7.     Insertion of a dowel (pile 11C).   Fixed to the dowel
    was the thermistor,  and 5 thermocouples at 0.3 m intervals.
    The personnel were,  from left to right, Messrs. Psarianos,
    MacGregor,  Miller and Finstein.   Photo by F.C. Miller.
                             14

-------
•END CAP
                                                    TO SAMPLER
                        V   "
                          \70  L
ENGTH  PROBE
 PIPE NOTES

   l" I D  S.S. OR PLATED  MILD STEEL
   4
 Figure 8.   Construction  of  the gas sampling probe.   The
     dimensions are in inches,  based on National Pipe  Thread

     (NPT) standards.
                                15

-------
 (2L vacuum  flask) packed with  glass wool,  and  thence  to  a  seven-point
multiplexing  system  (Figures 9,  10, 11  and 12).   The  multiplexing  system
sequentially  interrogated  up to  six different  pneumatic  inputs  using a
sample pump of nominal  l.QL/rain  flow.   A nominal  5L/min  purge pump main-
tained flow in the pneumatic lines that were not  being interrogated.
The seventh sample point was used as  an ambient system purge between
interrogations.  Thirty minute long sample cycles were initiated every
four hours  (Figure 13) .

      The gas sample collected by the multiplexer was split for passage
through an infrared  C02 Analyzer  (Beckman  model 865)  and a paramagnetic
oxygen analyzer  (BeckmSn model 755).  The  multiplexer, analyzer, and
dual channel  stripchart recorder  (Beckman  model 8720A) were housed in a
temperature-controlled  cabinet.

      Monitoring - manual .  A clamshell type pesthole digger having a
working diameter of  12.7 cm  (5 in.) was used to obtain samples.  The
sample material was put into a plastic  bag for transport to the labora-
tory.  Determinations were initiated  upon  return to the  laboratory,
usually within 4 hrs of sampling.

      Laboratory de t e rm inat i ons .  Independent "whole sample" and "non-
woodchip fraction" moisture content determinations were  usually perform-
ed.  The separation was accomplished  by removing non-woodchip material
from the woodchips by hand.  Except for Section 4 in which, the data re-
fer to the non-woodchip fraction, the data in the body of the report re-
fer to the whole sample.  A comparison  of  the moisture contents of the
two fractions is given in Appendix A- 3.

      To determine moisture content,  the material was dried at 104°C to
constant weight.   Moisture content is expressed on a wet weight basis
                 'C% moisture =  wet                     * «0)
      To prepare sample material for the pH determination, distilled
water was added to sieved material to make a slurry.  The determination
was by pH electrode.
                                   16

-------
     TO GAS
    ANALYZER
        o
  Pi
                            EXHAUST
S A M P I E
No I
                                          SAMPLE
                                           No 3
                                        A I R
Figure 9.
      The
      no.
 Gas multiplexing system, simplified pneumatic  flow di
actual multiplexer consisted of six sampling points.
4 represents the ambient system purge.
                                                 agram.
                                                 Valve
                   17

-------
                                           TO PURGE AND
                                           SAMPLE PUM^S
n
      TO SAMPLE INITIATER
               SAMPLE
               CYCLE
              IH TERMINATION
               TIMER
         SAMPLE
         TIMER
         No.]
         HN H G
         SAMPLE
         TIMER
         No. 2
         HN H G
         TIMER
         No. 3
         HN H G

LE
R
G
r


LE
R
G
}


LE
R
G
r










































N
J
r
I™


r~

P—
^
Figure  10.  Gas multiplexing system,  simplifipd  electrical
    schematic.   The actual  multiplexer consisted of six  sampling
    points.  Valve no.  4  represents  the ambient  system purge.

-------
  LEADS  FROM
  SAMPLE TIMERS
                                                     TO PURGE VALVE
Figure 11.   Fabrication of slave relay  (see  Figure 10).  Parts
    list:   Diodes (3), WER-17; resistor,  IK  10%  (76131RC PW1C);
    capacitor,  SPRAGUE TE-1509 20-150DC  USA-7317H;  relay, POTTER
    &  BRUNFIELD KA501 110 VDC.
                              19

-------
                            CYCLE

                            INITIATING
                            TIMER
                                                    POWER
                                                    IN
Figure 12.   Gas  sampling cycle  actuation system,
                                20

-------
         CYCLE INITIATION TIMER

         CYCLE CONTROL TIMER

         CYCLE CONTROL RELAY

         VALVE - SAMPLE No I

         VALVE-SAMPLE No 2

         VALVE-SAMPLE NO 3

         VALVE- PURGE

            t - doiod
            °-open
                                           8   10   12  14   16   18   20  22  24   26   26   30
                                                     TIME IN MINUTES
Figure  15.     
-------
                              SECTION  3

  COMPARISON BETWEEN THE RUTGERS AT)  BELTSVILLE  CONTROL  STRATEGIES:
         OBSERVATION AND INTERPRET .TION OF  SYSTEM BEHAVIOR
IMPORTANCE OF THE COMPARISON

     Comparative study of the Rutgers  (2-7) and Beltsville  (8-11)  strat-
egies has universal relevance to composting process design  and  control.
This is because i, composting processes are based on the self-heating
microbial ecosystem, ii. these strategies represent fundamentally  differ
ent approaches to the management o£ this ecosystem, iii. the comparative
analysis yields a coherent theory of the physical, chemical, and biolog-
ical interactions governing the dynamics of this ecosystem.

     The basic experimental design was to isolate strategy  as the  only
variable, by implementing both strategies in a common configuration
(static pile).  This permitted orderly interpretation of the data.  The
distinction between strategy and configuration was already  drawn.  (Sec-:-
tion 1) .

MATERIALS AND METHODS SPECIFIC TO SECTION 3

Ventilation

     For pile 9B (Beltsville Process) the blower was operated in the
vacuum- induced direction, solely as scheduled by timer.  A  condensate
trap v/as interposed between this blower and its piles (Figure 14} . For
piles 7, 8,  and 9A (Rutgers strategy) the blowers were operated in the
forced-pressure direction, as actuated by the temperature feedback con-
trol system.
    The physical features of the piles and the ambient weather conditions
are summarized in TABLE 2,  Variations in the control of the ventilation
system were as follows:
                          hr 500, 1.75/15
                          hr 500, 1.25/15
                          hr 120, 1.5/15; hr 120 - hr 240, 1.25/15;
                          hr 240 - 500, 0/1S
     Pile 9B:  time zero - hr 70, 4/15; hr 70 - hr 50J, 3/15
Pile 7:  time zero
Pile 8:  time zero
Pile 9A: time zero
                                 22

-------
                                    55 GALLON  DRUM
  WATER
  OUT
               fER
                 vs

                                                       EXHAUST
                                                       THRU
                                                       BLOWER
                                                  PUMP
                                                  WITH
                                                  FLQAT
                                                  SWITCH
Figure  14.   Condensate trap used for  the pile  (9B) managed
    according to the Beltsville process.
                            23

-------
      Pile 7:'  time zero - hr 56, position 1; hr 56-500, position 6
      Pile 8:   time zero - hr 500, position 6
      Pile 9A:  time zero - hr 44, position 1; hr 44-500, position 6
      Pile 9B:  not applicable

                                             j£jHE?JlilHZ£_££Bl££ii£Zl

      Pile 7:   time zero - hr 500, 45°C
      Pile 8:   time zero - hr 500, 4S°C
      Pile 9A:  time zero - hr 76, 4S°C; hr 76 -  hr 500,  48°C
      Pile 9B:  not applicable

      The cross-sectional representation of the  piles  are given in Fig-
 ures  15, 16,  17  and 18.


 RESULTS

      Three  of  the piles were controlled according to  the Rutgers  strategy
 (piles 7,  8, and 9A),  and one according to the  Beltsville Process (pile
 9B).  The  pile weights were:   pile  7,  6 tonnes;  pile  8,  27 tonnes;  pile
 9A, 36 tonnes; pile 9B, 36 tonnes.   Piles  9A and 9B  (Figure 19) were
 assembled within two days  of each other,  to  provide a direct comparison
 of  the alternative approaches to process  control.

 Blower Operation

      Blov/er operation  is  represented in Figures  20, 21,  22,  and 23.
 Temperature feedback control  (Rutgers  piles  only)  commenced as follows:
 pile  7,  hr  56; pile  8,  hr  10;  pile  9*\,  hr  10.  The maximum blower opera-
 tion  (%  time on)  and the  time at which  the maximum operation occurred
 was,  respectively:   601,  hr  110;  40%,  hr  100, 100%, hr  96  to 135.   In
 pile  7 the period  of feedback control  ended  at hr  344,  at  which time "de-
 mand" for the  blower wat  less  than  that scheduled  by  the  timer.   There-
 after, timer-scheduled  operation resumed.  In pile 8  the  return to  sch-
 edule operation  occurred  at  hr 170.  The timer to pile  9A  was disconnect-
 ed at hr 240,  eliminating  the  resumption of  timer-scheduled  blower  opera-
 tion when feedback  control ended.   Judging from  the dwindling amount of
 blov/er operation  time,  timer  scheduled  operation  for  this  pile would
 have resumed prior  to hr 380.

     The blower  to  the  Beltsville pile  was scheduled  for  operation  27%
 of the time early  in the run  (Figure 23).  At hr 70 this was decreased
 to 201,  as the 02 content  exceeded  that suggested for this process  (9).

 Pile Temperature

     In  the Rutgers piles  the median values were a few degrees higher
 than the  assigned set point  (e.g. for pile 7, 47°C vs 45°C) , and  only 10%
 to 13% of the  individual observations exceeded 60°C.   In the Beltsville
pile the  median was 70°C, and 91% of the individual observations  exceed-
ed 60°C.  The observations on temperature and other parameters are summar-
 ized in  TABLE  3.   The individual  temperature graphs are  given in Appendix


                                 24

-------
          TABLE  2.   PILE  DESCRIPTION  -  PHYSICAL  ASPECTS  AND WEATHER  CONDITIONS
Item Pile 7
Trial period
Start
End
Ambient air , j.
temperature (°C) '
High
Low
Mean
Range 2
Rainfall
Amt (cm)
Events (no.)
Pile dimensions in meters
(Woodchip base) 3.0 x

21 Apr
12 May

21
9
15
to 31

9.3
10
(LxWxH)
2.7 x 0.2
(Overall1) 4.9 x 4.3 x 2.0
Pile weight (tonnes)
Perforated flexhose
Segments (no.)
Lengths (m)
Blowers (no.)
6

1
2.7
1
Pile 8

27 May
9 June

27
15
21
10 to 31

3.3
6

13x2.1x0.2
15 x 4.3 x 2.0
27

6
5.6
6
Pile 9A

9 Jul
1 Aug

32
21
27
16 to 36

13.1
7

13 x 2.4 x 0.2
15x4.9x2.1
36

6
5.6
6
Pile 98

11 Jul
1 Aug

32
21
27
16 to 36

10.0
&

14 x 5.5 x 0.2
14 x 5.2 x 2. 1
36

2
12.2
1
Direction of
  ventilationi

Basis of
  control§
Forced pressure  Forced pressure  Forced pressure

Temp feedback    Temp feedback    Temp feedback
Vacuum induced

Timer schedule
*The dates refer to 1980.   tHigh = mean of the daily highs:  low = mean of the  daily lows;
 mean = mean of the daily means; range = overall  range of the daily highs and  lows.
 ^Dimensions before adding a covering of woodchips  25 to 30  era thick.   §  Forced pressure
 and temp feedback = Rutgers strategy; vacuum-induced and timer schedule  = Beltsville
 process.

-------
Figure 15.    Pile 7, cross-sectional representation:  textured area, woodchip cover
    and base; clear area, sludge-woodchip mixture; circle, perforated flexhose.  The
    numbers indicate monitoring and control positions: thermocouples, positions 1
    through 16; gas sampling probes, adjacent to positions 6 and 14; control ther-
    mistor, position 6 for most of the run (see Table III-3).   The blower was
    operated in the forced-pressure mode.

-------
                                LONGITUDINAL  SECTION
                              ^
                               I 8
                               I 5
                       26
                                      27
                                             28
                                             20
                                                     29
                                                             30
                                                                    31
              MAIM GR!D
                               'MAIN
                                GRID
AUXILIARY
GRID
            AUXILLARY GRID
Figure 16.    Pile 8,  cross-sectional representation:   textured area,  woodchip cover
     and  b; se;  clear area, sludge-woodchip mixture;  circles,  perforated flexhoses.  The
     numbers  indicate  monitoring and control positions:  thermocouples, positions 1
     through  31;  gas sampling probes, adjacent to positions  11, 20 and 28; control
     thermistor,  position 6.  The circles and tubes  in  the woodchip base represent
     flexhoses.   The blowers were operated in the forced-pressure mode.

-------
tsj
CO
         Figure 17.   Pile  ~>Ar cross-sectional representation:   textured area,  woodchip  cover
             and base; clear area, sludge-woodchip mixture; circles, perforated flexhoses.   The
             numbers indicate monitoring and control positions:  thermocouples,  positions 1
             through 19; gas sampling probes, adjacent to positions  7 and  16;  control
             thermistor position 6 for most of the run  (see text).  The thermocouple
             grid was positioned 6m from an end of the pile.  The blowers  were  operated  in  the
             forced-pressure mode.

-------
Figure 18.    Pile 9B, cross-sectional representation:  textured area, woodchip cover
    and base; clear area, sludge-woodchip mixture; circles, perforated flexhoses.  The
    numbers indicate monitoring positions: thermocouples, positions 1 through 15; gas
    sampling probes, adjacent to positions 7 and 14.  The thermocouple grid was
    positioned 6m from the end of the pile proximal to the blower.  The blower was
    operated in the vacuum-induced mode and controlled solely by timer.

-------
    J...-L..,,--
Figure 19.   Pile 9B (left) and pile 9A (right).  Photo by
    James A. filler.
                           30

-------
    100-f
                     SOD
200            300
   TIME IN HOURS
                                                                400
                                            500
Figure 20.  Pile  7,  blower  operation.   The baseline represents operation  as
   scheduled by timer,  and  the  area above the baseline represents operation  through
   the temperature-feedback control system.

-------
     z
     o
     u ?5-{:
     &
     t-
     J_
     z
     iii
     o
     £50-
     z
     o
     i-
     <
     er
     ec
     LJ25-
         i?
r^J'
o
_!
S3
o-
1 1
1 !
j ,>
(
*«
> , h - "i
r^ 1 , » s
"1 r- ( " J-l
U ' 'V ' , . \_
9 , • ' *i' i, '., T-
-.1' • . -."X
•,; ' . , ' 	 (, "' 	 , 1
| * >• f ' ' r
Q 100 200 300
V
1
tr,
K;|
1
400 BOO
                                        TIME IN HOURS
Figure 21.    Pile 8, blower operation.   The baseline represents  operation as

     scheduled by timer, and the  area  above the baseline represents  operation through

     the temperature-feedback control  system.

-------
              lOO—i
w
w
               0-
,
I
0
f ',''• k
I
iOO

I
200
	 :.-... 	 ,.j, 	 ,.r,,,,,,.
l
300
I (U^-lF-l-1 ""T ^JTT7 Tniih-i-
i
400

!
50C
                                                TIME IN HOURS
         Figure 22.   Pile 9A, blower operation.   The baseline (hr zero to 10)  represents
             operation as scheduled by  timer,  and  the area above the baseline represents
             operation through the temperature-feedback control system.  The timer was
             disconnected at hr.  240.

-------
   100-f
                    100
200            300
   TIME IN HOURS
                                                                  400
500
Figure 23.    Pile  9B,  blower operation  as scheduled by  timer.

-------
         TABLE  3.   SUMMARY  OF  THE  DATA FROM SECTION .3
           17
Temperature
  Period  (hrs)
  Range  (°C)
  Median  (°C)
  Observations  (no.)
  % Observations
Interstitial gases
  Period  (hrs)
  02 Range (!)
  02 Median  (%)
  Observation  (no.)
  C02 Range  (%)
  C02 Median (i)
  Observations  (no.)
Moisture content  (%)
  Time zero  (%)
  Selected observation
                (hr)
                (%)
                                  Process  control  strate
56-344
26-68
47
1102
11
10-170
-19-70
48
728
10
10-380
24-68
53
175S
13
100-500
45-82
70
1519
91
                          0-388      0-268
                          14.S-21    6.0-21
                          20.0      18.5
                           1S2        204
                           62

                          344
                           29
0-500
8.0-21
19.8
246
0-9
1.0
246
0-500
12-21
20.0
340
0-7
<0-0.1
240
 56

190
 24
 67

310
 33
 65

500
 61
For piles 7, 8, and 9A the summary is for the period of  temperature
feedback control.  For pile 9B the summary excludes the  period  of
temperature come-up.
                                 35

-------
 A-l.

      In  the  Rutgers  piles  the  data  from the  innermost vertical series
 of  probes  (positioned generally  in  a  line  above  an air duct)  indicate
 the establishment  o£ a systematic temperature  gradient in the direction
 (upward) of  airflow  (Figures  24 ,  25,  and 26).   In the Beltsville  pile
 the lower  members  of the  innermost  series  (not directly in a  line above
 an  air duct-see  Figure 18) recorded  a  weak  gradient  in the direction
 (d.wnward) of  airflow (Figure  27).  A gradient is absent at the upper
 end of this  series (Figure  28),  which represents  the  pile apex.  A more
 well-defined gradient was  established at the adjacent vertical series of
 probes (Figure 29),  which was  directly  in  a  line  above a duct-see Figure
 18).  This gradient  was shallow  and erratic,  however, compared to those
 in  the Rutgers piles.   The  temperature  gradient patterns are  summarized
 in  Figure  30.

      In  the  Rutgers  piles the  temperature  at the  thermistor position
 slightly "overshot"  the set-point at  around  the time  that blower  control
 passed from  timer-schedule  to  temperature  feedback (i.e.  at the start of
 the  period of  feedback control) .  The relatively  erratic temperature re-
 cord  of  Pile 9A during this transition  is  attributed  to control changes
 at  this  time (see  Pile Construction and Control,  and  Weather).  Nonethe-
 less, starting at  approximately  hr  140,  in this pile  also the tempera-
 ture  at  the  terraistor position stabilized  at the  set  point-level.   After
 termination  of feedback control, and  with  the  resumption of tL.-er-sch-
 eduled ventilation (piles 7 and  0 only) , the temperature  declii ed rapid-
 ly.   Where scheduled ventilation was  not resumed . (pile 9A)  the temperature
 fluctuated slightly,  but no particular  trend set  in.
     All of the. piles. were monitored for 02 „ but only piles  9A and  9B
were also monitored for C02-  Since the three Rutgers piles  behaved
similarly with respect to 02, only the observations for Pile 9A, which
involved both gases, are reported here (Figures 31, 32, 33 and 34).  The
other 02 data are given in Appendix A* 2.

     In the Rutgers pile the minimum 02 content (upper position) was 8%
(hr 26) , and in the Beltsville pile (upper position) this was 12%  (hr
70) .   Minimum QI values at the lower probes were 15% and 19% (hrs 380
and 70, respectively).  The f>2 an^ ^2 plots are generally complimentary
in that relatively high values of 02 and low values of C02 occurred simul-
taneously,  and vice-versa.

Moisjture Content

     The moisture content data reported here are for the whole sample
(includes woodchips) .  For a comparison with the non-woodchip fraction,
see Appendix A- 3.

     The respective starting moisture contents were (%) :  62, 56, 67,
and 65  (Figures 35 and 36) .  In the three Rutgers piles drying was sub-
stantial in that by termination the moisture content decreased to low
levels  (29%, 24% and 29%).   Relative to piles 7 and 8, the onset of dry-
ing in pile 9A was delayed.  Once drying commenced, however, it progress -


                                  36

-------
      80
PROBE 7-iS
PROBE 7-14  	
PROBE 7-6
                     100
     200           300
        TIME IN HOURS
400
                                                                            5OO
Figure 24.    Pile 7, temperature  at the innermost  vertical series of  thermocouples.

-------
                             PROBE 8-18  	
                             PROBE 8-11  	
                             PROBE 8-6  °=
                             PROBE 8-1   	
  -S
J ,
0
1 •__„ 	 II „. _ 1 1 I 1 1
iOO 200 30O 400
TIME m HOURS
1
500
Figure 25.     Pile 8,  temperature at  the innermost vertical series of
    (main grid).
iherrnocouDles

-------
      80|
     oSO
     LJ
     £T
     O
     t-
     <
     CC
         ?
      20 -

            PROBE 9AI8
            PROBE 9A15
            PROBE 9AI!
            PROBE SAG
            PROBE 9AI
     ;•?».„/

\-«  '
                      !OO
                  200            300
                    TIME IN HOURS
400
                                                                              500
Figure  26.     Pile  9A, temperature at the  innermost vertical series  of thermocouples.

-------
       80 —
         '3	

         r
       2o!f~
                              PROBE 98-10
                              PROBE 9B-6
                              PROBE 9B-I
         B-
                      100
200           3OO
   TIME SN HOURS
                                                               400
                                                                             500
Figure 27.     Pile 9B, temperature at the lowest three thermocouples cf the innermos-i
    vertical  series.

-------
                             PROBE 9B-I5
                             PROBE 9B-I3
                      iOO
                     200           300
                        TIME IN HOURS
                                                              400
                                                              500
Figure 28.
    vertica1
 Pile 9B, temperature  at the highest two  thermocouples of the innermost
series.

-------
       80!
     
-------
       20


       10


        0
      -10
     u
       20
       10
     2
     LU

     ^
     LU
     U.
     U.
     LU
     SK
     
-------
 20
 UJ
 S
 D


 §16
2
UJ
U
o:
UJ
CL


-12
z
UJ
o
o

K
O

CE
UJ
          CO?
               100
                           200          300

                              TIME IN HOURS
                                                 400
500
Figure 31.    Pile 9A, concentrations of 02  (upper curve) and
                        The  gas sampling probe  was adjacent  to
CC>2  (lower  curve) .
position  7.
                                44

-------

                100
                           200          300
                             TIME IN HOURS
400
            500
Figure 32.    Pile 9A, concentrations of 02 (upper curve)  and
    C02 (lower curve).  The  gas  sampling probe was adjacent  to
    position 16.
                               45

-------
   20
  z
  Ul
  o
  ac.
  LU
  Q.
  UJ
  o
  o

  E
  O

  o:
  UJ

            CO,
                 100
                             200          300

                                TIME IN HOURS
                             400
500
Figure 33.     pile  9B'
     C02  (lower curve).
     position 7.
concentrations  of (^ (upper curve)  and

 The gas sampling probe was adjacent to
                                46

-------
  20
 §16
 or
 iu
 Q.


 -12
 2
 O
                100
200          300
  TIME IN HOURS
                                                    400
                                                                 500
Fisure 34     Pile  9B,  concentrations of 02  (upper  curve)  and
    CO  (lower curve).   The gas sampling probe was  adjacent to
    position 14.
                               47

-------
         80
                        100
200           300
   TIME IN HOURS
                                                               400
                                                                            50O
Figure 55.    Piles  7  and  8, moisture content.  Samples taken from central  interior
    locations.  The  data refers  to  "whole samples" (includes woodchips).

-------
     20
100
                                 200           300
                                    TIME IN HOURS
                                                           400
                                                                        500
Figure 36.    Piles 9A and 9B,  moisture content.  Samples  taken  from central interior
    locations.   The data refers to "whole samples"  (includes  woodchips).

-------
 ed  rapidly.   The  Beltsville  pile  (9B)  dried  only  slightly,  with  the
 terminal  value  being  61$.
     The pH  data  are  given  in  Figures  37  and  38.   The  pH  increase  of  pile
 9A was  less  than  that of  piles  7  and  8 and  comparable  to  previously re-
 ported  piles (2) .  Pile 9B  was atypical in  that the pH decreased.

 L_e_aehajte

     Liquid  was last  seen issuing from pile 9A  on  day  3,  and  from  pile
 9B on day 17.


 REPRESENTATIVENESS  OF PILE  9B

     Piles 7, 8 and 9A represented  the Rutgers  strategy,  whereas only
 pile 9B was  constructed and managed according to the Beltsville pre-
 scription. The representativeness  of pile 9B was assessed by  comparing
 its behavior with that  of Beltsville-type piles reported  by others  (9,
 38-39).  The analysis  (see Appendix A-4) demonstrates  i.  a consistency
 of behavior  among piles managed according to the Beltsville prescription,
 ii. a consistency of  behavior among piles managed  according to the Rutgers
 strategy, iii. dissimilar behavior  between the  two  groups of  piles.  It
 is concluded that pile  9B adequately represented the Beltsville Process.

 GENERALIZED COMPARISON OF THE STRATEGIES

     The objective of  the Rutgers process control  strategy is to maxi-
 mize the rate of  microbial decomposition.   Operationally, this translates
 into the optimization  of  temperature via controlled ventilation.  For a
 brief period at the outset of processing (ea. ,  i day) the purpose of
 ventilation is to promote a rapid temperature ascent.  Thus,  the need to
 provide QZ for heat <~  .eration, through aerobic respiration,  temporarily
 conflicts with the :_eed to minimize heat removal.  The compromise is to
 actuate the blower, by timer, on  s, schedule to provide an adequate 0?
 level (ca. .  54) .

     When the temperature reaches a preset level (TggfJ  assigned to a
 controller and sensed via a thermistor in the pile, the purpose of venti-
 lation changes to that of matching heat removal to heat output, such that
pile temperature  is poised at a biologically favorable level  (<60°C) .
This is  achieved  automatically through establishment of an interaction
between pile and  blower, via the feedback o£ a temperature signal to the
controller.   With the onset of temperature feedback control the conflict
described above  disappears,  as the needs of 02 supply and heat removal
now coincide.  The period of feedback  control  ends at some indeterminate
time (typically  0-15 days),  when the loss of heat independent of venti-
lation by "demand" starts to exceed heat generation.

     The required ventilation capacity (e.g.,  blower horsepower)  is de-
fined by the peak demand for heat removal, which is primarily a function
of the abundance of readily metabolizable substrate in the waste.   The
forced-pressure  mode of ventilation is preferred for its greater effieien-


                                 50

-------
                      100
                                   200           300
                                      TIME  IN HOURS
                                                             400
Figure 37.    Piles  7  and  8,  pH.   Samples taken from central  interior locations

-------
                       PILE 9A 	
                       PILE 98 	
                                                                    —O
       4 (—
                                                                           ..JSl
                     100
                                  200
                                                300
                                                              400
                                     TIME
                                            HOURS
                                                                            500
Figure 38.    Piles  9A  and 9B,  pH.  Samples taken from central interior  locations

-------
 cyin removing heat and water vapor (7).

      In  the  Beltsville Process  the stated  purpose  of  ventilation  is  to
 maintain (^  at  a  level of  from  5%  to  15-t  (9) •   This is  accomplished  (8)
 through  use  of  a  timer to  schedule blower  actuation and deactuation.
 Usually  the  blower is  operated  201 of the  time,  but this  percentage  may
 be varied manually based on  02  level.  Pile  temperature typically peaks
 at 80°C  within  a  few days  and changes  little  for the  remainder  of this
 operational  stage,  which is  fixed  at  21 days.   A subsequent period of in-
 formal curing typically lasts several  months.   Blower size is standard-
 ized  at  1/3  hp    per 45 wet  tonnes of sludge  cakes, regardless  of the
 type  of  sludge  (e.g. raw,  waste  activated, anaerobically  digested).

 UNIFIED  INTERPRETATION OF  EVENTS INDUCED BY  THE  TWO STRATEGIES

      This interpretation is  based  on  the plug  flow of air through a  com-
 posting  matrix, and the interaction between  heat generation and tempera-
 ture.  V/e start by considering  two roles of  ventilation:  supplying  $2
 and removing heat.  In turn  this leads to  a  consideration of water re-
 moval.   Removal of C02 can be neglected for  the  present purposes.

      The  02  (% vol/vol) at any  given  point along the  airflow pathway re-
 presents  the balance between the rate  of microbial 02 uptake and  the rate
 of resupply  through ventilation.   Although the  two control strategies re-
 sult  in  similar 02  contents, this  reflects dissimilar rates of  uptake and
 resupply.

      In  the Rutgers strategy the sequence  o£ control-related events  is:
 02 is consumed through microbial activity, which generates heat;  the
 temperature increases,  resulting in a  signal actuating  the blower; venti-
 lation removes heat and resupplies  0%;  the temperature decreases,  deac-
 tuating  the blower.  The cycle  is  repeated until the period of  tempera-
 ture-feedback control  terminates,  reflecting a  low rate of heat genera-
 tion.  In a direct  sense the response  of the blower is  time-variable
 according to the  rate  of heat generation, as expressed  through  tempera-
 ture.  Indirectly,  however, the response is time-variable according to
 the rate of 02 upta .,  in  that  03  uptake and heat  generation both result
 from organic matter decomposition.   The rates of 02 uptake and  heat gen-
 eration are therefore  directly proportional.  Since the response  thres-
 hold  (Tset;)  is selected to maximize heat generation and 0? uptake, in-
 trinsic to the Rutgers  strategy is  a high rate of  ^2 uptake matched by
 a high rate of $2  resupply.  The result is a well-oxygenated pile.

     In the Beltsville  process the sequence is:  slight ventilation re-
 moves heat and resupplies  02 in slight amounts;  the temperature ascends
 to a level that suppress 02 uptake and other manifestations of raicrobial
 activity.  The rate of  resupply of 0->, though low,  suffices to maintain
 an oxygenated condition because of tne slight metabolic activity.   Thus,
 the stated objective of maintaining 92 at a level of 51 to 15%  (9) is
met by suppressing microbial metabolism (waste decomposition)  through
 operation at biologically  harsh temperatures.

     Ventilation should now be considered from the  vantage point of heat
 and water removal.  Removal through radiation and conduction is insigni-
 ficant or minor (4), and these mechanisms may be neglected for the pre-
 sent purposes.  Heat is transferred from the solid  phase  (composting

                                 S3

-------
 matrix)  to the gaseous phase (flowing airstream)  through the mecha-
 nisms of convection and vaporization of v/ater.   Vaporization is the
 dominant mechanism, removing approximately nine-fold more heat than
 convection (2).   For the transfer to occur an enthalpy (heat content)
 differential  must exist between the phases.   Because the matrix is the
 site of  heat  generation, its enthalpy is elevated relative to the air.
 As  the air flows through the matrix it progressively accumulates heat,
 decreasing the differential  and thus the potential for transfer.  This
 set of circumstances induces the progressive  storage of heat in the
 matrix material  along the axis  of airflow, resulting in the establish-
 ment 'of  a positive temperature  gradient in the  direction of airflow.
 The temperature  gradient, in turn,  maintains  an enthalpy differential
 between  the phases.

      Based on the relationship  between the enthalpy differential and
 the temperature  gradient, drying in the vertical  dimension is expected
 to  be relatively uniform. This is  because the  flowing air continu-
 ously approaches a saturated condition yet, provided that the adjacent
 matrix remains a site of heat generation,  does  not reach saturation.
 Consistent with  this expectation, repeated informal observation indi-
 cated uniform drying.

      In  contrast, ventilation of a  metabolically  inert mass (no heat
 generation) causes vaporization only insofar  as the inlet lir is
 unsaturated.   This is  the subject of Section  8.

      Meanwhile,  it  is  sufficient to  note that in the metabolically inert
 system a  discrete cooling-drying front  (evaporative  cooling)  is  expected
 to  migrate gradually through the pile,  and that this was  observed  in
 a ventilated  pile of well-curved, essentially inert, compost.   Relat-
 ing the  data  from the  pile of compost  to those  from the  Rutgers
 composting piles,  an estimated  95.5%  of the coraposting-associated  dry-
 ing is attributable  to  heat  generation  and the  -remainder  "(4.5$)  to
 unsaturation  of  the  inlet air.   Similarly, a theoretical  calculation
 attributes 95% to 98.2%  of the  composting  associated drying to  meta-
 bolic  heat generation  (Appendix A-S).

      Thus, drying during  composting  is  linked to organic  matter  decom-
 position,  in  that  the  vaporization  is driven almost  exclusively  by
 heat  generation.   Since  decomposition of putrescible material is the
 primary goal  of  waste  treatment, we  earlier suggested  (2-4),  that the
 course of  drying can  serve as a specific,  objective, sensitive,  and
 convenient indicator of process performance.  In this  application the
 moisture content  test  is  more sensitive than the volatile  solids test
 (Appendix A-6).   Figures  35  and 36 thus serve to represent  the compara-
 tive performance  of  the  two  static-pile composting processes.


 DISCUSSION

     These observations provide a coherent framework for  composting
process control  based on  interactions among microbial heat  generation,
 temperature, vaporization, and ventilation.  This is given  below.

     i.  The composting microbial ecosystem tends strongly  to self-limit
via excessive  accumulation of mcrtabolically generated heat, leading to
                                 54

-------
 inhibitively high tempera tut'o.   The threshold to significant inhibition
 is approximately S5°-60°C,  and  its severity increases sh'arply at higher
 temperatures.   Unless controlled through deliberate heat removal,  com-
 posting masses typically peak at GQ°C,  at which point the rate of de-
 composition is extremely low.

      ii.  This  self-limiting tendency must be countered if decomposition
 is to be  fostered.  Consequently,  the central problem in the design and
 control of composting facilities is heat removal in reference to a 60°C
 operational ceiling.

     iii.  A practical  means  of removing  heat from the composting mass  is
 through ventilation.   The main  ventilation-associated mechanism of heat
 removal (ca.  90$)  is  the vaporization of water.   Ventilation also  sup-
 plies 02  for  aerobic  decomposition (:,iain source of  heat).

      iv.  The  forced pressure-mode  of ventilation removes  heat more effi-
 cently than the  vacuum-induced  mode.

      v.   During  composting  the  rate of  heat generation is  time-variable.
 Hence,  to maintain a  given  temperature  this must be  matched  by corres-
 pondingly time-variable  heat  removal.   Implementation is  through tempera-
 ture  feedback  actuation  of  a  blower,  using  standard  control  equipment.
 Composting mass  and blower  thereby interact,  to  seek an  assigned set-
 point temperature.

      vi.  To achieve the  desired  operational ceiling  of 60°C,  it might  be
 necessary to assign a  lower set-point (e.g.  4S°C).

    vii.  The blower capacity  (head  and  volume) must  suffice  to meet peak
 demand  for  ventilation.   A  strong  waste  (e.g.  raw sewage sludge) demands
 more  ventilation than  a  weak  one (e.g.  digested  sludge).

   viii.   A  temperature gradient  is  established along  the axis  of air-
 flow.   This imposes a  height  limitation,  above which  a high  rate of de-
 compostion  is  not  obtainable.  With  the  sludge tested  herein,  the  limita-
 tion  was  approximately 2  meters.   Drying  is relatively uniform along this
 axis  of airflow.

      ix. Managed thusly,  decompostion and  drying  are related  in that the
 following chain of causation  is established: decomposition generates
 heat; the heat vaporizes water; the vaporization causes drying.  Pro-
 duction of metabolic water replenishes only ca.  10% of the water re-
 moved.

     x.  A consequence of this control strategy  is that the composting
mass is well-oxygenated  (typical 0? level,  171 v/v).  This is because
 approximately  9x more air is needed to remove heat than to supply 0? for
 aerobic respiration.

     This framework results  from a comparative analysis of two funda-
mentally different approaches to the management of the composting-micro-
bial ecosystem.  These are represented by the Rutgers strategy and the
strategy embodied in the Beltsville Process.  It may be doubted whether
a different analytical point  of departure could lead to a coherent frame-


                                 SS

-------
work.

     The study material was sewage sludge, and the configuration that of
unenclosed static pile.  The derived principles are nonetheless relevant
to other materials and configurations.  The static pile configuration,
however, is structurally and operationally simple, and is therefore pre-
ferred.
                                56

-------
                            SECTION  4

          SEQUENCE OF LIMITATIONS  INDUCED  BY CONTROL STRATEGY


INTRODUCTION

      The Rutgers strategy is to maximize  the rate of decomposition
via heat removal in reference to temperature.  Since the main heat
removal mechanism is the vaporization of water,  a strong drying
tendency is induced.  Consequently,  the prevention of inhibitively
high temperature promotes dryness, possibly to an inhibitive extent.

      Conversely, the Beltsville process focuses on the maintenance
of a minimal level of interstitial 02.  This leads, by default, to
inhibitively high temperatures and tne retention of water.

      The present Section concerns dryness as a  limiting factor.  The
experimental observations support  a  general discussion of the sequence
of limitations, and its practical  implications in sludge management.

MATERIALS AND METHODS SPECIFIC TO  SECTION  4

      These observations concern one pile  of the sludge-woodchip
mixture (Figure ?9), managed according to  the Rutgers strategy.  Depar-
tures from"the Materials and Methods described in Section 3 are that
the sludge-woodchip ratio was' 1 tonne sludge per l.Sm^ woodchips (1
ton per 1.8 yd^);  the infrared analyzer for C0» determination was a
Beckrnan Model 315; the material tested for moisture content had been
passed through a 0.64m (J in.)-sieve (ASTM E-ll specification).

      The trial period was 10 May  to 25 June, 1979.  Ambient air
temperatures during this period were (°C): mean of the daily highs,
24.6; mean of the daily lows, 14,5; mean of the daily means, 19.6;
range of the daily highs and lows  9 to 31.  Rainfall amounted to 26 cm
in 20 occurrences.

      The pile is  represented by Figure 39.  On four occasions (see
arrows,  Figure 40) water was added to the central portion of the pile,
through various procedures.  Occasion 1:  approximately 44SL (120 gal-
lons) of water was pumped through a gas sampling probe positioned
variously in the central portion of the pile.   Occasion 2:  four 176L
(20 gallon)  containers filled with water were carried to :he top of
the pile and emptied.   Occasion 3:  water was applied to the pile sur-
^ce  with a ^'re hose, at a low rate, for approximately 90 minutes
Occasion 4   watc  was applied to Chi pile surface with a fire hose,
 •  a  irod^-i.ie r..te, for approxima  \y 10 minutes.
                                57

-------
Figure 39.   Pile 4A, cross-sectional representation:  textured area, woodchip cover
    and base; clear area, sludge-woodchip mixture; circle, perforated  flexhose.  The
    numbers indicate monitoring and control positions: thermocouples,  positions  1
    through 14; control thermistor, position 1.  The blower was operated  in  the
    forced-pressure mode.   (This pile was designated 4A  in reference 4, and  A in
    reference 2.)

-------
       100
               100   200    300    400   500    600   700    800   9OO    1000   1100
                                        TIME IN HOURS
Figure -10.   Pile 4A, blower operation,,  and  temperature at the control thermistor.  The
     baseline represents operation  as  scheduled  by timer,  and the area above the base-
     line represents operation through the temperature-feedback control system.   The
     time of water addition is indicated by  arrows.

-------
RESULTS

Hour Zero to Hr 490

      Blower operation  first exceeded  that  schedule  by  timer  at
hr 12, indicating  that  the controller  had responded  to  a  thermistor
temperature of >4S°C  (Figure 40).  This marks  the  start of  the period
of temperature-feedback control.  Blower operation v/as  nearly contin-
uous from hr t'O to 150.   Feedback control terminated at hr  352,  at
which time blov/er  operation reverted to that  scheduled  by timer  (7%
of the time)„

      Based on 1020 observations during the period of feedback control,
the median pile temperature was 48°C,  and the  range  was 25°C  to  63°C,
Temperatures in excess  of 60°C accounted for  1.81  of these  data  points.
The temperature at the  thermistor control position is shovm in
Figure 40.

      The level of C0?  peaked at 144 just prior  to the  start  of  the
period of feedback control.  Thereafters CO,,  generally  varied between
2% and 4*.                                  L

      The starting moisture content of 761  decreased to ?.?.$ in IS
days (Figure 41),,  This refers to sieved material  (woodchips  removed).

Hour 490 to_Hr^lI4j)

      Immediately prior to the first water  addition  the temperature  at
the thermistor position had declined to well  below the  set  point value
(24 C observed vs. 45°C set point), and blower operation  was  solely
as scheduled by timer.  At this time the moisture  content was 22%.

      Water was added,  as described above,  to  the part  of the pile
containing the control  and monitoring  devices  (see Figure 39).   The
material absorbed water slowly and non-uniformly,  therefore repre-
sentative samples for the determination of moisture  content were not
obtainable in this part of the study.

      Approximately ten hours after the first water  addition  the
temperature re-ascended,  initiating a  second period  of  feedback  control.
Before the termination  of this period  the second \iater  addition was
made.   Shorrly thereafter feedback control  ceased, and  blower opera-
tion by timer-schedule  resumed.,   The third and fourth water additions
initiated -  lependent.,  successively weaker, periods  of  feedback  control.

      The comparative intensity of the successive periods of  feedback
control may be judged on  the basis of  the amount of  blower  operation
time.   Baseline blower  operation, as scheduled by timer,  is included
in this comparison.  Setting blower operation during  the  original
period (hr 10 to hr 348)  equal to 100, the other values are:  hr  519
to hr  69,  20.2;  hr 872 ro hr 946, 5.9; hr 1044 to hr 1070, 1.8.

      During the part of  the trial from hr 490 to hr  1140,  rainfall
amounted to  10.8 cis in  8  occurrences.  The heaviest  rainstorm,
                                 60

-------
                       100
200           300
   TIME IN HOURS
                                                             400
                                                                           500
Figure 41.  Pile 4A, moisture content.   Samples taken from central interior  locations.

-------
depositing 6.2 cm, started on hr 735 and ended on hr 755.  This
storm occurred between periods of temperature-feedback control  (see
Figure 40).  The rainfall wetted the pile only superficially, and
did not revive blower demand.

DISCUSSION

      Four factors can come to limit microbial activity in a compost-
ing mass.  These are:  0, depletion; development of excessively high
temperature; water depletion; substrate depletion.  Except for sub-
strate, whether a particular limitation comes into play or is bypassed
depends on the means of ventilation employed and whether water is
added (.Figure 42) .

      The sequence is discussed in reference to specific examples,
Oxygen depletion is exemplified in the composting of leaves in static
piles without mechanical ventilation, as reported elsewhere (40).
A summary of the pertinent observations follows.

      Soon after assembling leaves into a pile of substantial size,
e.g., L x W x H ° 7.6x3.1x1.8m (25x10x6 ft), the interior portion of
the pile became 0.,-deficient (defined as 0-, not detectable or barely
detectable).  The 02-deficient portion comprised roughly half the
total pile volume.  The outer portion (roughly half the volume)  re-
mained well-oxygenated.  The maximum temperature throughout the pile
rarely exceeded 60°C, presumably reflecting the slight generation of
heat fermentatively compared to O^-based metabolism.  Over the ten
month  observation period the pile shrunk to approximately half its
original volume.  The relative proportions of 0-,-deficient and oxygen-
ated volumes remained approximately equal.

      At termination the pile was disassembled for examination.   The
material in the oxygenated zone was damp and appeared to be humified,
whereas in the central core the material was wet, gave off an odor of
putrefaction, and retained evidence of the original leaf structure.
Thus, overall the leaves had not yet received adequate treatment.

      Nonetheless, it was concluded that satisfactory composting of
leaves is possible without mechanical ventilation (or agitation),
For this type of waste the need is to insure that the material in the
innermost core becomes oxygenated within the processing time available,
in that this signifies that all of the material has undergone thorough
decomposition.   Provided that the leaves are moist at the outset,
adequate decomposition throughout occurs over winter.

      An inadequate ventilation system,  in terms of blower capacity
and/or control system, leads to inhibitively high temperature, whereas
an adequate system leads to dryness and/or substrate depletion.   These
circumstances are represented by the Beltsville Process and the  Rutgers
control strategy, respectively, as was developed earlier.

      The practical implications  of this sequence are illustrated
through three hypothetical cases.   The first concerns a "dirty sludge,"
contaminated with heavy metals  and/or non-biodegradable industrial
                                62

-------
                                     MATERIAL ASSEMBLED
                                     FOR  COMPOSTING
NO FORMAL VENTILATION
      SYSTEM
INAPEQUATE VENTILATION
      SYSTEM
 ADEQUATE VENTILATION
        SYSTEM
OXYGEN  BECOMES
THE LIMITING FACTOR}
SYSTEM DEOXYGENATED
                              JL
 TEMPERATURE BECOMES
 THE LIMITING FACTOR;
 SYSTEM OXYGENATED
TEMPERATURE DOES  NOT
BECOME  THE LIMITING
FACTOR; SYSTEM OXYGENATED
                        WEAK TENDENCY
                           TO  DRY
                                STRONG TENDEP4CY
                                    TO DRY
   STABILIZATION
 PROGRESSES SLOWLY AS
 OXYGEN PENETRATES
     THE  MASS
      POORLY
    STABILIZED
      ORGANIC
      RESIDUE
                                            WATER NOT ADDED
                                            ACTIVITY DIMINISHES
                                            AS WATER BECOMES
                                            THE LIMITING FACTOR
                                           WATER ADDED
                                         ACTIVITY DIMINISHES
                                         AS  SUBSTRATE
                                         BECOMES DEPLETED
Figure  42.   Reprinted  bv permission from BIO/TECHNOLOGY,  Vol.  1,
    No.  4,  pp.  347-353.   Copyright ©  1983.   Nature Publishing  Co.
    Limitations  to biological  activity  induced  by different manage-
    ment strategies.
                                     63

-------
chemicals.  Such a sludge should not be considered as a feedstock
for compost production, yet composting might be the treatment process
of choice.  The process residue might be suitable as a landfill cover
material, or as a low grade solid fuel.  This residue is  lesireable
as a fuel, comps.red to the uncomposted sludge cake (Appendix B) .

      The second and third hypothetical cases involve a "clean  sludge,"
which affords a wider range of opportunities for ultimate disposal
via resource recovery.  Consider the production of a "rough" compost
for restricted bulk application to soil.  This product must be  aes-
thetically acceptable and remain so after application tr soil,-hence
it should be moderately well stabilized.  More extensive stabilization
than required to meet this need is undesireable as it adds to process-
ing costs and results in a loss of agronomically valuable nitrogen and
organic matter.  Production of a "rough" compost calls for a relatively
high initial moisture content  (consistent with reliable process "start-
up"), to prolong the period of microbial action prior to dryness.

      Finally, production of a highly stabilized compost for un-
restricted distribution calls for more extensive biological decomposi-
tion.  This might be accomplished in-place, as an extension of  a high
rate stage, by adding water to prevent premature dryness.   In this
manner the high rate stage v/ould gradually pass to an in-place curing
stage.   Alternatively, the material might be pennitted to become dry
in the high rate stage and then moved, and remoistened, for curing.
Other unit process flow schemes can be envisioned.
                                64

-------
                          SECTION  5


         RUTGERS  STRATEGY:  REPLACEMENT OF WOODCHI PS WITH
             RECYCLED  COMPOST AS THE  BULKING AGENT


INTRODUCTION

      Composting  requires gas exchange to remove  heat  and water  vapor
and to supply 02>  Mechanical agitation can provide the needed ex-
change only intermittently, and at high energy cost.   Thus,  agitation
does not lend itself to  rate maximization through controlled heat
removal, and its  main  role in composting is to mix and abrade the
material.  Ventilation by blower affords the only practical  approach
to rate maximization,  but this imposes a requirement for porosity to
permit the passage of  air.  Since  sludge cake by  itself lacks porosity,
it is commonly mixed with a "bulking  agent" having this property.

      The usual bulking  agent, woodchips, has serious  drawbacks.  In
routine Beltsville-type  operations, the purchase  of woodchips and
associated operations  (storage, translocation, mixing, screening)
represent perhaps one-third of the overall costs  (41).  Furthermore,
woodchip stockpiles are  colonized  by  Ajjjergillus  fumigjrtu_s_,  a fungus
which can infect  the human lung.   It  wouTdFe~3e'sTre¥ble, instead,  to
use internally generated recycled  compost as the  bulking agent,  while
retaining the structural and operational simplicity of the static pile
configuration.

      To dp so, the recycle must consist of i) stable  aggregates in
the physical sense, to impart porosity, ii) highly stabilized material
in the sense of supporting only slight metabolic  heat  generation, to
reserve most of the ventilation system's heat removal  capacity for  the
fresh sludge, iii) dry material, to absorb water  from  the sludge to
improve porosity.  Furthermore,  the  composting process itself should
promote drying so that once composting is initiated, porosity pro-
gressively improves.   These are precisely the tendencies intrinsic  to the
Rutgers strategy, hence this trial.

MATEuIALS AND METHODS SPECIFIC TO SECTION 5

      Screened material  (woodchips removed) from  piles 8 and 9A  \vas
used as the bulking agent.   Screening was by means of  a Royer Model 355
shredder-mixer .coupled with a Mogensen sizer (Royer Foundry and  Machine
Co.,  Kingston, PA).   After the screening the material  was stockpiled
in the open for approximately 4j and  3 months,  respectively, without
deliberate remoistening.   At the time of use the pile  was cool,  the
bulk of the material  had a moisture content of 391,  and the material
had a slight earthy aroma.

      Fresh sludge ct'ke and compost were fed by separate conveyer belts
into  the pug mill for mixing.   The feed rates were adjusted by eye to
                                65

-------
yield 6ne  batch  with  a  low  recycle  ratio  (sludge-rich),  one  batch
with an  intermediate  ratio  (roughly equal  proportions),  and  one
batch with a  high  ratio (compost-rich).  The  ratios  actually
obtained were  determined  by calculation bascid on  the moisture  content
of the sludge  (74%),  the  compost  (391), and  the mixture  (TABLE 4.).
Dry weight
Approx pile recycle ratio
wet weight (compost/compost
11A 5.4 0.3
11B 4.5 0.6
11C 3.6 0.8
Initial
+ Initial mois- approx. ,
69.5 5.3
61.5 5.3
52.0 5.3
     Each compost-sludge mixture  was  formed  into  s.  pile  over  parallel
segments of perforated  flexhose within  a woodchip bed  (Figure 43),
fitted with control and monitoring devices  (Figures  7  and  44),  and
covered with woodchips  (Figure 45).   (It was  thought advisable  to
insulate these small, free-standing,  piles with a woodchip  cover.)
Each pile was ventilated with one 1/3 hp blower operated in the forced
pressure mode.  Process control was based on  temperature feedback,
with the thermistor located  at position 1 (Figure 43), and  the  tem-
perature controller was set  to 45°C.  Timer-scheduled  operation was
0.75 min (uninterrupted) per IS min.

     The trial was started on 24  Oct  1980 and  terminated on 7 Nov,
at which time the woodchip cover  was  removed  (Figure 46).   Ambient
air temperatures during this period were (°C): mean  of the  daily highs,
14 ; mean of the daily  lows, 4°;  mean of the  daily means,  9°; range
of the daily highs and  lows, -2   to 21°.  Rainfall  amounted to  8.0 cm
in 3 occurrences.

RESULTS

Pj. c t o ri a 1_Reprg_servtat_io n

     The periods of temperature feedback control  ended on hrs 228,
186, and 102, respectively (see Blower  Operation, below),   On hr 330
the woodchip cover was  removed from all of the piles.  A pictorial
overview is provided by Figures 44 and  46,   Other pairs of  photos give
"before and after" closeups of each pile (pile 11A,  Figures 47  and 48;
pile 11B, Figures 49 and 50; pile 11C,  Figures 51 and  52).  Shrinkage
in volume is evident, being greatest  in pile  11A, intermediate  in
pile 11B, and least in pile 11C.  This  is also the order of sludge (and
water)  abundance in the mixtures  at time-zero.
                                 66

-------
Figure 45.    Piles  11A,  11B,  and  11C,  cross-sectional  representation:   textured  area,
    woodchip cover  and base;  clear  area,  sludge-compost mixture;  circles,  perforated
    flexhose.   The  numbers  indicate monitoring  and control  positions:   thermocouples,
    positions 1 through 12;  control thermistor,  position 1.   The  gas sampling probes
    were  positioned as follows: pile 11A,  adjacent to  positions  2 and  7;  pile 11B,
    adjacent to position 2;  pile  11C,  adjacent  to position  7.  The blower was
    operated in the forced-pressure mode.

-------
Figure 44.   Piles  11A,  11B, and  lie, before composting.  The
   dowels extend from the pile,  and the thermocouple leads are
   organized at subjunction boxes.  The thermistor was at the
   end of the  lowest dowel.  Modified oil drums served to cover
   the blowers at  the rear of the piles.  Photo by F.C. Miller.
                             68

-------
Figure 45.   Same as Fig. 44
   Photo by F.C. Mills?.   '
      covering with woodchips,
69

-------
70

-------
  1gy"^ -~^~^-*~~
  IH   -ill
'IM'
                       if.
  1   ~ t-                  i

  R./F'^.J'  '-1
Pila 1U,  before costing.   Photo by p.c. Mi
                71

-------
— _ _ _




If

'r
•*> ?
^
to-** I.-,'9 '- *%--''»
X. ^ J"1 > 1
^

^
f
j
t- 4
n
*"V f
r1

f ''
'
i-- • x , «
A ' K:> /'
' {C ~-^ w /
1 "
V— 1^'
*JS sv>'--!^ r^ J> '

-------
                                        $ J,'-'1 „' -
                                                       x^, -
49.    Pile 11B,  before composting.  Photo by  F.C.  Miller,
                      73

-------
   n
   1(0
           ii"
-T;
^f
. ^i
 p.
^i
i ^   /•"?  ••
  ii    -»  *
 , I \*   X *
 • f *-» M-'^i
  ^   f ^.. . l^
   o
   »*
   t:
   I,
       1C-'"
*" -*s

 K
- ^
- ^-^—

-V
          ^ d
 .-  *,  'j^r!   f'ii
 _*r~\ ^f>'^ 5"''

 ';   !' <":H-3 ? J>'j f 1
^'%U:/'"^"^,
 -^.,.^-?^^?^rv
     - '^i '""'  ' -''*
        \ i
   ^-*«iv"  -:^"5J:TJ
   |cv^^ii t^ii>«
   P     "    ' T
                                                    "J*^
       fr
                   ,«» tv
                 .  1S?f\'
                '   '"< f
PJgure  50.    Pile  11B, after composting,
    been removed.  Photo by F.C. Miller,
                            The woodchip  cover has
                             74

-------
    ni
     U,
•

                           X LF-
                           ll!
                                         xr
                                                    .,  N
                                             f- I   ]H
   JL
Figure 53.    Pile  11C,  before composting..  Photo by F.C. Miller.
                               75

-------
II
E|J,-_

'
                                             nt,
                   -.
                 -
                     k if    •
                                                   -     ,  :   -  ,--,-
Figure 52.
                                                                   cover
                                    76

-------
     The duration of  temperature  feedback  control  (Figures  5?,  54  and
 S3) was  longest  in pile  HA  (hr 22  to  22B~) ,  intermediate  in pzle  11B
 (hr 16 to  186) and briefest  in pile  11C  (hr  14  to  102).   In piles  HA
 and 11B  the peak demand  for  ventilation  utilized approximately  551  of
 the blower capacity.  The comparable value  for  pile  11C was 25%.   The
 comparative overall demand for ventilation was  assessed in  terras  of
 the total  blower operation time during the period  of feedback control
 (baseline  included).  Setting the greatest demand  equal to  100, the
 values were:  pile HA,  100; pile 11B, 92; pile 11C,  23.

 Pile Temperature

     The temperature  plots for the  thermistor and  overlying positions
 are given  in Figures  56, S7, and  S8,   At the thermistor position  the
 set point  temperature (45°C) was precisely maintained  during the period
 of feedback control.  With some exceptions,  notably  at probe 11B6,  the
 higher positions experienced higher  temperatures in  the usual pattern
 of a systematic  gradient.  The gradient  was  most clearly  established
 in pile  11C (compost-rich).  In other  parts  of  the pile the tempera-
 ture control was less precise (see Appendix  C).  The  temperature
 observations at  all of the positions during  the feedback  period,  and
 other data, are  summarized in TABLE  5.

 Pile Atmosphere

     The lowest  02 levels recorded during the period  of feedback
 control were as  follows:  HA (high probe position) r  14.51;  HB (low
 probe position), 16.8%;   11C  (high probe  position), 13.3%.   The cor-
 responding peak  CO., levels were 2.8%,  1,84 and  6.8%  (Figures 59, 60,
 61 and 62).

 Moisture Cont_erit

     The initial moisture contents  (HA, 70%; lib, 61%; 11C, 52%) re-
 flected the differences  in the sludge-compost mix ratios  (Figure 63),
 Water losses were positively correlated with the amount of  sludge in
 the mix.   The minimum moisture contents, observed prior to  the ter-
 minal observation, were: pile 11A, 29%; pile 11B, 21%; pile  11C, 23%,
 The terminal values were slightly higher.

Visual, Tactile, and Aesthetic J^ujy.ijj.ej;^

     At termination, the piles were bisected and examined.   The material
comprising the outer rim of the sludge-rich pile HA  ('-he "toes")  was
wet and pasty,  with an unpleasant odor.  This material had  not com-
posted appreciably.   The bulk of the pile however,  had composted ex-
tensively',  was dry,  and had a greyish  cast seemingly imparted by my-
celial growths,   This part the pile consisted of chunks of  material,
ranging in size from that of golfballs to boulders.  The  individual
units  v/ere resistant to  breakage by hand.  Unlike the outer  rim, the
bulk of tha material had no conspicuous odor.
                                  77

-------
                                        In..,,
                                     200           300
                                       TIME IN HOURS
400
                                                                            soo
Figure S "S     Pile  11A,  blower  operation.  The baseline represents  operation  as
    scheduled by  timer,  and  the area above the baseline represents  operation  through
    the temperature-feedback control system. The blower was operated  in  the forced-

    oressure mode.

-------
        100-
                                    200          30O
                                      TIME IN HOURS
                                                             400
Figure 54.    Pile 11B, blower operatior .  The baseline  represents operation as
    scheduled by timer, and the area above the baseline  represents operation through
    the temperature feedback control system.  The  blower was  operated in the forced-
    pressure mode.

-------
CO
O
                   icoi
                  Z
                  O
                  O
                  t-
                  3
                  O
ii   r  N
±<  J  ' s
>J  r   ,--'1

                                  100
                          200           30O
                            TIfc3E IN HOURS
                                                                                       500
           Figure  55.    Pile lie, blower operation.   The baseline represents  operation as
              scheduled by timer, and the  area  above the baseline represents  operation through
              the  temperature feedback control  system.   The blower was operated in the forced-
              pressure mode.

-------
                              PROBE IIAIO	
                              PROBE HAS  	
                              PROBE IIA I   —
                                   •N-x
j"
0
L 1 l 1 1 S i
SOO 200 300
TIME IN HOURS
1
400
I
500
Figure S6.
Pile 11A, temperature at the innermost  vertical series of  thermocouples.

-------
                             PROBE IIBIO
                             PROBE IIBS
                             PROBE IIS I

                      100
                    200
                                                300
                                      TIME IN HOURS
                                                400
• j cure
Pile 11B,  temperature at the  innermost vertical  series of thermocouples,

-------
      80 5-
              PROBE IICIO
              PROBE I!C 6
              PROBE !IC I
      20;
                                                *  '' »»
                                                \—*  %
R_ X- 	 ^. 	 S -X
1 ^^/ \^-
\ ^*+J^
r
L -I 	 j 	 i 	 i 	 i S ( S i
3 100 200 300 400
TIME IN HOURS
!•
f
\
f
i
5C





)0

Figure 58.
Pile 11C,  temperature at the  innermost vertical  series of thermocouples,

-------
            TABLE 5.  DATA SUMMARY FOR SECTION s'
Pile

Recycle ratio
Temperature (during period
Period (hr)
Range (°C)
Median (°C)
Observations (no.)
% observations * 60°C
Interstitial gases (time 0
0. Range (1)
Li
02 Median (%)
Observations (no.)
C02 Range (%)
C02 Median ($)
Observations (no,)
Moisture content ($)
Time- zero
150 hr
11A
0.3
of feedback control)
22-228
29-73
53
624
24
to hr 332)
12.8-21
20.8
168
<0.1-8.6
<0.1
168

70
32
11B
0.6

16-186
9-71
46
516
20

16.8-21
20.3
84
<0.1-1.8
0.2
84

61
28
11C
0.3

14-102
11-75
59
264
46

13.3-21
20.8
84
<0. 1-6.1
<0.1
84

52
24
Rutgers process control strategy applied to all of the piles.   Dry
weight recycle ratio s recycle/recycle * fresh sludge.
                               84

-------
   20
  5
  =>


  §'6
  
-------
   20
  UJ
  2
  §16
  z
  UJ
  o
  
-------
 20
UJ
5
D


§16
UJ
o
K.
-12
z
o
t-
<
CE
I—
Z
UJ
o
o
cr
O
         CO,
              100
                          200          300

                            TIME IN HOURS
                                                  400
                                       .00
Figure 61
           Pile 11B,
CCU  (lower  curve)
position  2 .
concentrations  of
The gas  sampling
                                           02  (upper curve)  and
                                          probe  was adjacent to
                              87

-------
   20
   o
   ce
   UJ
   0.
   z
   o
   t-
   
-------
      30
                        PILE IIA
                        PILE IIB
                        PILE 1IC
                      SOO
200           300
   TIME IN HOURS
                                                              400
                                                                            SOO
Figure 63.     Piles 11A, 11B, and  11C,  moisture content.  Samples  taken from central
    interior  locations.

-------
       At termination the material in piles 11B  (intermediate) and
11C  (compost-rich) was uniformly dry, granular,  and brown.  The
material in both of these piles had a slight earthy aroma.

DISCUSSION

       A noticeable difference in behavior attributable to the replace-
ment of woodchips with compost is that, lacking  the rigidity o£ the
woodchip matrix, these piles shrunk considerably in volume.  Other
differences that might also be attributable to the use of compost as
the bulking agent are: a relatively brief period of temperature feed-
back control (faster processing); less precise control of temperature;
faster drying.

       The major significance of this trial is that the possibility
of using recycled compost as the bulking agent in static pile con-
figuration was  demonstrated.  Others have demonstrated use of recycled
compost in this capacity in conjunction with "windrow composting,"
in which the mass is mechanically agitated (42).  Agitation is energy
intensive, however, and affords little control over temperature and
oxygen.

       It was anticipated that pile HA would fail, since it v/as so
rich in sludge  (recycle ratio, 0.3),   Yet, the bulk of the pile com-
posted satisfactorily, judging from all of the parameters.  The outer
material was isolated from the airstream, and did not compost.  For
routine use higher recycle ratios, as represented by piles 11B and 11CS
are indicated.   Nevertheless, the performance of pile HA indicates a
degree of processing resilience in conjunction with the use of re-
cycled compost.

       Owing to compactability leading to greater resistance to air-
flow, sludge-compo.st mixtures may be more subject to a height re-
striction than  sludge-woodchip mixtures.   This also indie?*fts use of
higher recycle  ratios.  Further field experimentation is .. dded to
define the height limitation.
                                  90

-------
                            SECTION  6


         COMPARISON  BETWEEN  THE  RUTGERS  STRATEGY AND  BELTSVILLE
                      ^ROCESS:   MATERIALS  BALANCE


 INTRODUCTION

      The  outcome of  processing in  terms  of materials  balance  impacts
 on  facility design  with  respect to  composting  area,  storage  area,  and
 machinery  needs  for materials handling  and transport.  Moreover  the
 amount  and nature of  the process  residue, relative to  the  sludge,
 strongly influences the  possibilities for ultimate disposal/resource
 recovery.  As  such, materials balance is  a major determinant of
 construction and operating  costs, disposal options,  and  indeed of  the
 utility of composting as a  waste  treatment technology.

      In Section 3, the  Rutgers and BeltsviTle approaches  were compared
 in  terms of blower  operation, temperature, i ,  and C02  levels,  and
 moisture content.   This  comparison  is extended herein, based on  certain
 of  these data  and related analyses.  Also, materials balance is  esti-
 mated for  the  piles described in  Section  5.  These employed  recycled
 compost  as the bulking agent, and were  managed according to  the  Rutgers
 strategy.

 PROCEDURE  FOR  CALCULATING AIRFLOW AND MATERIALS BALANCE

      An airflow delivery of 807 m3/hr  (28,500 ft3/hr) per blower  was
 assumed, based on the manufacturer's specifications  at 5.1 cm  (2 in.)
 of water head.

      The  mass of the mixture after composting was calculated based on
 the equation:
where:

     M- s initial mass

    MC. - initial moisture content

    MC,p s final moisture content
      k = 8.07  (Rutgers strategy), or 8.33  (Beltsville Process),

The constant k  is the ratio mass water vaporized/mass solids decomposed,
as derived in Appendix A-6.  With air exiting from the composting mass
at. 60°C and 100% relative humidity (representative of the Rutgers
strategy), the  constant is 8.07; at 70°C and 1001 RH (representative
of the Beltsville Process), it is 8.33.  Mixture refers to sludge and
woodchips , or sludge and recycled process residue ("compost").
                                 91

-------
       Rearranging:

           M     „  M.[k(l-MC.)  -  MC.]
                                                                   (ii)
 Since  the  values  for  M.  and  MC-  are  known,  the  initial  weights of dry
 matter  and water  can  be  calculated.   Having derived Mr., and knowing
 MC.p, the  final  weights of dry matter and  water  can be  calculated.   At
 this stage the  estimates refer to  the mixture of sludge and woodchips ,
 or  sludge  and recycled compost.

      With respect  to the sludge-woodchip mixture  the  estimate was
 converted  to total  sludge solids decomposition,  and to  sludge  volatile
 solids  decomposition, based  on a sludge-woodchip ratio  o£ 1 tonne
 sludge  cake per 1.7m3 woodchips  (1 ton per  2 yd3) ,  and  the  following
 nominal characteristics:  sludge  cake moisture content,  75.51;  sludge
 cake volatile solids  content,  73.51;  woodchip bulk density, 234 kg/ra3
 (394 ib/yd-5)  (43).  Also, it was assumed  for the purpose  of calcula-
 tion that  the woodchips  did  not  decompose.  To  estimate the volatile
 solids  decomposition  of  the  mixture  of sludge and recycled  compost, the
 compost was assigned  a volatile  solids content  of 54.5% based  on values
 given  in TABLE  6.

 AIRFLOW AND MATERIALS BALANCE  ESTIMATES

      The  airflow and materials balance estimates for  the piles using
 woodchips  as the bulking agent are given  in TABLE 6.   For piles 7,  8,
 and 9A, the mean total air delivery  -during  12.4  days of composting  was
 9,9SOm3/tonne (319,000 £t3/ton). For pile 9B this was  20.8  days and
 2330 m3/tonne (74,500 ft3/ton) .  Thus, mean air  delivery  to the Rutgers
 piles was  4.3x  greater.   The  comparable values  for mean and peak
.delivery were Sx greater  and  15x greater, respectively.   The need  for
 more air reflects the more extensive  waste  decomposition.

      As is characteristic of  feedback control  (Rutgers) , mean and  peak
 usage differed, being 37.0 and 95. Im^/tonne-hr ,  respectively (1,190 and
 3,050 ft3/ton-hr) „  With  timer control (Beltsville)  mean  and peak
 differed slightly,  only  because th-?  blower  schedule  was  adjusted manu-
 ally.   This was as  prescribed  (9) , in response  to high  levels  of 02.

      With application of the  Rutgers strategy,  a mean  of 16.01 of
 the sludge-woodchip mixture was decomposed  in 12.4  days.  The  compar-
 able estimate for the Beltsville Process  is 4.3%,  in 20.8 days.   Assum-
 ing for calculation purposes  that the woodchips  were not  decomposed
microbd.ally (a  reasonable approximation at  these  temperatures  and
 times), total sludge  solids decomposition amounted  to 41,8%  (mean
value)   and 11.3%, respectively.  In  terms of sludge volatile solids
 the estimates are 56.9%  and  15.4%.    Regardless of the fraction (sludge-
woodchips  mixture,  or total sludge solids ,  or sludge volatile  solids),
 the Rutgers strategy  induced  an estimated 307x more decomposition in
 approximately half  the time.    With respect  to v/ater the removals
amounted to 78.21 and 19.4%,   indicating 4x  more  removal in  approxi-
mately half the time.
                                 92

-------
                                                                       *
TABLE 6.   EFFECT OF CONTROL STRATEGY ON AIR USAGE AND MATERIALS BALANCE
_ .._ . . ..


Period (hr)
•?
Air delivery (m /initial wet
Total
Mean (hr"l)
P O Jl V1 r Tl T* "" ^ ^
i JL
Air delivery (ft /initial wet
Total
Mean (hr"1)
Peak (hr"1)


Pile 7
0-344
tonne)
13,100
38.1
81.2
ton x 10
421
1.22
2.60
Material decomposed (i of initial dry
Overall sludge-woodchip
mixture
Sludge total solidst
Sludge volatile solids*
Water removed (4 of initial)

15.9
41.6
56.6
78.9
Process
___Rutj£erj3___
Pile 8
0-170

7,180
42.1
71. 2
3)
230
1.35
2.28
weight)

11.8
30.8
41.8
74.8
control strategy^"

Pile 9A
0-380

9,580
30.9
133

307
0.990
4.28


20.4
53.1
72.3
80.9
Beltsville
Pi.le 9B
0-500

2,330
4.7
5.6

74.5
0.149'
0.192


4.3
11.3
15.4
19.4
  Certain entries  given  in both  metric  and  English  units.

  For  piles  7  and  8  the  calculation  is  based  on  the period  from  time-
  zero to the  cessation  of temperature  feedback  control  ( =  resumption
  of timer-scheduled blower operation).   For  pile 9A it  was  based  on
  the  period  from  time-zero to hr  380  (scheduled operation  not resumed
  because the  timer  was  disconnected during feedback control).   For
  pile 9B it was based on  the period from time-zero to hr 500.   Where
  necessary,  the corresponding value for  MCf  was derived through inter-
  polation.

  The  calculation  was made as if decomposition of woodchips were nil.
                               93

-------
      The materials balance estimates for the piles using recycled
compost bulking agent are given in TABLE 7.  Performance was compar-
able to that obtained with woodchip bulking agent in conjunction with
the Rutgers strategy  (compare to TABLE 7, piles 7, 8, and 9A).
Except that air delivery to pile 11B was somewhat high, the pattern
of results is consistent with the strength of the mixtures.  Thus
pile 11A, which had more fresh sludge than pile 11C, exerted a
greater demand for ventilation and lost more solids and water.  Being
richer in sludge, its potential for heat generation, blov/er demand,
and vaporization was greater.

DISCUSSION

      Composting is a robust process such that it is possible (though
not desireable) to design and operate a facility without benefit of
a coherent process control strategy.  The process1 robustness stems
from two factors.  First, microbial self-heating  (the underlying
phenomenon) commences spontaneously, even if conditions of nutrition,
moisture content, and gas exchange are only marginally adequate.  This
reflects the non-specificity, ubiquity, and rapid growth of the microbes
capable of initiating the process and carrying it forward (18).   Second,
composting is resistant to outright process failure (24),   This stems
from the tendency of the climax population, apparently consisting of
thermophilic members of the genus Bacillus (44-45) to elevate the
temperature to the edge of its tol¥FIHc¥~Timit (^ 7S°C).  In this
state decomposition proceeds, but at a mere fraction of the rate
achievable at 60°C and less.

      This is evidenced herein in that the Rutgers strategy, compared
to the Beltsville strategy, induced 3.7x more decomposition and 4.Ox
more water removal in approximately half the time. 'This outcome reflects
the fundamental difference between these approaches.   The Rutgers
strategy is to maximize the rate of decomposition through ventilative
heat removal in reference to temperature.   The Beltsville process,  by
default,  suppresses decomposition through inhibitively high temperature.

      The Rutgers strategy is expected to result in raor' cost-effective
composting in terms of facility construction and routi^d operation.  The
reasoning is that i.  less facility time/space is required, ii.  less
process residue is produced,  decreasing materials handling operations,
iii.  the residue is easier to handle,  store,  and transport,   Morsover?
this strategy improves composting's utility as a waste treatment tech-
nology, owing to the production of a process  residue that is more
amenable to resource recovery/ultimate disposal.

-------
TABLE 7.  RECYCLED PROCESS RESIDUE AS BULKING AGENT:  AIR USAGE AND

          MATERIALS BALANCE*


Period (hr)
Air de]ivery (m /initial wet
Total
Mean (hr'1)
Peak fhr-1}
Air delivery (ft /initial wet
Total
Mean (hr'1)
Peak fhr-lj

11A
0-228
tonne)
9,770
42.8
86.2
ton x 11
313
1.37
2.76
Material decomposed (1 of initial dry
Overall sludge -recycle
mixture
Volatile solids
Water removed (% of initial)

24.7
36.4
85.4
Pile
11B
0-186

10,100
54.3
119
>3)
323
1.74
3.83
weight)

14.8
23.8
76.1

11C
0-102

3,370
33.1
57.1

108
1.06
1.83


7.8
13.4
58.3
 Rutgers process control strategy applied to all piles.   The calculation
 is based on the period from time-zero to cessation of temperature feed-
 back control (resumption of timer-scheduled blower operation).  Certain
 entries given in both metric and English units. *

-------
                             SECTION  7
                                                             *
        MATHEMATICAL DESCRIPTION OF PROCESS  CONTROL  DYNAMICS


      In this Section  composting process  control  dynamics  are  reduced
to mathematic form.  The mathematical description  focuses on the
generation and removal of heat; an extra-mathematical  constraint  is  the
effect of temperature on heat  generation.  Tiiis  development  leads  to
testable predictions of system behavior.

      Overall heat removal may be expressed  in units of  energy/time:


                       fit 3 ficonv
where:  (J*.    -  total heat removal
              3  convective heat re

              -  conductive heat removal
         Q.     3  convective  heat  removal
          cor,j


         ^r    =•  radiant heat  removal.

 A calculation based on  a ventilated  field-scale  mass  indicates that

 Qconj .'o 0.02 Q  , and that Q  is  small  and  sometimes  exceeded by radiant

 gains  (4).  Thus, all but convective removal  may be  neglected in the
 analysis .

       The relationship  governing convective removal  is:

                         Sconv s 2 £out  - hin)                       Cii)

 where:  m     -  dry air mass  flow  (mass dry  air/time)

         h     K  outlet air enthalpy (energy /mass  dry air)
         ""OU 1C

         _h.    °  inlet  air enthalpy  (energy /mass dry  air).

 The enthalpy of the air is a function of its  temperature  and the amount
 of water vapor it contains.  Note that  to  maintain a  quasi-steady state
 (temperature - constant) , -Qt (and hence QCQnv) must match heat genera-
 tion.   Note also that the goal is to maximize £Lonv  in  a  sustained
 fashion (quasi-steady state maintained), as this is  equivalent to
 maximizing decomposition rate.
      Convective removal can be subdivided into tv/o parts:

                                                                   (iii)
                                   * ^dac
 where:  0^    -  heat removal through vaporization

* We~~tEarnr~D~rT Peter ~Fi
 developed in Sections 7 and 8.

                                  96

-------
           E  removal  through  dry  air  convection.

At  composting  temperatures the dominant  mechanism  of heat  removal  is
vaporizatio.;,  with 0 ^  9 Ud   (2).   Removal  of  heat through vaporiza-
tion  thus  can  be  expressed at?

                     QVH 0.9 m

Note  that  the  goal of maximizing flconv pertains  equally  to


      The overall  rate o£ wp.ter removal is  expressed  as:

                     I  s » ("out • "in3

where:v  B  mass vapor flow  (mass  moisture/time)

       to  -  humidity ratio (mass moisture/mass  dry air).

With  inlet  (ambient) conditions  of 20°C  and  50%  relative humidity,
approximately  96% of the vaporization is driven  by heat  generation,
and the  remainder by inlet air unsaturation  (Appendix A-5).   Thus we
may focus  on expression  (iv).

      The only  part of expression (iv) corresponding  to a manipulable
physical analogue is the coefficient  m,  which corresponds  to ventila-
tion  rate.  Thus, m  represents a means of  matching heat  removal  to
heat  generation in reference  to  a constant,  activity-promoting,
temperature.

      Now consider in theory  two  ways  of  controlling  m which  predict
markedly different behavior  patterns  (TABLE  8).  In  the  first  (see
"Prediction" columns), designated R,  m is  varied such that the pile's
outlet temperature ascends to 60°C and" this value  is subsequently
maintained as  the operational ceiling.  The need for a variable  m
st(5ins  from the variable  rate  of  heat  generation  as caused by popula-
tion  shifts, nutrient depletion, available water depletion,  and  other
unidirectional changes characteristic of batch culture.  A period of
vigorous heat  generation ensues, yielding  a large  0  .  Consequently
the material dries rapidly,  ultimately terminating activity  for  lack
of microbially available water (Section 4).  The pile is well-oxy-
genated, as approximately 9-fold more air  is needed  to remove heat than
to supply 02 for respiration  (Appendix D-l).

     The -econd predictive theoretical control approach, designated B-,
does not involve deliberate  temperature management.  Rather, m is con-
trolled such that interstitial 02 is not less than 5% (v/v) .  This is
easily accomplished through minimal ventilation on a fixed schedule
as it  leads, by default,  to  inhibitively high temperature, suppressing
02 consumption.  Although water  removal per unit air is comparatively
high  (large h  t), owing to a small m overall removal (0 ) is slight.
Thus, approacR B leads to inhibitively high temperature^nd  a prolonged
period of low-level activity.  Moreover,  attempting  to increase  0
by increasing m is not successful.   Since heat output, is slight, the
increased ventilation cools the pile, decreasing h
                                                 ~

                                 97

-------
If.'
                 TABLE 8.   BEHAVIOR OF THE COMPOSTING ECOSYSTEM AS INDUCED BY DIFFERENT APPROACHES TO THE
                           CONTROL OF m (VENTILATION MANAGEMENT).  PREDICTION BASED ON EXPRESSION iv
[fly - 0.9 m (hQut - hin)l, AS CONSTRAINED BY THE INTERACTION BETWEEN HEAT GENERA-
TION AND TEMPERATURE, 'BSERVATION BASFD ON SIDE-BY-SIDE 36 TONNE PILES (PILES
9A AND 9B).



m (m air delivered/wet metric ton)
Total
Peak (hr'1) 1
Mean (hr'1) J
Temperature (°C)
Outlet air
Pile: median (range)
Pile: lobs > 60°C (number of obs)
Heat
Qv
h . (kj /kg dry air) #
— o y £
hin (kj/kg dry air)$
Decomposition
Sludge volatile solids decrease (1)
Water removal
Moisture content decrease (5)
Mass H20 removed (% of initial)
kg H?0/kg dry air
Prediction

Anoroach R Aporoach B'

• Large Small
Peak>mean Peak=mean


60 (by design) SO (by default)1



Large Small
477 1560
74.9 74.9

Large Smal^

Large Small
Large Small
0.1404 0.5394
Observation
Rutgers ^ Seltsville
strategy r>rocessi"

9660 2350
135 6.06*
31.5 4.69*


53(24-68} 70(45-82)
13(1755) 91(1519)





72.3 15.4

67 -»• 29 65 -> 61
8C.9 19.4

                   kg H20/ra^  ambient  air
                   (continued)
0.05S6
0.0793

-------
TABLE 8. (continued)
Processing period (days)
Respiration
Rate 0- uptake
0,, level (§ v/v)
CG^ level (% v/v)
Prediction
*
Approach R
Short

High
High
Low
Approach B
Long

Low
High
Low
Observation
Rutgers Beltsi'ille
Strategy Process
15. S 21

.161 .16%
1,4 % i/45
 Time-variable, interactive, blower operation (temperature feedback control) in reference
 to an operational ceiling of 60 C.
"^Fixed schedule blower operation.
'^Difference reflects manual adjustment of schedule (9).
sThe ecosystem brings itself to the edge of its temperature tolerance limit.
 Relative humidity (RH) assumed to be 1005.
^Temperature and RH assumed to be 20°C and 50%.
S

-------
      The  physical  analogues  of  R  and  B  are  the  Rutgers  and Beltsvi1le
 approaches,  as  already compared in  detail  in  previous Sections.   In
 TABLE 8,  the  comparison is developed  in terms of equation (iv) .   Note
 that  where  existing  data  permit evaluation, prediction  i? con firmer7
 by  observation.

      Additional  confirmation  comes  from responses  to sudden chanees
 in  ventilation.  In  the Rutgers strategy heai generation  is intense,
 hence accidental loss  of  blower function is predicted to  induce  a
 tainperatm a upsurge.   This response was obser/ed numerous times.   In
 the Beltsville process  heat  generation  is weak,  he:ice a sudden  -'ncrease
 in  ventilation should  induce  a  temperature  downturn.  Moreover,  increased
 ventilation is not expected  to  substantially  enhance drying, as  the de-
 creased h   +  tends t-o  offset  the  increased m, moderating  any change in
        —-                                  ~*"
 0  .  This behavior was obseived  in an  unsuccessful  attempt  to  improve

 drying  in field-scale Beltsville pilt~ by  increasing ventilation  4-fold
 partway through the process cycle  (10) >  as described in Appendix  D-2.
      In the Rutgers strategy the system  is not permitted to  self-limit
via inhiliitively high temperature, but rather is prompted to do so via
substrate and water depletion.  Our experience is that demand for ven-
tilation terminated on day 7.1-IS.8 with a mixture of primary sewage
sludge and wnodchips, and on day 4.3-9.5 with a mixture of the sludge
and recycled compost.  While the immediate cause of termination was
dryness, the addition of water provoked  only a weak revival  of demand.

     Thus, the essence of composting process control is given by the
expression Qr = 0.9 m (h    - Jj-n)j as constrained by the interaction
between heat generation and temperature.  Mathematically, in isolation
from this constraint, it might seem thai; a high value of Qy  is obtain-
able through a high value of n, or h,,, . , or both.  This is unrealistic
                             -~     — out
however, for two reasons.  First, ordinary values of _h.  dictate that

an arbitrarily high m leads to a low h   t. (see Secticn 8 - next). Second,
values Oj,' h  t representing temperatures higher than 60°C inhibit heat

generation.   Consequently, the role of m is defined as that of matching
heat ~emoval to heat generation in reference to temperature, such that
maximal sustainable values of h  t and 0  are realized.

     This mathematical development constitutes a formal rationalization
for on-demand ventilation via temperature feedback control.   This is
the basis of the Rutgers strategy for composting process design and
control.
                                100

-------
                              SECTION 8


 DRYING ASSOCIATED WITH COMPOSTING,  AND NON-BIOLOGICAL AIR DRYING


 INTRODUCTION

     The analysis of Section 7 is  extended  here'-in  to  include  the'be-
 havior of a biologically inert, ventilated,  pile  o£  organic  material
 ("air drying").   This  leads to a  reinterpretation of the  Rutgers and
 the  Beltsville approaches,  in a broader context.


 THEORETICAL TREATMENT

 Simplifying Assumptions

     Consider the air drying process  as  a closed system  comprised
 of a pile (not mechanically'agitated)  of biologically inert  material
 and  the air which is passed through  the pile.  Neglecting gains  and
 losses  of heat through the  mechanisms  of radiation and  conduction,
 and  fractional gains resulting from  the passage of air  through the
 matrix,  the system (pile  and  air) experiences no  change in enthalpy
 (- isenthalpic).   However,  if the inlet air  is unsaturated,  a redis-
 tribution of enthalpy  occurs  within  the system through  the vaporiza-
 tion of water  from the matrix into the  flowing air.   This  results
 in a decrease  in the enthalpy of  the pile  exactly balanced Ly an
 increase  in the  enthalpy  of the air.

     Similarly, consider a static  composting  pile managed  according
 to the  Rutgers strategy and the air passed through it as  a closed
 system,  accepting the  same  simplification with respect  to  radiant,
 conductive,  and  frictional  factors.  Unlike  air drying, the compost-
 ing  system  is  not  isenthalpic; rather,  the system's enthalpy increases.
 This  is  because  chemical  bonds are broken through biological action,
 releasing heat.   In common  with air drying any unsaturation of the
 inlet air causes  a  redistribution of enthalpy.  This  contributes
 negligibly  to  *he  enthalpy  increase of  the air, however, compared to
 biological  heat  generation.

    Two  further  simplifications are made for the purposes  of this
 exercise.   First,  the redistribution of  enthalpy originating in the
 unsaturation of  the inlet air  is neglected.  That is, in the air
 drying examples  the air is  treated as if it were isenthalpic.  Rela-
 tive to actual conditions this overstates the temperature  decrease
 experienced by the air, and understates  the water removal per unit
mass  of air.   It does not affect the composting examples with respect
 to these factors.  Rather,  the effect is a small overstatement of
 the mass of air needed to accomplish a  given -mount of cooling.   These
 simplifications are adopted to permit the use of standard psychometric
data  (46).


                               101

-------
      Next,  outlet  air  RH values  must- be  assigned.   For  air drying
 this  is  not  a problem  as a  necessary condition (short of "break-
 through")  is an  outlet  RH of  1001.   For  composting  an outlet  RH
 of  100%  is  adopted,  in1 the  belief  that this  is a  valid  approximation.

      Neglect of  the  enthalpy  redistribution  originating in unsaturated
 inlet  air  introduces a  bias favoring composting relative to air dry-
 ing, with  respect  to the removal of  water  per  unit  mass of air.   The
 stipulation  of saturated composting  outlet air might introduce  a
 further  bias in  this direction.  We  believe  that  the bias(es)  are
 minor  in the context of the exercise, and  that the  hypothetical
 examples provide useful approximations of  the  difference in perfor-
 mance  to be  expected of air drying and composting,  with respect to
 water  removal.   Regardless, independent  verification based on  field
 data  is  offered  later  in this  Section.
     As was  already  seen  in Section  7,  regardless  of whether  the
driving force  is heat  generation, unsaturation  of  the  inlet air,  or
both, vaporization can be described  as  follows:
where:    v a mass vapor flow  (mass moisture/time)

          m s dry air mass flow  (mass dry  air/time)

          uj - humidity ratio  (mass moisture/mass dry air)


The outcome of this relationship is exemplified in the non-biological
air drying system using various  inlet conditions (TABLE  9) .

     If the inlet air is saturated (e.g. 7.2°C- 100% RH) no change
in temperature and RH occurs with passage  through the matrix,  and
no moisture is removed (Aui s 0).  If the inlet air RH is less  than
1001, water is vaporized from  the aqueous  matrix into the  flowing
gaseous phase until an RH of 100% is reached.  The vaporization
from the matrix causes it to cool, and this  is translated  into a
cooling of the air as its temperature equilibrates with  that of the
matrix.  Thus, in the air drying process the magnitude of  moisture
removal is solely dependent on inlet air conditions, and the only
factor subject to process control is m. This may be increased  to
compensate for poor drying air.  For example, roughly equal vapori-
zation rates are obtainable with air at 32°C-20$ RH or 32°C-90% RH,
by using a 10- fold greater m for the more  humid air.

     Heat drying (see last entry in TABLE  9) is a special  case of
air drying, in which the ambient air is preheated in an operation
external to the pile.  A gas-fired hot air generator, for  example,
might be used to condition the inlet air.  Coropared to the ambient
air, this results in a larger Am.
                              102

-------
TABLE 9.  HYPOTHETICAL CHANGES IN AIR AS EFFECTED BY THE AIR DRYING PROCESS (BASED ON
          ISENTHALPIC CONDITIONS - SEE TEXT)"
Inlet air
Process
Air drying
Air drying
Air drying
Air drying
Air drying
Heat
dryingt
T
4.4
7.2
7.2
32
32

60
RH
111
60
100
20
20
90

10
ID
(leg H20/
kg dry air)
.0031
.0064
.0012
.006
.028

.013
Inlet and
outlet air Outlet air
h
(kJ/kg
dry air)
30.2
41.2
28.8
65. 1
121

111
1.7
7.2
0
17.3
30,8

29,4
RH
HI
100
100
100
100
100

100
(kg H20/
kg dry air)
.0042
.0064
.0038
.0123
.0286

.026
u/m = Ato'
Tkg H20/
kg dry air)
.0011
0
.0026
.0063
.0006

.0124
ft
   Values of h and w derived from Reference 46.

t  The  inlet conditions are representative of crpp drying applications  (47)
   is a special case of air drying  (See text).
Heat drying

-------
      Whereas in air drying (and its variant heat drying) the unsatura-
 tion of the inlet air is the  only force driving vaporization, in
 composting this factor is minor.   Here the major factor is the meta-
 bolic generation of heat in the matrix, as follows.

      Metabolic  activity in the aqueous matrix phase  generates heat,
 establishing an enthalpy differential between the solid-liquid matrix
 and the gaseous phases (see Section 3).  The differential is sustained
 by the  flow, which brings in  cooler,  drier, air.  The differential
 drives  vaporization,  and establishes  a positive temperature gradient
 along the  axis  of airflow.   Thus, as  long as heat generation persists,
 the enthalpy of the air increases with passage through the pile.   The
 upper limit to  this condition is  defined by the heat  generation-
 temperature interaction.   The major components of the enthalpy increase
 are i)  an  increase in the air temperature, and ii) an increase in the
 water content per unit mass of air.

      Therefore,  in composting the rate of vaporization is  a function
 of heat  flow.   As was derived in  Section 7S this leads to  the approxi-
 mation:

            Qv " °-9 » (iout-hin)  " °-9 £ A^                      (ii)

 where:      0 =»  heat  flow associated  with vaporization (energy/
                 time)
         -out  " outlet  a^r  -nthalpy  (energy/mass dry  air)

               = inlet air enthalpy  (energy/mass dry air).
An ef^'jct: of inlet conditions on waiter removal per unit air during
composting is implicit in expression  ii,  in that ambient  temperature
and RH determine the value of h- .   (But, variations in h.  are
automatically compensated for Tn the  Rutgers strategy through tempera-
ture-feedback adjustment of m. )  The  effect of inlet conditions is
small, however, as ceen in TABLE 10=  For example, only 1.2 times as
much moisture is removed per unit mass of air by the "best ambient
drying air" (7.2°C - 20% RH) than the "worst air" (32°C - 90% RH) .
This illustrates that, given informed process control, composting
process performance is not sensitive  to ambient conditions.   (Extreme
cold is a separate potential problem  - see later.)

     Thus, the amount of moisture removed per unit mass of air is much
greater through composting than air dry ig.   In this hypothetical exer
cise the amount of water removed through composting is larger by the
following factors:   135x,<«>, S8x, 23x, and 207x (compare the Au's in
TABLES 9 and 10).   The factor of infinity is obtained when the inlet
air is saturated.   Compared to the heat drying example, the composting
Aw's are lOx to 12x greater.

Continuum Among Air Drying, Rutgers Strategy, and Beltsville Process

     It is now possible to define the differences among air drying,
the Rutgers  composting process control strategy,  and the Beltsville
                                104

-------
 TABLE 10.   HYPOTHETICAL CHANGES  IN AIR AS EFFECTED BY THE  COMPOSTING PROCESS

Inlet air

T
RH
h

0 (kJ/kg
(
4.
7.
7.
32
32
Q.
4
2
2


r ••"
60
100
20
20
90
dry
30
41
2S
65
121
air)
.2
.2
.8
.1


(kg H20/
kg dry air)
.0031
.0064 •
.0012
.006
.028
Tt
1 '
( C)
60
60
60
60
60
Outlet air

RH
111
100
100
100
100
100
h
(kJ?kg
dry air)
477
477
477
477
477
to
(kg H20/
kg dry air)
.152
.152
.152
.152
.152
v/m=Au

Tkg H20/
kg dry
.149
.146
.150
.146
.124
air)





Values of h and w derived from Reference 46,  and where necessary by calculation.



Based on the outlet temperature characteristic of the Rutgers strategy.

-------
composting process in terms of ventilation.  This is done in reference
to expression ii.  The hypothetical exercise involves a static pile of
suitably porous, energy-rich, organic material, different ambient air
conditions, and manipulation of m.  The constant in expression ii is
temperature-dependent and would Have to be changed as appropriate,
were numerical solutions sought.  This is not a source of error in the
present, qualitative exercise, however, as its outcome is not affected
by the value of the constant.

     Cases 1 and 2 involve a very high, arbitrary, value of m, such
that the material is suspended in the airstream ("fluid bed"J.  In
Case 1 the ambient air (inlet air) temperature is 1°C and its RH is
100%.  Because the temperature is biologically unfavorable, heat
generation is very slight.  Because any heat that is generated is
quickly removed, the temperature does not. increase to a more favorable
level.  Consequently, for all practical purposes 0. ° 0, and the
system's behavior is described by expression i.  However v also is
zero, as this is the special case of saturated inlet air.  Were the
RH<100$, evaporative cooling would ensue, v would be grea    than
zero, and 0  would be zero.

     Case 2 also involves a very high m, but differs in that the
ambient air is 30 C-100% RH.  Since the temperature favors biological
activity heat generation ensues, but the heat is promptly removed and
no appreciable temperature elevation results.  Nonetheless, the heat
generated drives vaporization, h  t>n.  » and the system's behavior is
described by expression ii.  Although Cases 1 and 2 are o£ theoretical
interest, they would represent extravagant use of energy for ventila-
tion.

     Cases 3 and 4 involve composting in the ordinary sense, in that
a temperature elevation is experienced.   Expression ii pertains in
both eases.

     Case 3 is that of the Rutgers control strategy, in which the
value of m is continuously adjusted through fee.dback control to seek
an outlet temperature of 60°C.  Either set of inlet conditions (1°C
-1001 RH, or 30°C -1001 RH) initiates self-heating, although a slow
start is experienced with the colder air.   Onr.e underway, a quasi
steady-state is  established in which a sustained high rate of decom-
position and vaporization is realized.

     Case 4 is that of the Beltsville Process,  in which m is set at
some  low fixed value consistent with the maintenance of a minimal
residual 02 level.   This  leads to the highest value of h  t support-
able  by the system (but note the low m),  and a commensurate inhibi-
tively high temperature (^80°C).   Consequently, a quasi steady-state
is established characterized by low rates  of decomposition and vapori-
zation.   This condition is signified by a  low £) .
                                106

-------
 FIELD  EXPERIMENTATION

 Ml!J:££iiLLl_Ji£J:LJ!2^

     The  experimental  study concerned  a  single  pile  formed  of material
 previously  composted by  the Rutgers  method.   This  was  a  mixture of
 the  screened  (woodchip-free)  material  from  piles  8 and 9A (Section 3)
 which  had not  been used  in  the  11  series  piles  (Section  S),  and all
 of the material  from piles  11B  and 11C.   Prior  to  its  use in the
 present study  the  material  was  stored  in  the  open  for  3i  to  8j months.
 During this period much  of  it became moist, presumably resulting in
 further stabilization  through informal curing.

     This material was  formed into a pile of  approximately  2.2 tonnes,
 designated  as  pile 12,   It  had  a starting moisture content  of 65$.
 The  pile  was fitted with one  blower, thermocouples,  and  a gas sampling
 probe  (Figure  64).  Neither a thermistor  nor  a  timer was  required,  as
 the  experimental plan  called  for continuous operation  of the blower
 (100%  time  on).  The forced pressure mode of  ventilation v/as employed.
 Samples for the  determination of moisture content  were taken regularly
 from the  part  of the pile slightly below  that represented by position  S,
 On some occasions  the  lower part of  the pile  represented by  position 1
 v/as  also  sampled.

     Time-zero was 27  February  1981  and  termination  was  on  18 March
 1981.   During  this period the ambient  temperatures were  (°C):  mean
 of the daily highs, 9°;  mean  o£ the  daily lows,  zero0; mean  of the
 daily  means, 4.5°;  range of the daily  highs and  lows,  -7.8°  to 13.9 .
 Rainfall  amounted  to 7 cm in  four  occurrences.

 Results

     The  temperatures  of the  ambient air  and  pile  position 1 are
 plotted in the lov/er graph  in Figure 65,  and  those of  positions  1 and •
 5 are  plotted  in the upper  graph in  this  figure.   (The other tempera-
 ture data are  given in Appendix E.)  To help  bring out the trends the
 •plotted data are also tabulated in the form of the mean  differential
 values  for each  24  hour  interval (TABLE 11),

     Position  1 tended to be  cooler  than  the  ambient air  during  the
 first  240 hours, notwithstanding invervals to the  contrary  (e.g. hr
 130  to  160) and intervals of  identical temperatures  (e.g. hr 180 to
 200).   Starting at  hr 280 position 1 tended to be  warmer  than  the
 ambient air.  Positions  1 and S v/ere at similar temperatures during
 the  first 190 hours, and thereafter  position  S was cooler.   Position 5
was generally cooler than the ambient air.

     The  0, and C02 levels  in the pile v/ere not distinguishable  from
the ambient values.

     The moisture  content data  are given  in Figure 66.   At the upper
sampling  level, the moisture  content did  not  change during the experi-
mental period.   The lower level experienced a moisture content decrease
subsequent to hr 140.   At hr  467 an  unusually low  value  (11.2%)  v/as
noted.

                                107

-------
o
oo
        Figure 64.      Pile 12, Cross-sectional representation:  textured area,  woodchip  cover
            and base; clear area, compost; circle, perforated  flexhose.  The  numbers  indicate
            monitoring positions: thermocouples, positions  1 through  6; gas sampling  probe,
            adjacent to position 5.  The blower was operated continuously  (100%  time  on)  in
            the forced-pressure mode.

-------
         15
                                                         t 1  1 1 I  II 1 ! J	I	) \  I I  ! 1  t i
                                                        LJLjLJLJL-jl—JLJLJLJLJ-JLJ-JUJrf.»A^L^i«JU -,
         -5
                                          HOURS
Figure  65.
     and 5.
Pile  12, temperature  external  to the  pile  (ambient)  and at  positions 1

-------
TABLE 11.  TEMPERATURE DIFFERENTIALS AMONG AMBIENT, POSITION 1,
           AND POSITION S PROBES*

Interval
(hr)
0-23
24-47
48-71
72-95
96-119
120-143
144=167
168-191
192-215
216 239
2^0-263
264-287
288-311
312-335
336-359
360-383
384-407
408-431
432-455
Position 1
-ambient
™l!£L_
-3
-1.8
-1.8
-I.I
-0.6
-0.2
1.2
-0.5
-0.5
-0.4
-0.5
-0.1
1.3
-0.4
0.3
1.6
2.3
3.0
3.0
Position S
-position 1
(°C)
0.6
0.2
0.7
0.7
0.2
-0.1
0.1
0
-0.7
-0.6
-1.1
-0.2
-0.7
1.1
-2.4
-3.4
-5.3
-1.4
-2.8
Position 5
-ambient
___ffiL_
-2.4
-1.6
-1.1
-0.4
-0.4
-0.3
1.3
-0.5
-1.2
-1.0
-1.6
-0.3
0.6
-1.5
-.2.1
-1.8
-3.0
-1.6
0.2
ft
 The temperature was  recorded every hour.   The  differential values
 are based on the mean of the twenty-four  hour  interval.
                                110

-------
      701
      60
      50
     52
     o
     u
     o
     e:
     u
     a
      30
      20
                          7© CM HIGH LEVEL

                          40 CB3 HIGH LEVEL
                                         -1
                      100
200           300

   TIME IN HOURS
                                                               400
                                                                             500
Figure 66.      Pile 12, moisture content,  samples taken  from central  interior

    locations.

-------
     At  termination  the pile was bisected  for visual  inspection
 (Figure  67).  A hemispherical dry  zone, with its -origin  at  the
 ventilation duct,  comprised approximately  half  tha cross-sectional
 area.  This terminated in a 2 cm wide transition zone, evidencing
 a moisture gradient.  The upper half of the cross-sectional area
 was  frankly moist.   With respect to other  characteristics  (granula-
 tion, odor, color),  the material at termination did not  differ
 noticeably from the  starting material.

 DISCUSSION
     The subject of the experiment was a ventilated pile of £omp_os_t_1
as distinct from the usual ££JS££s^ijig pile. The purpose was to
investigate air drying withautTEe~complication of biological heat
generation.  To approach this condition the pile was formed of material
previously composted by the Rutgers method, and subsequently cured
informally for a prolonged period.  As such, the material was depleted
o£ readily metabolizable substrate, and potentially supportive of
only slight raicrobial activity.  The timing of the trial was fortui-
tous in  that the ambient air temperature was unfavorably low for
microbial activity.  Furthermore , the strong ventilation imposed
(though not comparable to hypothetical cases 1 and 2) prevented heat
storage, which might otherwise have elevated the temperature to a
biologically more favorable level.  Judging from these conditions and
from the pile's behavior as discussed below, heat generation was
negligible as intended.

     In the virtual absence of heat generation the system is expected
to behave as follows.  A well-defined cooling-drying front develops
at the air inlet point and expands radially along the axis of airflow.
Below the front the moisture content of the material comes into equi-
librium with the RH of the ambient air; above it the moisture content
is unchanged from time- zero.  Below the front the temperature of the
material and interstitial air is at the ambient level (except for
frictional heat input - see later); above it the temperature is less
than ambient.

     This behavior is predicted because the drying and cooling both
result from vaporization — driven solely by unsaturation of the inlet
air.   In this  circumstance the unsaturated condition is abruptly ter-
minated through the action of two complimentary forces at the drying-
cooling front:   1)  the transfer of moisture into the gaseous phase,
saturating it,  and 2) the cooling of the gaseous phase as it comes
into temperature equilibrium with the matrix, decreasing its capacity
to hold moisture at saturation.  In combination, these forces bring
vaporization and cooling to a sharp halt5 hence a narrow, well-defined
front.   This cooling-drying front migrates through the pile with time,
leaving in its  wake an expanding region of air-dried material.   The
material ahead  of the front is undried.
                                112

-------
by P. C.  Millr
        3t t«.in.tion.  Note
         and drV lo«er area.  Photo
113

-------
     These predictions were verified  in  the  experimental  observations.
Cooling  is evident  in the  temperature  differentials between:  position
1  and  ambient; position  S  a,:d  1; positions 5  and  ambient.   The  pattern
of the differentials is  suggestive of  a  cooling front passing posi-
tion 1 roughly midway through  the experimental time period,  but not
reaching position 5 by termination.  This is  inferred from  the  sign
change,  from negative to positive, in  the differential between  posi-
tion 1 and the ambient temperature, and  the  opposite change  in  the
differential between positions S and 1.  Thus the temperature de-
creased  and the decrease was discontinuous,  as expected of  a  system
in which biological heat generation is negligible.

     In.  stipulating isenthalpie air in the hypothetical air  drying
example, it was noted that this exagf  ~ates  the temperature  decrease
(see Simplifying Assumptions) .  This '!._,  because the exercise  is based
on stan3ar3~psycHomet,ric 3at¥, which do  not  take  into account the inter-
action between the gaseous phase and the liquid-solid phase  (the air
is  cooled by the matrix).  Additionally  there is  a second source of
error  in the same direction, in the form of  frictional heat  input,
resulting from the  forcing of  the air  through the niitrix.   In the
virtual  absence of biological  heat generation, frictional heat  could
be  significant.  Neither of these errors can be quantified at present.

     It  is possible to develop an exercise, not without its  own diffi-
culties, comparing the theoretical isenthalpic cooling and the  observed
cooling.  The exercise is as follows.

     During the experimental period the mean ambient temperature (mean
of  tae daily means) was 4.5°C.  Relative humidity was not measured,
and the ambient value no doubt fluctuated widely.  Nevertheless, if
an  RH of 60% is taken as a, representative ambient value, and  if inlet
conditions were consta\ ^ at 4.5°C - 60% RH, an isenthalpic temperature
decrease of 2.S°C is predicted.  The data offer three opportunities
for comparison to the hypothetical value, as follows: position  1 minus
ambient, from time- zero to hr  215; position S minus position  1,  from
hr  216 to termination; position S minus ambient, for the entire period.
The observed mean differentials are, respectively, (°C) : -0.8,  -1.9
and -1.0.  Thus observation conforms to theory in that the means of
the data are negative, and the absolute values are less than  2,5.
              at hr 280, above- ambient temperatures were noted at
positio;  1 (TABLE 11).  This is attributed to frictional heat input.

     Thus, aespite the complications inherent to this exercise, the
temperature data are suggestive of the migration of a cooling front
through the pile.

     The moisture content data provide equally suggestive data for the
drying component of a cooling-drying front ("igure 66).

     Perhaps the most compelling evidence, however, is in the form
of the visual appearance at termination (Figure 67).  This revealed
unambiguously that a distinct, radial, drying front had indeed passed
position 1 but had not reached position 5.  It is concluded that the
observations are as predicted by the model of the non-biological air
drying process (expression i).
                               114

-------
Water
     Water  removal  per  se  is  a  goal  in  sewage  sludge  treatment.   It
 is  therefore  of  interest to  compare  composting and  air drying  from
 this narrow perspective.   This  comparison  neglects  the other treatment
 goals  advanced through  composting.

     Consider the two approaches  to  composting.   In the Rutgers  stra-
 tegy,  one-third  less water was  removed  per unit  air delivery.  This
 reflects the  characteristically lower value of hQut-  However, owing
 to  the  high proportion  of  water removed and the  speed of its removal,
 only the Rutgers strategy  affords  a  useful means  of drying sludge.
 (Also,  see  Appendix D-2).

     Compared to non-biological air  drying, the  biological system
 (Rutgers Strategy)  removed 22. 3x  more water per  unit  air delivered,
 This reflects the dominant role in composting  of  heat as the force
 driving vaporization.   Unsaturation  of  the inlet  air plays only  a
 minor  role.   In  air drying,  however, inlet air unsaturation is the
 only operative mechanism.  In addition  to  its  efficiency in terms of
 air usage,  composting removes a higher  proportion of  the water,  and the
 removal is  faster.  This .comparison  is  expressed  in terms of the cost
 of  pumping  air (TABLE 12.)


 TABLE  12.   COST  OF  WATER REMOVAL  THROUGH BIOLOGICAL AND NON-BIOLOGICAL
            MEANS





Process
Composting-
Rutgers d
strategy
Composting-
Beltsville
processt
Non-biological
air drying*
Water
removed/
unit air
delivered
(ton x
10-6/ft3


1.85


2.48

0.0828



Process-
ing time
-iiazsJL.


9.8


20.8

18.8
Propor-
tion of
•. initial
water re-
moved
_£!!_


75.7


19.4

46.7


Cost o£ air
delivery/ur.



it
water removed
C$/ton)5


0.31


0.21

6.33









 Based on piles 7, 8, 9A, 11A, 11BS 11C.

"("Based on pile 9B,
±Based on pile 12; assumes that half the material was air dried at
 termination, and half unchanged from initial moisture content (see
 Figures 66 and 67).

§Based on electricity @ $0.06/kw-hr.
                                115

-------
                                SECTION  9


                COMPARISON  BETWEEN THE RUTGERS  AND  BELTSVILLE
                 CONTROL  STRATEGIES:  EFFECT  ON CURING  STAGE


 INTRODUCTION

     The greatest benefit  from  formal process  control  is  realized
 early in the composting, when readily metabolizable  organic sub-
 strates are most abundant  and the potential  for decomposition  is
 greatest.  As readily  available  substrate  is depleted  it  becomes
 feasible to continue the processing, if necessary  to do so, on a
 less formal basis.  The  earlier  period  may be  called the  high  rate
 stage, and the  later period the  curing  stage.

     The present section concerns the curing of materials previously
 composted by the Rutgers and Beltsville methods.

 MATERIALS AND METHODS  SPECIFIC TO SECTION  9

     Material from piles 9A and  9B  (Section  3) was screened, to
 remove woodehips, with a Royer Model 365 shredder/mixer,  coupled
 with a Mogensen sizer  (Royer Foundry and Machine Co.,  Kingston, PA).
 Screened-material was  transported from  the primary composting  site
 at Camden, N.J, to New Brunswick, N.J., to form two separate curing
 piles.   Each pile consisted of approximately 3m^ of material in a
 conical shape.  A stainless steel dial  thermometer of  1 meter  length
 w°" inserted into the ^center.of  each pile, where it remained through-
 out the trial.  The experimental chronology and related matters are
 given in TABLES 13 and 14.

     On two occasions the piles  were remoistened with  water from a
 garden hose.   As part of this operation the material v/as  turned and
 mixed by shovel.

     Samples  v/ere removed periodically  from_the pile interior  and
 tested qualitatively for NHt,  NO;*  and N03-   The reagents were
 as described in Standard Methods under  items 132B and  134 (48).
Devardas alloy was used as a reductant  in  the  test for NO^.

     Samples  from the pile interior v/ere subjected to  an  odor  test
on the  day of sampling.  The material was placed into  1 pint
 (ca.O.SJl)  screw-cap jars, such that they were  one-third full.   The
jars were coded to conceal sample identity, and randomly  selected
 individuals (excluding project personnel) were asked to evaluate the
samples  on a  sc.  ' ,• of -5 to +5 (-5 « most unpleasant;  0 - neutral;
 + 5 = most pleasant).   Tv/enty people evaluated  the first set of
samples, and  30 people the second set.
                                 116

-------
       TABLE 13.  PERTINENT DATES, ELAPSED TIME, AND WEATHER
Time zero (parent pile formation)

     Rutgers:  9 Jul 1980
     Beltsville:  11 Jul 1980

Parent pile screening*

     Rutgers:  day 36
     Beltsville:   day 26

Curing pile formationt

     Rutgers:  day 40
     Beltsville:   day 26

Termination (10 Nov 1980)

     Rutgers:  day 124
     Beltsville:   day 122  .
Weather (1 Aug to 10 Nov)*

     Ajnbient air temperature (°C)

        High:   25
        Low:    14
        Mean:   20
        Range:   -S to 37

     Rainfall

        Amount (era):  21.6
        Occurrences (no.):  IS
*At Camden, N.J.

^At. New Brunswick, N.J.

 This is taken to represent the conditions following the start of the
 curing stage.  See Table 2 for conditions during the high rate stage.
I
 High = mean of the daily highs; low - mean of the daily lows; mean =•
 mean of the daily means; range ™ overall range of the daily highs and
 lows.
                                117

-------
TABLE 14.  MANIPULATIONS OF THE CURING PILES, AND CHARACTER OF
           THE MATERIAL
Moisture content (!) and water addition
     Rutgers parent pile: day 0 to 19, see Figure 36;  day 24, 251
     Rutgers curing pile: day 76, water added;  day 84, water added;
                          day 90, 631; day 107, 64*
     Deltsville parent pile:   day 0 to 21, see  Figure  36;  day 22, 65$
     Beltsville curing pile:   day 74, water added; day 8"2,  water added;
                              day 88, 584; day  105, 53%
Odor test (mean score)
     Rutgers curing pile:  day 86, • 1.30; day  91, + 0.66
     Beltsville curing pile:   day 84, - 3.15; day 89,  - 2.29
First detection of NO,
     Rutgers curing pile:  day 86
     Beltsville curing pile:   day 113
First detection of NO:
     Rutgers curing pila:  day 99
     Beltsville curing pile:   day 123

*
 The piles  were turned as part of the water addition operation.
                               118

-------
RESULTS

     Because of  its physical accessibility  the parent  Beltsville
pile (9B) was  screened  first, permitting  the earlier formation  of
its derivative curing pile  (TABLE 13).  This curing pile was moist
from the outset  (see Figure 68 and TABLE  14).  A subsequent period
of rainfall delayed the screening of the  Rutgers pile  (9A) , but
the material eventually obtained was from the interior of  the pile
(unaffected by the rain).  Hence this derivative curing pile was
dry at the outset  (Figure 36 and TABLE 14) .  In addition to the
moisture content data,  the dry condition  of this curing pile over
much of the trial was indicated by visual and tactile  examination
of the samples obtained periodically for  the nitrogen  spot tests.
On day 76 the material was wetted by the  addition o£ water and
turning,

     The Rutgers curing pile cooled more  quickly and in a more
regular pattern  than its Beltsville counterpart (Figure 68).  In
the Rutgers pile a slight, temporary, temperature descent coincided
v/ith the first water addition-turning operation (day 76), but not
the second such  operation.  In the Beltsville pile sharp descents
and re-ascents coincided with both addition-turning operations.

     Based on the average test panel score, at the time of the
first odor test  the material from the Rutgers curing pile was less
unpleasant than  the Beltsville material (TABLE 14).  At the time of
the second odor  test the freshly sampled  material from the Rutgers
pile was rated at the low end of the "pleasant" range.  Although
the material from the Beltsville pile was improved v/ith respect
to odor, it still v/as considered distinctly "unpleasant."
     The end products of both steps of nitrification  (NOZ  and
NOj)  appeared earlier in the Rutgers pile  (TABLE 14).

DISCUSSION

     In the parent pile 9A there was intensive heat generation
(decomposition) and vaporization, hence the derivative curing pile
was formed of moderately well-stabilized, dry, material  (TABLE 14).
Because of the dryness, activity was probably slight from approxi-
mately day IS to day 76 (when the material was wetted).  Thus,
dryness presumably delayed curing by as much as two months.   In
routine operation timely water addition would be indicated to prevent
a curing hiatus.

     In contrast, parent pile 9B experienced inhibitively high
temperatures, with correspondingly slight decomposition and vaporiza-
tion.   Its derivative curing pile was formed of poorly stabilized,
moist,  material.  Although direct observations are lacking,  we
suspect that activity in this pile v/as O^-limited for a part of the
curing period.
                                 119

-------
ts)

O
                80
                60
a:

£40
2
bJ
                20 I
                                       PILE 9A-Q

                                       PILE 9B-0
                                                                                      -4
                                                                        1
                  40
                 60
80            100
   TIME IN DAYS
                                                                         120
                                                                        140
        Figure 68.   Curing piles, temperature in  the center of  the pile,

-------
     The circumstances at the start of the curing period may be
summarized as follows.  The material derived from pile 9A was ready
for curing, as it had reached a moderately well-stabilized condition.
Curing was delayed, however, by dry-ness.  The material derived from
pile 9B was not ready for curing, as prior stabilization was slight.

     Despite the dryness-induced curing hiatus, 9A material was first
to reach a well-cured condition judging by the cooling rate, odor,
and onset of nitrification.  (Although cooling is generally a sign
of substrate depletion, in the present case the issue is clouded by
the effect of the dryness on biological heat generation.)  Odor is of
obvious practical operational interest, as well as being indicative
of the degree of stabilization.  This odor test result is consistent
with other formal observations on odor (2, 29).

     The validity of nitrification as a sign of organic matter
stability is widely appreciated in the sewage treatment field, among
other fields (49).  This stems from several characteristics of the
responsible bacteria, such as their chemoautotrophic nature, sensi-
tivity to elevated temperatures, and slow grov/th.

     In routine practice the addition of water could be made part
of the transferral operation from the site of the high rate operation
to the curing site.  Alternatively, water might be added during the
high rate stage, as needed to sustain microbial action.  In this
manner the high rate stage would pass to the curing stage without
moving the material.  This mode of operation can be visualized by
reference to Figure 42.

     In designing a water addition program, it should be recognized
that the material becomes progressively more difficult to wet as it
dries.   As a. rough approximation, it might be advisable to add water
as the moisture content decreases to perhaps 40%.

     In employing composting as a waste treatment  technology it is
generally desireable, in the initial processing,  to strive for a
maximal decomposition rate.   Whether it is necessary to subsequently
cure the material is a site-specific matter.   This depends on the
nature of the waste and the intended avenue of ultimate disposal/.
resource-recovery.
                                 121

-------
                           SECTION  10


     RUTGERS CONTROL STRATEGY: DIAGNOSIS OF  PROCESSING  FAILURE


 INTRODUCTION  .

       The original intent of this  trial was  to examine the effect
 of wetness on composting by the Rutgers strategy, by isolating
 starting moisture content as the variable.  One batch  o£ sludge cake
 and  woodchips was mixed without addition of water, to  serve as the
 control, while water was added to two batches.  During composting,
 however, all three piles behaved similarly, exhibiting inhibitively
 high temperatures, low rates of activity and  decomposition, and slight
 drying.  These patterns indicated processing  failure by the standards
 of the Rutgers strategy.

       The suspected immediate cause of failure was inadequate pene-
 tration of air into the composting  mass, leading to inadequate heat
 removal.  Both a low amount of bulking agent  in the mix, and an overly
 large woodchip ventilation base contributed to poor air penetration.
 This outcome, though, lead to an unplanned  interpretative opportunity
 to contrast the behavior of a failed pile with that of a successful
 one  from a different trial.

 MATERIALS AND METHODS SPECIFIC TO SECTION 10

       Each pile consisted of approximately 6 tonnes of the sludge-
 v/oodchip mixture (excluding added water).    Pile 6A v/as formed using
 unammended sludge cake and woodchips (no water added).   For pile 6B
 tap  water was added from a garden hose at a "moderate  rate" as the
 sludge and woodchips were mixed in  the pug mill, and for pile 6C the
 rate of addition was "fast."  The ratio of.sludge to woodchips was the
 same in all piles (1 tonne sludge to 1.35m  woodchips).  The added
 water is not taken into account in  this ratio.  The controller set
 point for all of the piles was 45 C.

       The trial period was  13 March 1980  to  3 April 1980.   Ambient
 air  temperatures during this period were (°C):  mean of the daily
 highs,  13;  mean of the daily lows,  2; mean of the daily means, 11;
 range of the daily highs and lows,  -3 to 20.  Precipitation amounted
 to 14.4 cm in 1-2 occurrences.

       In a special terminal test (3 April)  of the penetration of
 air into the sludge-woodchip mixture of pile  6B, hot air was substituted
 for ambient air by use of a  kerosene-fired catalytic space  heater rated
at 24 x 106gm cal/hr (95,000 btu/hr).  The heated air was introduced
 to the  inlet of the blower,  which resulted in  air at 130°C  being intro-
duced to the flexhose.   Prior to the start of this test five additional
 thermocouples  were positioned at 0.3m intervals in the  woodchip base
                                122

-------
 midway  between  the  top  and  bottom of the base,  in a line perpendicular
 to  the  flexhose.

 RESULTS

        All  three  piles  failed.   One  of these  (6B) is compared to a
 previously  described  successful  pile of comparable size  (pile 7 - see
 Section  3).   Certain  details  not included herein are recorded in Sec-
 tion  3,  Appendices  A-l  and  A- 2 ,  and  Appendix  F.   Piles 6B and 7 are
 shown in cross-section  (Figure 69).

 Blowr
        For pile  6B  the  period  of  timer-scheduled  operation  lasted  until
 hr  138,  at which  point  temperature-feedback  control  came  into  play
 (Figure  70).   Blower  operation was  demanded  for  lOOt  of the time,  and
 the  demand did not  subside,

        For pile  7 the initial  timer period lasted until hr  56,  followed
 by  feedback control.  Demand built  to  a  peak of  65%,  at hr  110,  then
 gradually subsided.   Demand terminated at hr 344,  with the  resumption
 of  timer-scheduled  operation.

 Temperature

        In pile 6B the time-zero temperature  was  8-11  C  (Figure  71).
 At most  of the positions  the temperature ascended  gradually, though at
 an accelerating rate, so  that  at  the control  thermistor  (positio.  I)
 138  hours elapsed before  reaching .cet-point  (45  C) .   The  temperature
 pattern  at some of  the  outermo  t  petitions was erratic.   Feedback  con-
 trol did not arrest the temperature  ascent.   Whereas  the  temperature
 at the  thermistor should  have  stabilized at  45 C,  it  did  so  at  68°C.
 Elsewhere, higher temperatures  generally prevailed.

        In pile 7 the  time-zero  temperature was 18-22°C.   The tempera-
 ture ascended  faster, such that at  the control thermistor (position
 6) 56 hours elapsed before reaching  set-point  (45°C).  The  onset of
 temperature feedback  control arrested  the ascent  at the thermistor
 at 4S°C, and elsewhere  the design ceiling of  60°C  was rarely exceeded.

 Oxygen

       In pile 6B the lov/est 02 level  observed v/as 10% (Figure  72).
 The comparable observation for pile  7 v/as 14%.  Thus, in  both piles
 oxygenated conditions were maintained.

Moijture Content

       In pile 6B the moisture content decreased  from 72% to 68%
 (Figure 72),   The comparable decrease  in pile 7 was from  62% to  28%.
Thus, the drying tendency was strongly expressed only in  pile 7.
                                 123

-------
          PILE SB
PILE 7
Figure 69.   Cross-sectional representation of piles 6B and 7 (see also Section 3).  Tex-
   tured area,  woodchip cover and base;  clear area, sludge-woodchip mixture; circles,
   perforated flexhose; positions 1-16,  thermocouples in a 0.3 m x 0.5 m grid.  In pile
   6B the thermistor was at position 1,  and in pile 7 it was at position 6.  The wood-
   chip base dimensions (L x W x H)  were:  pile 6B, 5.2 m x 4.6 m x 25 cm; pile 7,
   3.1 m x  2.7  m x 25 cm.   The bottom-most level of the sludge- VOCK' chip mixture had
   areal dimensions (L x W) of 4.9 m x 4.3 m (both piles).  The piles  ere slightly
   longer in the axis parallel to the flexhose than in cross-section.

-------
         100
        P60
        fe

        a eo
        S40
        cc
        Q
         20
        O

        CD
            PILE 6B
                                                                  .  s
                  100
                        200
                               300
                                      400     0      100
                                         TIME IN HOURS
                                                           200
                                                                  300
                                                                         400
                                                                               500
Figure 70   Reprinted  by permission from TOXIC  AND HAZARDOUS WASTE:   Proceedings of the
   15th Mid-Atlantic  Conference, pp. 463-471.   Copyright c 1983.   Buckneli  Universi.v
   Blower ooeration,  mean of four hour  intervals (see also Section ^;.

-------
           PILE  6B
        801- PROBE 68-1 —
           PROBE 6B-I3 —
PILE 7
PROBE 7-6
PROBE 7-14
                       200
                              300
                                     400     0
                                        TIME IN HOURS
                                                    100
                                                           aoo
                                                                  300
                                                                         400
                                                                                500
Fioure 71   Reprinted bv permission  from TOXIC AND HAZARDOUS  WASTE:  Proceedings  o, the

   15th Mid-Atlantic Conference, pp.  463-471.  Copyright  c  1983.   Bucknell  University
   Pi^e temperature at selected  interior positions, plotted every four hours.   Contiol

   thermistors  were at positions 6B-1  and 7-6 (see also  Section j) -

-------
          80
          70
          SO
        S so
        o:
          40
          30
          20
                                            20
                                             16
                                             12
                                            02 CONCENTRATION
                                            PILE 6B--
                                            PILE 7 —
                  100
                        200
                               300
                                      400      0
                                         TIME IN HOURS
                                                    100
                                                           200
                                                                  300
                                                                        400
                                                                               5OO
Figure 72.  Reprinted by permission from  TOXIC AND HAZARDOUS WASTE:   Proceedings of the
   15th Mid-Atlantic  Conference, pp. 465-471.   Copyright, c 1985.   Bucknell  University.
                              samples taken  from the pile interior  (see  also Section 5.)
Left, moisture content  of
           concentration of 0
   probes were  at  position
                           2  (v:v) in gas
                         15 (pile 6B) and
samples taken  ac  four
position 6  (pile  7).
hour intervals.   Sampling

-------
       With  the  recognition that pile 6B  failed, an effort was made
 to  diagnose  the  responsible operational flaw(s).  The suspected
 immediate  cause  of  failure was inadequate penetration of air into the
 mass,  fading  to inadequate heat removal.  The extent of air pene-
 tration was  tested, on day 20, with the use of externally heated
 air  (130 C measured in the flexhose duct  between blower and pile).

       Thirty  minrt.es of continuous input of heated air resulted  in
 a 33,8°C temperature elevation in tV^e woodchip base at the monitor-
 ing position nearest the flexhose, and a  10.3 C elevation at the
 furthest base  position (Figure 73).  The maximum elevation in the
 sludge-woodchip  mixture itself (1.1°C) was immediately above the  flex-
 hose.  This  indicates that most of the air passed horizontally through
 the woodchip base,  rather than passing up through the sludge-woodchip
 mixture.

 DISCUSSION

       Two of  the factors influencing air penetration into the compost-
 ing mass are the  porosity of the sludge-bulking agent mixture, and the
 design of the  woodchip base.  Porosity is affected by the sludge-bulk-
 ing agent ratio  and by the moisture content, among other factors.  Both
 insufficient porosity and poor base design probably contributed to the
 failure of pile  B.

       In routine practice, the more favorable porosity represented by
 pile 7's sludge:  bulking agent ratio and initial moisture content
 should be an operational goal.  However, variations in po.-osity are
 unavoidable.   In  contrast, woodchip base construction is repeatable
 without variation, and its design should be optimized.  The base design
 represented by pile 7 is superior in that it suppresses short-circuit-
 ing, directing the air to the overlying mass.

       One aspect of the comparative performance of these piles ic
 explained by ambient temperature.  This is the length of time needed
 to reach the set-point temperature, which was  2.5x lengthier for
 pile 6B than pile 7.  The lengthier "come-up"  is attributable to the
 lower, biologically unfavorable,  time-zero ambient temperature.  The
 explanation of the other performance aspects resides in the factors
 governing the  composting ecosystem, as influenced by feedback ventila-
 tion control.

       With adequate air penetration,  represented by pile 7,  demand
 for heat removal is met via blower operation.   This induces an inter-
action between pile and blower manifested in time-variable blower opera-
 tion, leading to the regulation of pile temperature.  This results in
 rapid decomposition, rapid drying, and an oxygenated condition.  The
 first two factors are intimately linked, in that decomposition gener-
 ates heat - which vaporizes the water.  The oxygenated condition re-
 flects the balance between a high rate of oxygen consumption  and a
commensurately high rate  of oxygen resupply.
                                128

-------
Figure 73.   Reprinted by permission from TOXIC AND HAZARDOUS WASTE:   Proceedings of the
   15th Mid-Atlantic Conference,  pp.  463-471.   Copyright c 1983.   Bucknell University.
   Pile 6B; temperature changes after 30 minutes continuous input of externally heated
   air.

-------
       Non-adequate air peretration, represented by pile 6B, unlinks
blower demand and heat removal, in that the air  (or much of it) is
short-circuited.  Hence, pile and blower do not  interact.  The con-
sequent failure to arrest the temperature ascent leads to inhibitive
levels; consequently, decomposition and drying are slight.  An oxy-
genated condition prevails, because the slight rate of oxygen re-
supply suffices to roughly match the slight rate of consumption.

       Finally, these observations are put into  a broader perspective
by comparing the nominal behavior of a composting pile managed accord-
ing to the Rutgers strategy (.e.g.  pile 7), the nominal behavior of a
Beltsville-type pile (e.g. pile 933), and a failed pile intended as a
Rutgers pile (6B) (TABLE IS).   Nominal Rutgers behavior requires a
ventilation system adequate in three respects.   First, the blower must
respond, in a time-variable fashion, to the needToTneat removal in
reference to temperature (temperature-feedback control).  Second, the
blower must be adequately sized, to meet the peak demand fo*r~veritila-
t^on'  Third, the air must pass through the mass reasonably freely.
The system represented by pile 7 was adequate with respect to all three
factors; the pile 6B system was deficient with respect to the third
factor; the pile 9B system was deficient with respect to the first and
second factors.

       Thus, piles 6B and 9B had different deficiencies in terms of
their ventilation systems.  The outcome was similar, however, in that
excessive heat accumulation was manifested in inhibitively high tem-
perature coupled vrith a high level of 0^.   This demonstrates that a
processing failure by Rutgers  standards resembles nominal performance
by Beltsville standards.
                                130

-------
    TABLE  15.   NOMINAL  AND  FAILED  BEHAVIOR OF COMPOSTING SYSTEMS
Nominal behavior
Temperature
Drying tendency
Demand for ventilation
Level of peak demand
Duration of peak demand
Interstitial atmosphere
Temperature gradient
Rutgers
Regulated, as
intended
Strong
Builds gradually
to peak
Usually <100%
Hours or days,
then dwindles'
Oxygenated1*
Decisively
established
Beltsville
Regulation not
attempted, iinhibi-
tively high
Weak
N/A
N/A
N/A
t
Oxygenated
Weakly
established
Failed
behavior
Rutgers
Regulation
failed, inhibi-
tively high
Nil
Euilt instan-
taneously to
peak
100%
Sustained
indefinitely
f
Oxygenated
Weakly
established
 Result of rapid 02 uptake balanced by rapid resupply (see Section



^Result of slow 02 uptake balanced by slow resupply (see Section 3)
3).

-------
                            SECTION  11

                       PATHOGEN  INACTIVATION
 INTRODUCTION
      Composting  can  serve  as  a waste  treatment  technology  owing  to
 its capacity  to  stabilize  and sanitize  the  material.  The  components
 of stabilization are the decomposition  of putrescible matter,  the
 reduction  of  volume  and weight,  and the removal  of  the water.  All
 of these are  advanced  through maximization  o£ the rate of  decomposi-
 tion.  The material  is sanitized  through biological antagonisms  that
 inactivate or destroy  pathogenic  organisms,  and  through  temperature-
 inactivation.  Biological  antagonisms,  though poorly understood, are
 presumably also  promoted through  maximization of the decomposition
 rate, as this is  synonymous with  the  general level  of biological
 activity.  To this extent, therefore, the goals  of  stabilization and
 sanitation are advanced in tandem. With respect  to  temperature-in-
 activation, however,  a potential  for  conflict exists.  Inactivation
 through this  mechanism is positively related to temperature throughout
 the range of  possible  composting  temperatures (peak ~ 80 C), whereas
 the threshold to  a significant decrease in  decomposition rate  is
 approximately 55  -60°C.

     This potential  conflict  would be aggravated by an administrative
 regulation calling for "the fastest possible pathogen inactivation"
 (15).  A regulation  so worded would mandate  the  earliest possible
 attainment of the highest possible temperature.  However,  the  Federal
 interim final criteria for pathogen reduction (50)  are worded  in a
 mann<;r that permits  a degree of  operational  flexibility  in  meeting  the
 indicated goals.  The criteria are for  "significant pathogen reduc-
 tion" (at least  40°C for 120  consecutive hrs and, within this  period,
 at least 55 C for 4  hrs), and "further  pathogen  reduction"  (at least
 55°C for 72 consecutive hrs).

     These criteria  can be met with little or no penalty in terms
 of stabilization and water removal through tactics  involving either
 the high rate stage or the curing stage.  The former is  demonstrated
 experimentally in the present section;  the latter is noted in discussing
 the problem of monitoring regulatory compliance.

MATERIALS AWD METHODS SPECIFIC TO THIS  SECTION

     The trial involved a 6 tonne pile  of the sludge-woodchip  mix-
 ture (pile 13).   In  addition  to the usual thermocouple grid in the
mixture, four thermocouples were positioned in the woodchip bed
 (Figure 74).   (Note that the  mixture was accidentally placed in an
off-center orientation relative to the  flexhose duct.)   The controller


                               132

-------
Figure 74.  Pile 13,  cross-sectional representation:   textured area,  woodchip cover and
    base;  clear area,  sludae-woodchip mixture;  circle,  perforated flexhose.   The
    numbers indicate the monitoring and control positions:  thermocouples,  positions 1
    through 19; gas  sampling probes,  adjacent to positions  ^ and 16;  control
    thermistor, position 5.   The blower was operated in the forced-pressure  mode.
    (The sludge-woodchip mixture was accidentally offset realtive to the woodchip  base
    and this pile,  lacking the usual slight longitudinal axis,  was nearly symetric.)

-------
 set  point  was  4S°C.   The  experimental  plan  was  to  shut  down the  blower
 before  termination  of the  period  of  temperature  feedback  control,  and
 this  was done  at  hr 176.

      The trial  period was  27  February  1981  to  18 March  1981.   Ambient
 air  temperatures  during this  period  were  (°C):   mean  of the daily
 highs,  9;  mean  o£ the daily lows,  zero; mean  of  the daily means, 4.5;
 range of the daily  highs  and  lows, -7.8 to  13.9.   Precipitation
 amounted to 7.0 cm  in 4 occurrences.

 RESULTS
     Temperature  feedback  control  commenced  at hr  48  (Figure  75).
 Blower operation  time  increased  sharply,  until reaching  1001  at  hr  66.
 The  I time on  started  decreasing at  hr  138,  reaching  approximately
 61%  at hr 176,  At  this  time  the blower was  deliberately shut  dovm.
     The temperature record at the  innermost  series of probes  is
presented in  Figure 76  (all of the  temperature data are  in Appendix
G) .  Early in the period of temperature  feedback control  (hr SO to
6S) the temperature ascent at position S  (site of  the control  thermistor)
paused at approximately the set point value  (4S°C).  The  ascent resumed,
however, despite 1001 blower on time, reaching a peak of  58QC  at  hr  100.
The temperature subsequently declined and, starting at hr 140, stabi-
lized at the  set point level.  As is characteristic of the Rutgers
strategy, a positive temperature gradient in  the direction of  airflow
became established,  Deliberste blower shutdown at hr 176 marks the
start of a new temperature ascent.

     Other temperature records representative of this trial are shown
in Figure 77.  Probe 1, positioned  in the woodchip base adjacent  to
the flexhose duct, experienced only a slight  temperature  ascent (10°C
to 27°C) prior to blower shutdown,  at which point  there was a  sharp
upturn in the temperature  (27 C to  7i , 3C).  Within  the sludge-woodchip
mixture the sharpest post-shutdown  temperature increase was at posi-
tion 6 (34°C to 78°C).  Two positions near the edge of the pile,  14
and 9, did not increase in temperature subsequent  to blower shutdown,
but rather decreased.

     The temperature data  from all  of the positions are summarized in
TABLE 16 to show the temperature increase resulting from blower shut-
down.   The mean increase (positions 9 and 14 omitted) v/as 27.7°C  in
20.4 hours (!„ 36°C/hr) .

Gas __An_aly^sejS_._

     Prior to shutdown C02 was below the limit of detection and 0?
was at the ambient concentration (Figures 78  and 79).  Abrupt  changes
(CO-, increase and 0,,, decrease) coincided with the cessation of
mechanical ventilation.
                                134

-------
          100-tl
         2
         O

         £
         CJ
         a:
         2
         O
         (-40

         o:

         Q

         ET
         LU
                         iOO
                                     200           300
                                        TIME IN HOURS
                                                                400
                                                                             500
Figure 75.  Pile  .13,  blower operation.  The baseline  represents operation as scheduled
    by timer,  and  the area above the baseline represents  blower operation through the
    temperature-feedback control system.  The blower, which  was operated in the
    forced-pressure  mode,  was shutdown at hr 176.

-------
     ao
        r
    oSO
PROBE 13-le
PROBE 13-15
PROBE 13-10
PROSE \$-
                                     TIME IN HOURS
                                                             400
                                               500
Figure  76. Pile 13, temperature at the  innermost vertical  series of thermocoup
                                                     >les.

-------
Figure 77.   Pile  13,
temperature at  re-
presentative posi-
tions.   Top graph,
position 14;  middle
graph, position  6;
bottom graph,
position 1.
oo



60



40



20



 0

80
                         60
                         80 ~
                         60 -
                         40 -
                         20 -
                                  100
                                         200     300
                                         TIME IN HOURS
                               400
                                       500
                              137

-------
                TABLE 16.  PILE 13:
TB4PERATURE PEAKS RESULTING FROM SIWIDCMN OP THE B3JWER
             AT HR 176

Therrao-
couplea
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Pre-shutdown
Temp
iM
27
36
57
53
57
65
65
63
67
64
72
73
72
68
72
73
78
73
74
jpeak
Time
Ihrl
175
110
128
100
100
90
100
100
90
92
85
85
88
88
85
85
85
90
85
Temp (°C)
just prior
to shutdown
27
18
27
31
45
34
37
42
49
54
44
40
47
47
61
50
48
62
49
Post-shutdown
Temp
CQ
71
67
66
58
75
72
66
60
-t
76
75
73
65
-t
74
76
67
76
71
peak
Time
(hr)
240
204
200
186
206
200
190
185
-
208
194
192
189
_
200
194
194
200
197

Differential (°C)**
44
49
39
27
30
38
29
18
-
22
31
33
18
-
13
26
19
14
22
 *   Thermocouples 1 through  4 were  positioned in the woodchip  base;  the other thermocouples  were
    in the sludge-woodchip mixture  (see  Figure .75).

"*   Post-shutdown peak temperature  minus temperature just prior  to shutdown.

 t   The temperature decreased after blower shutdown.

-------
                100
                            200          300
                              TIME IN HOURS
                               400
                                           500
•igure  78.  Pile  13,
    (lower curve) .
    position 6.
concentrations of  02  (upper curve)  and CO.
The gas sampling probe  was adjacent to
                              139

-------
                 100
       200         300
          TIME IN HOURS
                                                    400
                                                                500
Figure 79,  Pile 13,
     (lower  curve).
     position  16,
concentrations of  02  (upper  curve)  and C0?
The gas sampling probe  was  adjacent to
                             140

-------
     During  the  first  176 hrs  the moisture content decreased  from
 604 to 43%  (Figure  80).  In the post-shutdown period  a  further
 decrease  to  27$  occurred.

 DISCUSSION

     During  the  period of comparability  (up to hr 176)  the  temperature
 control achieved in this pile  was less precise than in  the  previous
 Rutgers piles.   This is attributed  to the accidental  off-center
 position  of  the  sludge-woodchip mixture  relative to the  flexhose
 (Figure 74), presumably leading to  uneven air distribution.   Thus
 two factors  were non-ideal with respect  to the stabilization-water
 removal objectives: the non-deliberate offset position  of the compost-
 ing pile  relative to the flexhose,  and the deliberate early shutdown
 of the blower  (hr 176).  Despite these factors process  performance
 was satisfactory, judging from the  moisture content decrease  and
 informal  observations of the odor and visual appearance  of  the material.
 Questions of pathogen  inactivation  aside, the satisfactory  performance
 of pile 13 is  indicative of the reliability ol' the Rutgers  strategy,
 given indifferent routine field practice.

     As intended, deliberate blower shutdown induced  a  sharp  tempera-
 ture upturn.   The exceptions (sites 9 and 14) were near  the pile's
 outer edge.  The  ambient temperature during this trial  (mean, 4.5°C)
 was the coldest  of the entire  investigation.

     These temperature data were analyzed with respect  to the federal
 interim final  criteria for pathogen reduction in static-pile  compost-
 ing (48).   Three  time periods  are considered separately  (TABLE 17).
 During the first period the temperatures at positions 1  through 4  (the
 woodchip  base) and position 6  djd not meet either of  the criteria.
 The other positions met either the  criterion for significant pathogen
 reduction (spr)  or the criterion for further pathogen reduction (fpr).
 During the post-shutdown period positions 7, 8, 9, 13,  14s  and 17
 did not meet either of the criteria.  Considering the entire trial
 period, only position 4 (in the woodchip bed) did not meet  the cri-
 terion for further pathogen reduction, and this position met the
 criterion for  significant reduction.  Thus, the induction of harsh
 temperatures through deliberate blower shutdown improved performance
with respect to  the federal interim criteria for pathogen inactivation..

     The other Rutgers piles (7, 8,  9A,  11A, 11B, 11C) were not
deliberately subjected to harsh terminal temperatures and did not
perform as well with respect to the federal criteria  (TABLE 18).
Overall,   76% of  the monitoring positions in these piles met the
significant reduction criterion;, and 411 met the further reduction
criterion.  The  Beitsville pile (9B) met the criterion for  further
pathogen reduction at all of the monitoring positions.  Pile 6A met
the criterion  for further pathogen reduction at 12 of the 13 sites
 (see footnote  - TABLE 17).

-------
       805
                      100
200            300
   TlfcSE IN HOURS
                                                              400
                                                                           SCO
Figure 80.  Pile 13, moisture content.   Samples taker  from central interior locations

-------
TABLE 17.   PILE 13:   TEMPERATURE DATA
TERMS OF THE FEDERAL INTERIM FINAL CRITERIA
Pre- shutdown
(hr zero to hr 176)
Thermo-
couple
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
a
Spr =
Fpr =
Hrs Hrs
£40°C ->55 C
None
None
75
70
130
92
124
132
126
130
130
128
128
136
264
132
128
118
' 125
significant
consecutive
further p&tl
None
None
14
None
28
55
72
66
81
74
82
88
92
86
190
90
97
98
98
pathogen
period ^
Criter-n
ion met
None
None
None
None
Spr
None
Fpr
Spr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
reduction
5SdC).
tiogen reduction (72
Hrg
155
128
122
112
230
130
114
31
12
280
236
122
86
6
132
236
118
270
250
(120
Post-shutdown
' (hr 176 to hr 445)
Hrg
120
95
100
5
102
98
40
6
None
121
105
42
20
None
90
104
38
130
90
consecutive hrs
Entire trial period
(hr zero to hr 445)
Criter-* Hrs Sirs
ion met ±40°C >-S5°C
Fpr
Fpr
Fpr
Spr
Fpr
Fpr
None
None
None
Fpr
Fpr
' Spr
None
None
Fpr
Fpr
None
Fpr
Fpr
i 40°C,
155
128
197
182
360
222
238
163
138
410
366
250
214
142
396
368
246
388
215
plus 4
12G
95
114
5
130
153
112
72
81
175
187
130
112
88
280
194
135
228
188
Cnter-e
ion met
Fpr
Fpr
Fpr
Spr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
Fpr
hrs within the
consecutive hrs L 55°C).

-------
TABLE IS.  ALL PILES:  SUMMARY OF THE TEMPERATURE RECORDS WITH RESPECT TO THE FEDERAL
           INTERIM FINAL CRITERIA
Pile
(report
Section)



7(3)
8(3)
9A (3)
4A (4)
11A (5)
11B (5)
11C (5)
13 (11)
9B (3)
6A (10)

Control
strategy



Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Rutgers
Beltsville
N/A*
Mean
ambient
temp ( C)



15
21
27
14.5
9
9
9
4.5
27
11
Terminal
temp
elevation



No
No
No
No
No
No
No
Yes*
N/A
N/A


Temp
Total
(no. )

15
31
19
12
12
12
12
19
IS
13


monitoring


sites
Perforraance
with respect to
stabilization
and water removal
Meeting indicated
criterion
Spr
12
20
17
11
11
8
7
19
15
*
12V
(no . )
Fpr
6
6
13
2
6
6
7
18
15
12*


Good
Good
Good
Good
Good
Good
Good
Good
Poor
Failed
   Terminal harsh temperatures induced  (present Section)

   The intention was to Jnanage this pile according to the Rutgers process
   ing failed because of inadequate air penetration (see Section 10).
but the process-
   One position  (position 5 - see Figure 74) did not literally meet either criterion, but
   it experienced the following temperatures: peak, 71°C; 49 consecutive hrs at ^  55°G;
   116 hrs  (in three intervals) at > 40°C.

-------
     In contrast, with respect to the stabilization-water removal
objective the Rutgers piles performed well, the Beltsville pile
performed poorly, and pile 6A failed.

     These observations should be put into perspective.  Except for
the present Section, this investigation concerns the stabilization-
water removal objective.  As such, the thermocouples were not posi-
tioned specifically for the purpose of determining compliance with
the Federal criteria.  For example, pile 9B (Beltsville process)
lacked a series of thermocouples next to the interface with the
woodchip cover  (compare Figures 17 and 18).  The material in this
region is suspect because of its proximity to the pile's edge, and
because the vacuum- induced direction of ventilation would further
promote coolness.  Another deficiency, from the viewpoint of monitor-
ing for compliance with the Federal criteria, is the lack of probes
in the pile "toes."  This refers to the part of the pile forming
a triangle with the woodchip base or the concrete pad.  Because of
the direction of ventilation, the "toes" of Beltsville piles are
particularly subject to coolness.  These considerations illustrate
the obvious point that the reliability of "compliance data" is
strongly dependent on the frequency of monitoring and the selection
of positions for the monitoring.

     The problem of monitoring may be less intractable in routine
practice than might appear from a consideration of these freestand-
ing piles.   This is because the extended pile geometry would pre-
sumably be used for the Rutgers strategy in static pile configuration,
as it is in routine Beltsville- type operation.  The- ratio of exposed
edges to interior volume is less with such geometry.

     Despite the accumulation of considerable routine opera.tional
experience with the Beltsville process , published systematic studies
of the capability of this process to ..cr.ply with the Federal interim
final criteria seem to be lacking.   Furthermore, other than pile 9B,
we know of no published report on the Beltsville process which provides
sufficiently detailed data to support an interpretation of the time-
temperature observations in reference to the criteria.

     An independent side-by-side comparison of the Rutgers and Belts-
ville approaches, involving a mixture of refuse and sewage sludge,
included tests for Salmonella as one of the many points of compari-
son (34).   All of tfieTIsts relating to stabilization, including
changes in moisture content, indicated that the Rutgers strategy
gave superior performance.   With respect to the decrease in
both processes performed comparably.
     TABLE 19 summarizes our thoughts regarding the relative merits
of the Rutgers and Beltsville approaches with respect to the pathogen
inactivation objective.  Some of the points are necessarily specula-
tive.  The present state of knowledge does not indicate that one, or
the other, of these approaches is superior with respect to sanitation.
Both are highly effective in this respect.
                               145

-------
TABLE 19.  RELATIVE MERITS OF THE RUTGERS HIGH RATE STAGE AND THE BELTSVILLE
           "ACTIVE STAGE" WITH RESPECT TO SANITATION.
     P_o_int_ of  comparison
          Comments
Harsh temperatures
(e.g. 7SDC-8QOC)
Biological antagonisms
Cool spots
Dryness as a sanitizing agent
Observed Salmonella  reduction
Potential for regrowth  of
Salmonella
Intrinsic to  the  Beltsville pro-
cess; inducible as  a  terminal step
of the Rutgers strategy.

Presumably more effective in the
Rutgers strategy.

"Jogs*":  Beltsville more  subject
to pro¥lem.

Outer_e_dge_sj_  Beltsville more
sTTEJFcTto~~p*r o b 1 e m.

Interior_(near_f 1exhose) :   Rut-
gers mor£j~s~u B j e~c tto~p~FoFlem (but
not if terminal harsh temperatures
induced).

Operative only in the Rutgers
strategy.

Rutgers and Beltsville  approaches
performed comparably  (34).

Less potential in Rutgers
strategy, as  substrate  is  more
thoroughly decomposed.
                                  146

-------
     Finally, the potential role of the curing stage in assuring
product safety should be considered.   The advantages o£ emphasizing
the curing stage to insure sanitation, relative to the high rate
stage, are as follows.  1) The curing stage is closer, with respect
to time, to the point of end-product  usage.  This decreases the
opportunity for post-processing recontamination or regrowth.   2)
Large well-insulated curing piles presumably will gradually self-
heat to harsh temperatures, even though formed of material thoroughly
stabilized and dried in.the high-rate stage.  3)  Large piles  have a
relatively low surface to volume ratio.  Thi^ minimizes the volume
of "edge material" exposed to ambient conditions.  4)  It seems prob-
able from item 3) that, in routine practice, monitoring of the curing
stage would provide the more reliable time-temperature data to insure
pathogen inactivation.

     Composting is an excellent means of sanitizing waste.  Regard-
less, we believe that the high-rate stage should  be managed primarily
to accomplish stabilization and water removal objectives.   If the
intended use of the end-product requires additional assurance of
pathogen reduction this can be achieved through a terminal harsh
temperature phase of processing, or through a curing strategy designed
with product safety assurance  as one of its objectives.
                                147

-------
                             SECTION  12
                  UNIFORM  PROVISION OF AIR ALONG  THE  LENGTH
                          OF A  COMPOSTING PILE"
INTRODUCTION
      If the composting pile  is  relatively  long,  and  the  area  of  air
holes is uniform along the  length  of  the  duct,  delivery of  air to
the pile is non-uniform.  This is  because more  air  exits  from  the  part
of the Juct close to the blower  than  from the distant part.  With  the
control thermistor placed midway between  the ends of  the  pile,  the
part of the pile close to the blower  is overventilated  (too  cool)  and
the more distant part underventilated  (too  hot).

      A means of providing  air uniformly  in the longitudinal dimension
is to vary the size and/or  spacing of  the air holes along the  length
of the duct, such that the  amount  of  air vented is  similar  regardless
of distance from the blower.  A  computer program, presented  herein,
was developed to help the user determine the number of evenly  spaced
holes, of a given hole size, necessary to supply  a  given  amount of
air to a 5 footTsection of  duct.   Each succeeding S foot  section is
treated separately.  A spacing scheme  for one to  20 holes per  section
is stored in the program.   Depending  on the length  of the pile, it
may.be advantageous to increase  the size of the holes in  progressive
sections of the duct.

CALCULATION4

      The airflow out of each hole in  the duct  is calculated through
the use of two equations  (52).   The  first involves  the pressure drop
in the duct, as follows:
                 D5
where:    f - the Moody friction factor  (dimensionless)

          L = the length of duct (ft)

          p = fluid density (slugs/ft3)

          q - flow (ft/sec)

          D ~ duct diameter (ft).

Comments on these factors are as follows :
*An early, fTal7e^T~v^i'sTon~~6^f"'tKrs"^5ection was  published  (51).   A
 correction was  submitted  to  the publisher.
tTo change feet  to meters, multiply  by  0.305.
tV/e thank the following individuals  for  kindly  reviewing  the calcula
 tions: James H. Miller, Roberto C.  Leon, Dr. Robert C. Ahlert.
                                 148

-------
 Ntoody  friction  factor  (f).   Tables  of values of this factor are found
 in"TiuT3~3ynariu"cs TeTcHooKS .   Also,  the Fanning friction
 factor may be used  (Fanning  factor  = Moody factor/4).   For the pre-
 sent application  (commercial  pipe  serving as duct,  high velocity) the
 Moody  friction  factor  is  roughly  0.02.

                      This  refers  to  the total length of duct,  includ-
 _____
 TngpaFalTel~Ten~gtEs where  used.

                     The value used  in  the  program is  p a 0.00238
                This  is the mean  cross-sectional  velocity.
 £iajnete£__(Dj_._  The user selects  the  duct  diameter.

       The second equation  is used  to  calculate  the  airflow out of
 each hole, as follows:
  /
,y
        0 => 0.00?ld2     duct "exit                                (ii)
        e               ___  _
 where: q  = the exit flow  (ft/3sec)

    P   6                                                 2f
     duct ° the pressure in the duct at  that point  (Ib/ft  )

        d = hole diameter  (in.)
    P
     exit s the pressure in the pile

        K c constant

 Comments on these factors are as follows:

 Pressure in the pile.   Since instantaneous pressure should  not  vary
 aTo~rf g~THe-TengtTioT~t h e pile, it is set equal to zero.

 Constant.   In this program K is set equal to 1.5.

 COMPUTER PROGRAM - BACKGROUND

 List of thingj_tjia/t the user^jnust Jcnow to enter the program

                              Clb/ft ).  The pressure at the entrance
 _.^.--.--
• t"6 ^He~ fTFsT~JeTtTon~T2^°aucT~mu s t be kn n wn .   The pressure at this
 point  is  the  pressure at the blower (fc ^. , as specified by the
 manufacturer)  minus the pressure drop over the distance to the first
 section.   The  user should also subtract the  pressure loss from the
 ductwork  to the  outer edge of the pile (the  pile backpressure).  If
 this is not known, a conservative estimate is 1.0 Ib/ft2.  The
 To change  slugs/ft3  to  kg/m3,  multiply by 3.99 x 10"4.
tTo change  Ib/ft2  to  kgm/ra2,  multiply by 4.873.
                                 149

-------
pressure at the exit of the first section  is  the  entrance pressure  at
the second section, etc.   The pressure at  the blower must provide
adequate pressure to the terminal section  of  duct.  This is  satisfied
when the number of holes required in the terminal section is  <20,
and the required hole diameter is reasonably  less than  the duct  (pipe)
I.D..  The use of a blower delivering higher  pressure than necessary
yields a solution involving fewer holes and/or smaller  holes,  but this
represents an uneconomical approach.  A_suggested first approximation
for pressure at the blower is 6.0 Ib/ft .

                 to duct (cfs) * .  The estimation  of the required air
                   ~
                         the" piTe is based on pile 9A  (Section 3)
as this is the largest pile  in our experience (40 ton) ,"*'  and  as  it
yields the largest  (hence the most conservative) estimate.  Further-
more, the estimate  is conservative for use with other  sludges, as the
experimental material was primary sludge with a high volatile solids
content (^ 75%).  This pile  was ventilated by six 1/3  hp  blowers at
each end of the pile.  The blowers were operated in unison.   Each
blower discharged into a 20  foot length of perforated  duct, which was
capped at the distal end.  Since the total length of perforated  duct
was 120 feet, for calculation purposes there are twenty-four  S foot
lengths .
                                                       $
      The peak demand exerted was 71.3 cfm per wet ton  (initial
weight) ,  which is rounded off to 80 cfm per ton for purposes  of  the
estimation.  Since  the pile  weighed approximately 40 tons, the total
peak demand was 3200 cfm^.   The total length of pipe was  120  feet,
arranged in three parallel branches of equal length.   Thus there were
twenty-four S foot  sections, with each section requiring  133.3 cfm.
For use in the program this  is converted to cfs (2.22  cfs) .

                          The use of standard pipe ID's is recommended.
                  This depends on duct size, material, and airflow,
                  .

             , L__in . J2nds of an inch.  Different values should be entered
                 ™   the~*piro'gf am to find the best combination of number
and size of holes for each S foot section of duct.  An input of 99
will end the program.
 To change from cfs to m-^/sec, multiply by 2.83 x 10"2.
-j.
 To change ton to tonne, multiply, by 1.1023.
t                           ^
 To change from cfm/ton to m /sec-tonne (metric), multiply by 4.232 x
 10-4.

 To change from cfm to m /sec, multiply by 4.720 x 10" .
                                ISO

-------
 Output
      XXXX/32  inch*       XX holes         M£I§,    XXX.XXPSF
      The  hole  size     The  number of  Exit flow    Pressure at the
      (inputted)        holes  selected  from the    end of the
                       by computer      holes       section
 Two  output  values  are  provided each '.time the  program is run.  The
 first  is  the value  for the number of holes giving slightly less than
 the  desired exit  flow.   The  second is the value for the number of
 holes  giving slightly  greater  than the desired exit flow.
 Npjte_s
       a.  The  first  step in  putting the program on line is to es-
 tablish the file  on  hole spacing.   This is provided in TABLE 20
 (the "H.DAT" file).
       b.  All  of  the values  inputted to run the program must be real
 numbers,  rather than integers.   Alv/ays include a decimal point.
 List o f
AL  (20,21)  Array storing distance  between  each  hole  for  1-20  holes.
      Does not change  through  the program.
PREF  Duct pressure  in PSF.  Changes  in  line  290 }  330  and 240.
PREM  Duct pressure  in PSF.  Does not  change.  Used for program  to
      remember initial pressure.
CFS   Flow in pipe  (CFS) .  Changes  in  lines 270  and 310.
CFSM  Flow in pipe  (CFS).  Does not change  (used like  PREMJ
FF    Friction factor.   Does not change.
DUCT  Duct diameter  in feet.   Does  not change.
EXF   Exit flow from holes In  CFS.  Does not  change.
I (First use)  Counter for filling  "AL"  array  (Line ISO).
J     Counter for filling "AL" array  (Line  150)  and line  320.
DIA   Exit hole diameter (32nd1 s of an inch).  Changes in line 180.
N     Counter used in pulling  data  from  "AL",,  (Depicts number of
      holes.)  Changes in lines 200 and  220.
FL    Exit flow from holes (CFS).  Changes  in  lines 250 and  300.
FS    Exit flow from holes (CFS) 0  Acts  to  remember most  recent  value
      of "FL".  Changes  in line 230.
 to change from inches to cm, multiply by 2.54.
                                151

-------
                    TABLE 20.  "H. DAT' »iOLE DATA)  FILE
0.
0.
0.
0.
0.
0.
0.
0.
0.
5. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0o 0. 0
2.5 2.5 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
1.667 1.667 1.667 0. 0. 0. 0. 0. 0. 0. 0. 0
1.25 1.25 1.25 1.25 0. 0. 0. 0. 0. 0. 0. 0.
1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0, 0. 0. 0
0. 1.667 0. 1.667 0. 1.667 0. 0. 0. 0. 0. 0
.7083 .7083 .7083 .7083 .7083 .7083 .7502 0
0. 1.25 0. 1.25 0. 1.25 0. 1.25 0. 0. 0. 0.
.5521 .5521 .5521 .5521 .5521 .5521 .5521 .
. 0.
0.
. 0.
0.
. o.
. 0.
. 0.
0.
5521
0
0
0
0
0
0
0
0
0 0.
0.
. 0.
0.
. 0.
. 0.
. 0.
0.
5832
0
0
0
0
0
0
0
0
0
. 0
0.
. 0
0.
. 0
. 0
. 0
0.
. 0
0.
. 0
0.
. 0
. 0
0.
. 0
. 0
0.
. 0
. 0
0.
. 0
. 0.
0.
. 0.
. 0. 0. 0. 0. 0.
0.
. 0. 0. 0. 0'. 0. 0.
  0.
0. 0. 1. 0. 1. 0.  1.  0.  1.  0.  1. 0.  0. 0. 0. 0. 0. 0. 0. 0. 0.
0. .4583 .4583 .4583  .4583  .4583 .4583 .4583 .4583 .4583 .4583 .4170 0. 0. 0. 0. 0. 0.
  0. 0. 0.
0. 0. .8333 0. .8333  0.  .8333  0. .8333 0. .8333 0. .83?5 0. 0. 0. 0. 0. 0. 0. 0.
0. .3854 .3854 .3854  .3854  .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3854 .3752 0. 0.
  0.0.0.0.0.
0. 0. .7083 0. .7083  0.  .7083  0. .7083 0. .7083 0. .7083 0. .7502 0. 0. 0. 0. 0. 0.
0. .3333 .3333 .3333  .3333  .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333 .3333
  ,3338 0.  0. 0.  0. 0.
0. 0. .625  0. .625 0. .625  0.  .625 0. .625 0.  .625 0. .625 0. .625 0. 0. 0. 0.
0. .2917 .2917 .2917  .2917  .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917 .2917
  .2517 .2917 .3328 0.  0.  0.
0. 0. .5521 0. .5521  0.  .5521  0. .5521 .0. .5521 0. .5521 0. .5521 0. .5521 0. .5832 0.
  0.
0. .2604 .2604 .2604  .2604  .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604 .2604
  .2504 .2604 .2604 .2604  .3128 0.
0. 0. .5 0. .5 0.  .5  0.  .5  0.  .5 0.  .5 0. .5 0. .5 0. .5 0. .5
0. 0. 0. 0. 0. 0.  0.  0.  0.  0.  0. 0.  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.

-------
 PRES  Duct pressure in PSF.  Acts like "FS".  Changes- in line 240.
 I (Second use)  Counter to pull data from "AL".  Changes in line  280.
 NC    Simply N-l used to modify output of liie 370.  Changes in line 360,
 THE PROGRAM
 @ Type Pipos, for
 10    REAL   AL(20,21)
 20    WRITE(Sf440)
 30    READ(5,*)DUCT
 40    WRITE(5,4SO)
 SO    READCS/3 EXF
 60    WRITE(5,400)
 70    READ(5,*)PREF
 80    PREM^PREF
 90    WRITE(5,410)
100    READ(S,*)CFS
110    CFSM^CFS
120    WRITE(5,420)
130    READ(5,*)FF
140    OPEN (UNIT=29,DEVICE='dsk:'SFILE^'H.DAT')
150    READC29/)    ((AL(I ,J) ,J=1,21) ,1 = 1,20)
160    CLOSE(UNIT=29)
170    WRITEC5,430)
180    READ(S,*)DIA
190    IF(DIA,EQ.99.)GO    TO   390
200    N=l
210    GO   TO   250
220    N=N+1
230    FS=FL
240    PRES=PREF
250    FL^O
260    PREF-PREM
270    CFS=CFSM
280    DO   3101=1,N
290    PREF=PREF-(AL(N;I)MCFS**2)/(DUCT**Sr'FF/519)
                                153

-------
300
310
320
330
340
350
360
370
380
390
400
410
420
430

440
450

460

470
FL=FL+(0. 000126* (DIA**2) * (PREF**0.
CFS=CFSM-FL
J-N + 1
S))


PREF=PREF- (AL(N,J)A (CFS**Z) / (DUCT**S) *FF/519)
IF(FL.LT.EXF)GO TO 220
WRITE (S ,460) N,DIA,FL,PREF
NC=N-1
WRITE(5,460)NC,DIA,FS,PRES
GO TO 170
STOP
FORMAT (' ', 'INPUT ENTRANCE
FORMAT (' ', 'INPUT ENTRANCE
FORMAT (' ', 'INPUT FRICTION
FORMAT (' ', 'INPUT DIA IN
INCH OR 99 TO END')
FORMAT (' ' , 'INPUT DUCT
FORMAT (' ', 'INPUT EXIT
OF DUCT IN CFS')
FORMAT (' ',12,' HOLES' ,F4.0,'
CFS' ,F8.2' , PSF')
END






PRESSURE IN PSF1)
FLOW IN CFS')
FACTOR')
32NDS OF AN

DIAMETER IN FEET')
FLOW PER 5 FT

732 IN.',F7.2',


DISCUSSION

      In the developmental phase o£ the Rutgers strategy non-uniform
distribution of air was not a significant problem,  as this involved
relatively small (pilot scale)  piles.   Moreover,  the largest pile
(9A-40 tons) was served by blowers at  both ends of the pile, and the
longest continuous length of perforeated duct was only 20 feet.   Routine
operation, however, would involve longer piles, and a facility layout
having blowers at only one end  is preferred.   In  the absence of a
specific design remedy, this would result in  non-uniform ventilation and
degraded performance.   One such remedy was developed herein.
                                1S4

-------
                      CONCLUSIONS


The composting system tends to accumulate metabolically generated
heat excessively,  leading to  in.iibitively high temperature.  The
threshold to significant inhibition  is approximately 60°C,  and
its severity increases sharply at higher temperatures.  At  80°C
(common peak temperature) the rate of decomposition is extremely
low.

This tendency can  be controlled through ventilative heat removal
in reference to temperature.  The main mechanism o£ heat removal
is evaporative cooling; establishing  a drying tendency.  Implemen-
tation is via temperature feedback control of a blower(s),  using
standard  (non-proprietary) equipment.  The forced-pressure  mode
of ventilation is  moi3 efficient than the vacuum-induced mode.
In this manner an  operational ceiling of 60°C is maintained.

Blower capacity (head and volume) must suffice to meet peak de-
mand for ventilation, &s expressed through feedback control.  A
strong waste (e.g. raw sewage sludge) demands more ventilation
than a. weak one (e.g. digested sludge).

A temperature gradient is established along the axis of airflow,
whereas drying is  relatively uniform.  The temperature gradier.t
imposes a height limitation, above which a high rate of decom-
position is not realizable.

Managed thusly, decomposition and drying are related in that the
decomposition generates heat, the heat vaporizes water, and the
vaporization causes' drying.  Hence,  the stronger the drying
tendency the faster the sludge decomposition.

A consequence of temperature feedback control is that the compost-
ing mass is well-oxygenated, because more air is needed to  remove
heat than to resupply 0^.

This strategy permits the use of recycled compost as the bulking
agent, in static pile configuration.

This strategy,  compared to a conventional approach, resulted in
4x more sludge  decomposition in half the time

The cost of ventilation for water removal through composting and
non-biological  air drying was as follows:  composting,  $0.32/tonne
water removed ($0.31/ton);  air drying, $6.43/tonne water removed
($6.33/ton).  This difference results from the biological genera-
tion of heat at the expense of putrescible organic material in
the sludge.


                           155

-------
                       RECOMMENDATIONS


Maximization of decomposition rate should be the explicit goal of
composting process design and control.

Rate maximization should be approached through temperature feedback
control of a blower(s).

The rate of decomposition should be assessed in terms of the demand
for ventilation, and the course o£ drying.

This strategy (v/hieh focuses on temperature feedback control) should
be implemented at lowest possible construction cost, consistent with
operational considerations.

The unenclosed static pile configuration is structurally simple and
eminently suitable for implementation, and should be the preferi'ed
configuration.
                             156

-------
                             REFERENCES


1.  Finstein, M.S., and F.C. Miller.  Distinction Between Composting  (the
    Process) and Compost  (the Product).  Letter, BioCycle, 23 (6):56,  1982.

2.  MacGregor, S.T., F.C. Miller, K.M. Psarianos, and M.S. Finstein.  Com-
    posting Process Control Based on  Interaction Between Microbial Heat
    Output and Temperature.  Appl. Environ. Microbiol., 41:1321-1330,
    1981.

3.  Finstein, M.S., F.C. Miller, P.P. Strom, S.T. MacGregor, and K.M.
    Psarianos.  Composting Ecosystem Management for Waste Treatment. Bio/
    Technology, 1:347-353, 1983.

4.  Finstein, M.S., J. Cirello, S.T. MacGregor, F.C. Miller, and K.M.
    Psarianos.  Sludge Composting and Utilization: Rational Approach to
    Process Control.  Report to U.S. EPA, N.J. DEP, C.C. MUAS pp. 211,
    New Jersey Experiment Station, New Brunswick (U.S. Dept. Commerce,
    NTIS, Springfield, VA, No. PB 82 136243), 1980.

5.  Miller, F.C., LJ.T. MacGregor, M.S. Finstein, and J. Cirello.  Bio-
    logical Drying of Sewage Sludge - A New Composting Process.  In:
    Proceedings of the 1980 National Conference on Environmental Engineer-
    ing 3  Amer. Soc. Civil Engineers, New York, New York, p. 40-49, 1980.

6.  Finstein, M.S., J. Cirello, S.T. MacGregor, F.C. Miller, D.J. Suler,
    and P.F.  Strom.  Discussion of Paper by R.T. Haug (J.  Water Pollut.
    Control Fed.,  51:2189-2206, 1979)".  J.  Water Pollut. Control Fed.
    52:2037-2041, 1980.

7.  Miller, F.C., S.T. MacGregor, K.M. Psarianos, and M.S. Finstein.
    Direction of Ventilation in Composting Wastewater Sludge.  J. Water
    Pollut. Control Fed.,  54:111-113, 1982.

8.  Epstein,  E., G.B.  Willson, W.D.  Surge,  D.C. Muller, and N.K. Enkiri
    A Forced Aeration System for Composting Wastewater Sludge.   <7.  Water
    Pollut. Control Fed.   48:688-694, 1976.

9.  Willson,  G.B.,  J.F.  Parr,  E. Epstein, P.B.  Marsh, R.C. Chaney,  D.
    Calacicco,  W.D. Burge, L.J. Sikoras  C.E. Teste,  and S. Hornick.
    Manual for Composting Sewage Sludge  by the Beltsville  Aerated-Pile
    Method.  U.S.  EPA/U.S.D.A. Report, EPA-600/8-80-022, MERL/ORD,
    Cincinnati, pp. 65,  1980.

10.  Willson,  G.B.  and  J.C. Thompson.  Dewatering of Sludge Compost  Piles.
    In:  Proceedings of the National  Conference  on Municipal and Industrial
    Sludge Composting,  Hazardous Materials  Control  Research Institute,
    Silver Spring,  MD.f  p. 46-54, 1980.

                                157

-------
11. Sikora, J., G.B. Willson, D. Calacicco, and J.F. Parr.  Materials
    Balance in Aerated Static Pile Composting.  J. Water Pollut. Control
    Fed. 53:1702-1707, 1981.

12. Anonymous.  Composting Processes to Stabilize and Disinfect Municipal
    Sewage Sludge.  EPA Water Programs Operations (WH-547) 430/9-81-011
    MCD-79, Washington, D.C., pp. 43, 1981.

13. Willson-, G.D. and D. Dalmat.  Sewage Sludge Composting in the U.S.A.
    BioCyele, 24 (5) : 20-23, 1983.

14. Suler, D.J., and M.S. Finstein.  Effect of Temperature, Aeration, and
    Moisture on CO? Formation in Bench-Scale Continuously Thermophilic
    Composting of Solid Waste.  Appl. Environ. Microbiol.* 33:34S-350,
    1977.

IS. Finstein, M.S.   Composting Process Temperature:  Conflict Between
    the Fastest Possible Disinfection and Organic Matter Stabilization.
    Report distributed at the "Workshops on the Health Effects and Legal
    Implications of Sewage Sludge Composting," Cambridge, MA. pp. 12, 1978,
16. Finstein,- M.S.. Composting Hazardous Wastes.
    1366-1367, 1979.
Letter, Science^  204:
17. Finstein, M.S..  Composting Microbial Ecosystem:  Implications for Pro-
    cess Design and Control. Compost Science/Land Utilization, 21(4):
    25-39, 1980.

18. Finstein, M.S., J. Cirello, D.J. Suler, M.C. Morris, and'P.P. Strom.
    Microbial Ecosystems Responsible for Anaerobic Digestion and Com-
    posting.  J. Water Pollut.  Control Fed., 52:2675-2685, 1980.

19. Finstein, M.S.. Heat Output - Temperature Interaction in Composting -
    A Summary Report distributed at the "EPA Enclosed Municipal Sludge
    Composting Meeting," Cincinnati, OH, pp. 5, 1981.

20. Finstein, Melvin S., Frederick C. Miller, Steven T. MacGregor, and
    Kevin M. Psarianos.  Rational Design and Control of Composting
    Facilities.  In:  Proceedings of the Conference on Energy  Conserva-
    tion:  Retrofit of Municipal Wastewater Treatment Facilities,
    Argonne National Laboratory ANL/CNSV-TM-95, p. 19-25, 1982.

21. Finstein. M.S., F.C. Miller, S.T. MacGregor, K.M. Psarianos.  Com-
    posting Sludge: A Means of Linking Decomposition and Water Removal.
    In:  Proceedings of The Sisth Mid-America Conference on Environmental
    Engineering Design, University of Missouri, Columbia, MO,  p. 112-119,
    1982.

22. Psarianos, K.M., S.T. MacGregor, F.C.  Miller, M.S. Finstein.  Design
    of Composting Ventilation Systems for Uniform Air Distribution.
    BioCycle,  24(2):  27-31, 1983.

23. Miller, Frederick C., Steven T. MacGregor, Kevin M. Psarianos, and
    M.S. Finstein.   Static Pile Sludge Composting with Recycled Compost
    as the Bulking Agent.  In:  Industrial  Vasts, Proceedings of the
    Fourteenth Mid-Atlantic Conferences  Ann Arbor Science, p.  35-44.  1982.

                                158

-------
 24. Miller, Frederick C., Steven T. MacGregor, Kevin M. Psarianos, and
    M.S. Finstein.  A Composting Processing Failure:  Diagnosis and Remedy.
    In: Toxic and Hazardous Waste: Proceedings of the Fifteen th Mid-
    Atlantic Industrial Waste Conference, Ann Arbor Science, p, 463-471,
    1983.

 25. Finstein, M.S..  Economic Motives for Managing the Composting Micro-
    bial Ecosystem.  Bio/Technology 1:341-342, 1983.

 26. Miller, Frederick C., and Melvin S. Finstein.  Equipment for Control
    and Monitoring of High Rate Composting.  In: Proceedings of the In-
    ternational Symposium on Biological Reclamation and Land Utilization
    of Urban Wastes, Universities o£ Naples and Pisa, p. 551-560, 1983.

 27. Finstein, M.S., F.C. Miller, and P.P. Strom.  Evaluation of Compost-
    ing Process Performance.  In: Proceedings of the Conference on Com-
    posting of Solid Wastes and Slurries, The University of Leeds, 1983,
    In Press.

 28. Strom, P.P., F.C. Miller, and M.S. Finstein.  Problem of Scale in
    Composting Research.  In: Proceedings of the Conference on Composting
    of Solid Wastes and Slurries, The University of Leeds, 1983, In Press.

 29. Finstein, M.S. and F.C. Miller.  Principles of Composting Leading to
    Maximization of Decomposition Rate, Odor Control, and Cost Effective-
    ness.  In:  Proceedings of the Seminar on Composting of Agricultural
    and Other Wastes, Commission of the European Economic Community,
    Oxford University, 1984, In Press.

 30. Miller, F.C. and M.S. Finstein.  Materials Balance in the Composting
    of Sewage Sludge as Affected by Process Control Strategy.  Submitted
    for publication.

 31. Vestal, J.  Robie, and Vicky L.  McKinley.  Operational Parameters  of
    Composting:  Microbial Activity in Composting Municipal Sludge.   Pre-
    liminary Report to EPA, Contract No.  CR-897852-01-0, pp. 145, undated.

 32. MeKinley, Vicky L., and J.  Robie Vestal.  Biokinetic Analysis of
    Adaption and Succession: Microbial Activity in Composting Municipal
    Serfage Sludge.  Appl.  Environ.  Microbiol.    47:933-941, 1984.

 33. Hoitink,  H.A,J., and G.A. Kuter.  Factors Affecting Composting of
    Municipal Sludge in a Bioreactor.   Draft Report to EPA-MERL-ORD
    CR-807791-010) ,  pp. 100, undated.

34. DeBertoldi, M.,  G. Vallini,  A.  Pera,  and F.  Zucconi.  Comparison  of
    Three Windrow Compost Systems.   BioCycle 23(2):45-50,  1982.

35. Stentiford, E.I., D.D.  Mara, P.L.  Taylor, and T.G.  Leton.  Forced
    Aeration Co-Composting of Domestic Refuse and Sewage Sludge.   In:
    Proceedings of the Conference on Compacting  of Solid Wastes  and
    Slurries,  The  University of Leeds, 1983, In  Press.

36. Seton,  T.G., P.L. Taylor, D.D.  Mara,  and E.I.  Stentiford.  Tempera-
    ture  and Oxygen  Control of Refuse/Sludge Aerated Static Pile  Systems.
    In:  Proceedings  of the Conference  on  Composting  of Solid Wastes and
    Slurries,  The  University of Leeds, 1983, In  Press.
                               1S9

-------
 37. Toth, S. and N. Nocitra.  Sludge Composting and Utilization: Chemical
    Composition and Agricultural Value of Sewage Sludge Compost.  Re-
    port to U.S. EPA, N.J. DEP, C.C. MUA, pp. 115, New Jersey Experiment
    Station, New Brunswick, N.J. undated.

 38. Willson, G.B., J,F. Parr, and D.C. Casey.  Criteria for Effective
    Composting of Sewage Sludge in Aerated Piles and for Maximum Effici-
    ency of Site Utilization.  In: Pvoceedingo of the National Conference
    on Deoign of Municipal Sludge Composting Facilities, Information
    Transfer, Inc., Rockville, MD, p. 79-87, 1978.

 39. Olver, W.M. Jr., Static Pile Composting of Municipal Sewage Sludge;
    The Process as Conducted at Danger, Maine.  U.S. Environmental Pro-
    tection Agency, Washington, D.C. (Grant No. 803828), undated manu-
    script.

 40. Strom, P.P., M.L. Morris, and M.S. Finstein.  Leaf Composting Through
    Appropriate, Low-Level, Technology.  Compost Science,,  21(6):44-48,
    1980.

 41. Kasper, V.  Jr., and D.A. Derr.  Sludge Composting and Utilization:
    An Economic Analysis o£ the Cainden Sludge Composting Facility.  Re-
    port to U.S. EPA, N.J. DEP, C.C. MUA, pp. 342, New Jersey Experiment
    Station, New Brunswick, 1981.

 42, lacoboni, M.D., T.J. LeBrun, and J. Livingston.   Windrow and Static
    Pile Composting of Municipal Sewage Sludge.  Report to EPA-MERL-ORD
    Contract No. 14-12-150), pp. 124-, 1982,

 43. Alcock, R., G.H. Nieswand, M.E.  Singley, M.P.  Bolan, and B.L.  WMtson.
    Sludge Composting and Utilization: Systems Analysis of the Cainden
    Composting Operation.  Report to U.S. EPA, N.J.  DEP, C.C.  MUA, pp.
    128, New Jersey Experiment Station, New Brunswick,  1981.

 44. Strom,  P.P. The Thermophilic Bacterial Populations  of Refuse Com-
    posting as  Affected by Temperature.  Ph.D. Thesis,  Rutgers University,
    New Brunswick, N.J., 1978.

 45. Strom,  P.P. and M.S. Finstein.  Thermophilic Bacterial Populations
    of Solid Waste Composting.  Abst.  Annu,  Meet.  Amer.  Sos. MioTobiol.
    Q89, 1979.

46. American Society of Heating, Refrigeration, and  Air Conditioning
    Engineers,  Inc..  Psychometric Chart Nos. 1 and  3,  1963.

47. Anonymous.   Drying Corn at the County Elevator.   Univ  Illinois Co-
    operative Extension Service Circular 1053, March 1972.

48. American Public Health Association.   Standard  Methods  for  the  Ex-
    amination of Water and Wastewater,  13th  Edition, 1971.

49. Alexander,  M..   Introduction to  Soil Microbiology,  2nd Edition,  John
    Wiley § Sons,  N.Y.  1977.
                                160

-------
50.  Federal  Register.   Environmental Protection Agency.   Part IX, p.
    53438-53468,  September 1979.

51.  Psarianos,  K.M.,  S.T.  MacGr^gor, F.C.  Miller,  and M.S.  Finstein.
    Design of Composting Ventilation System for Uniform  Air Distribu-
    tion.   BioCyele,  24(2):27-31,  1983.   (Reprinted in Managing Sludge
    By  Compoeting,   p.  281-287,  The JG Press,  Inc., Einmaus, PA. (1984),
    under  the title  Ventilation  System for Uniform Air Distribution.)

52.  Crane  Technical  Paper #410:   Flow of Fluids.   Engineering Division,
    Crane  Co.,  Chicago, 1975.
                                 161

-------
        80
       60
     o
     o
     2: 40
     UJ
     H
       80
       60
     o; 40
       20
        APPENDIX A-l

Temperature Observations  for

   Piles 7, 8, 9A and 9B
                   100       200       300
                            TIME IN HOURS
                            400
          500
                   100
         200      300
         TIME IN HOURS
400
500
Figure 81.  Pile 7, temperature at position  1  (upper  plot)  and
    position 2  (lower plot).
                             162

-------
        QO






        60



      o_



      I 40
      UJ
      h-



        20






         0
        80







        60








      I 4°
      UJ
      H-




        20
           0
                    100       200       300
                             TIME IN HOURS
100
                            400
                             500
200       300

TIME IN HOURS
400
500
Figure 82.   Pile 1,  temperature  at  position 3 (upper plot)  and
    position 4  (lower plot).
                              163

-------
        00
        60
      u
        20
                    100
200      300
TIME IN HOURS
400
500
        80
        60
        40
        20
                   100
20C      300
TIME IN HOURS
400
500
Figure 83.  Pile 7, temperature  at  position 5 (upper plot) and
    position 6  (lower plot).
                              164

-------
         00
         60
      u
      9
      K.  40
      2
      UJ
      t-
         20
           0
100       200       300
         TIME IN HOURS
400
SOO
         80
         60
      o
      @
      UJ
                              "T
                   T
                    100      200       300
                             TIME IN  HOURS
                                     SOO
Figure 84.  Pile 7,  temperature at position 7  (upper  plot)  and
    position 8  (lower  plot).
                              165

-------
         80



         60

       u
       fi,

       I' 40



         20
         80
       a; 40
       UJ



         20
          0
                     100
                              JL
          J
200       300
TIME IN HOURS
                     100       200      300
                              TIME  IN HOURS
400
500
                   400
          500
Figure 85.   Pile 7, temperature at position 9 (upper plot) and
    position 10 (lower plot.) .
                               166

-------
          80
         60
       o
       o
       §•  40
       UJ
       i-
          20
                     100
200       300

TIME IN HOURS
400
                                      500
         80
         60
       o
       a: 40
       a
       UJ
       I-
          20
           0
            0
100       200      300

         TIME IN HOU^S
                   400
         300
Figure 86.   Pile 7, temperature at position 11  (upper  plot)  and

     position 12  (lower  plot).
                              167

-------
        80
        GO
      o
      0
      Si 40
        20
         0
                    100
200      300
TIME IN HOURS
400
500
                    100       200      300
                             TIME  IN HOURS
                   400
          500
Figure 87.  Pile 7, temperature  at  position 13 (upper plot) and
    position 14  (lower plot).
                              168

-------
         80
        60
      o
      UJ
      K
         40
         20
                    100
         200       300
         TIME IN HOURS
                   400
          500
        80
        60
        40
         20
          0
           0
100
200      300
TIME IN HOURS
400
500
Figure 88.  Pile 7,  temperature  at position 15 (upper plot) and
    position 16  (lower  plot).
                             169

-------
         BO
         60
       o
       o
         40
         20
                    100      200       300
                             TIME IN HOURS
                   400
          500
         80
         60
       o: 40
       S
       UJ
         20
                    100
200      300

TIME IN HOURS
400
500
Figure 89.  Pile 8, temperature  at position 1 (upper plot) and
    position 2  (lower plot).
                             170

-------
         80
         60
       o
       o
         20
         00
         60
       I
       UJ
         20
                     100       200       300
                              TIME IN HOURS
400
500
                     100       200       300
                              TIME IN HOURS
400
500
Figure 90.    Pile 8,  temperature at position  3  (upper plot)
    position  4  (lower plot).
              and
                              171

-------
       80
       60
       4°
       20
                  100      200       300
                           TIME IN HOURS
                   400
500
       80
       60
       40
     UJ
       20
                  100
200      300
TIME IN HOURS
                                              400
500
Figure 91.   Pile 8, temperature  at position 5 (upper plot) and
    position 6  (lower plot).
                              172

-------
         80
         60
       u
       -' 40
         20

                    100      200       300
                             TIME IN HOURS
400
500
         80
         60
       a: 40
         20
                    100      200       300
                             TIME IN HOURS
400
500
Figure 92.    Pile 8,  temperature at position  7  (upper plot)  and
    position  8  (low^.r plot).
                             173

-------
      o
      o
         80
         60
         40
         20
                    iOO      200      300
                             TIME IN HOURS
                   400
          500
         80
         60
       o
       1 40
         20
          0
                    100
200      300
TIME IN HOURS
400
                                                         300
Figure 93.   Pile 8, temperature  at position 9 (upper plot) and
    position 10  (lower plot).
                             174

-------
        00
        60
      UJ
        20
          0
100       200      300
         TIME IN HOURS
400
500
        80
        60
      CJ
        40
      UJ
        20
                   100
         200      300
         TIME  IN HOURS
40O
'600
Figure 94.   Pile  8,  temperature at position 11  (upper  plot)  and
    position 12  (lower  plot).
                              175

-------
        80
        60
      o
      0
      UJ
        20
                   100
200       300
TIME IN HOURS
400
500
        80
        60


      u

      a: 40
      UJ



        20
         O
                   100
200      300
TIME IN HOURS
400
500
Figure 95.   Pile  8,  temperature at position  13  (upper plot)  and
    position 14  (lower plot).
                              176

-------
         80
        60
      o
      o
      2! 40
      UJ
      H
                    100       200       300
                             TIME IN HOURS
                   400
          500
        80
        60
        40
        20
                    100
200      300
TIME IN HOURS
400
SOO
Figure 96.   Pile  8,  temperature at position 15  (upper  plot)  and
    position 16  (lower  plot).
                              177

-------
        60 -
        60  ~
     o
     o
     I  4°
        20
                  —r

                   100       200       300       400

                            TIME IN HOURS
                            500
        80 -
        60
     a;  40

     HJ
     H




        20
                   100
200      300

TIME IN HOURS
400
500
Figure 97.   Pile 8' temperature  at position 17  (upper  plot)  and

    position 18  (lower plot).
                              178

-------
         80 ~
         60  ~
      o
      o
      a:  40
      UJ
      (-
         20
                    100
200       300
TIME IN HOURS
400
500
         60
         60
      o
         40
         20
                    100
200       300
TIME IN HOURS
400
500
Figure 98.   Pile S, temperature  at  position 19 (upper plot) and
    position 20  (lower plot).
                             179

-------
         80
         60


       a
       g_

       a 40
       s
       UJ
         20
            0
100
200       300
TIME IN HOURS
400
                                                           500
         80
         60
       o: 40
       S
       UJ
         20
                     100
         200      300
         TIME IN HOURS
                   400
          500
Figure 99.   Pile  8,  temperature at position  21  (upper plot)  and
    position 22  (lower  plot).
                               180

-------
         80
        60
        40
      UJ
        20
                    100
         200       300
         TIME IN HOURS
                   400
500
        80
        60
      o
       : 40
        20
           0
100
200      300
TIME IN HOURS
                                               400
500
Figure 100.   Pile 8,  temperature at position  23  (upper  plot)  and
    position  24  (lower plot).
                             181

-------
         80
         60
       o
       0
       a: 40
         20
          0 L.
           0
100
200       300

TIME IN HOURS
400
500
         80
         60
      a:  40
         20
                    100
         200      300

         TIME IN HOURS
                   400
         500
Figure 101.   Pile  8,  temperature at position  25  (upper plot)  and

    position  26  (lower plot).
                              182

-------
       UJ
          80 h
          60
          40
          20
          80
          60


       CJ

       a:  4O
       UJ
       (-


          20
            0
                     100
100
         200       300
         TIME IN HOURS
                   400
          500
200      300
TIME IN HOURS
400
SOO
Figure 102.   Pile  8,  temperature at position  27  (upper plot) and
    position  28  (lower plot).
                             183

-------
         80
         60
       o
       o
         4C
       ui
                     100       20O       300
                              TIME IN HOURS
400
500
         80
         60
       o
       o
         40
         20
                     IOO       20O       300      400
                              TIME IN HOURS
         500
Figure 103.  Pile  8,  temperature  at position 29 (upper plot)  and
    position 30  (lower plot).
                             184

-------
       80
       60
     o
     UJ
     H
       20
                  100
200      300
TIME IN HOURS
400
500
Figure 104.   Pile  8,  temperature at position 31
                             185

-------
          60
       o
       o
          4°
          20
                     100       200      300
                              TIME IN HOURS
                            400
          500
         SO
       CJ
         40
          20
            0
100       200      300
         TIME IN HOURS
400
500
Figure 105.   Pile 9A,  temperature at position 1 (upper plot) and
     position 2  (lower  plot).
                              186

-------
        o
        UJ
        H
          40
          20
            0
          80



          SO



        K 40
        s
        t-

          20
100       200      300
         TIME IN HOURS
                     100
400
500
         200      300      400       300
         TIME  IN HOURS
Figure 106.   Pile 9A,  temperature at position 3  (upper  plot)  and
    position  4  (lower  plot).
                               187

-------
         so
         60
       o
       0
       UJ
       (-
         40
         20
                     100      200       300
                             TIME IN  HOURS
400
500
         90
         60


       CJ
       S»-
       a: 40
       UJ
       H


         80
                    100       200       300
                              TIME IN HOURS
400
300
Figure 107,  Pile  9A,  temperature at position  5  (upper plot)  and
    position 6  (lower  plot).
                              188

-------
        80
        60
      CJ
        40
        20
                    100       200       300
                             TIME IN HOURS
                            400
          500
        80
        60
      3:
      UJ
      t-
        40
        20
           0
100       200      300
         TIME IN HOURS
400
300
Figure 108.  Pile 9A, temperature at position 7  (upper plot)  and

    position 8  (lower plot).
                              189

-------
        80
        60 ~
      o
      o
      o: 40
      UJ
      I-


        20
                   100      200       300
                            TIME IN HOURS
                   400
          500
        80
        60


      CJ

      a 40
      UJ



        20
                   100
200      300
TIME IN HOURS
400
Figure 109.  Pile 9A,  temperature at position 9  (upper  plot)  and
    position 10  (lower  plot).
                             190

-------
         80
      o
      0
      I 40
      UJ
        20
         0
                    100
200      300

TIME IN HOURS
400
500
        60
      UJ
      i-
        20
                    100
?,00      300

TIME IN HOURS
                                               400
         SOO
Figure 110.  Pile 9A, temperature at position 11  (upper  plot)  and

    position 12  (lower plot).
                              191

-------
          60
        L)J
        h-
          20
                      100
200      300       400
TIME IN HOURS
          500
          80
          60

          40
        UJ
          20
           0
                      100
200      300
TIME IN HOURS
400
500
Figure 111.   Pile  9A'  temperature at position 13 (upper plot.)  and
     position 14  (lower plot).
                               192

-------
         80
         60
       o
       0
         4°
         20
                    100      200       300
                             TIME IN HOURS
                  400
          500
         80
         60
       o
       o
         40
       UJ
         20
                    100
200      300
TIME IN HOURS
400
500
Figure 112.   Pile  9A,  temperatnie at position 15  (upper  plot)  and
    position  16  (lower plot).
                              193

-------
        80
      o
      o
      9=" 4r
      UJ

      t-
        20
         0
                   100      200       300

                            TIME IN HOURS
                            400
                            500
        00
      o
      o
        40
      UJ
      t-
        20
          0
100
200      300

TIME IN HOUR3
                                               400
800
Figure 113. Pile  9A,  temperature at position  17  (upper plot) and

   position 18  (lower plot).
                             194

-------
       80 -
     o
                   100      200       300
                           TIME IN HOURS
400
500
Figure 114. Pile  9A,  temperature at position 19
                            195

-------
         80
         60
       o
       0
       £ 40
       UJ
         20
" too"
                              200       300
                              TIME IN .UOURS
                   400
800
         €0


       'G
       o

       o: 40
       UJ
       i_


         20
                     100
200      300
TIME IN HOURS
                                                400
900
Figure 115.  Pile  9B,  temperature at position  I  (upper plot)  and
    position 2  (lower  plot).
                              196

-------
        80
        60
        40
      UJ
        20
                    100
200       300
TIME IN HOURS
400
500
        60

      e>

      a: 40

      I—


        20
                    100
200      300
TIME IN HOURS
400
300
Figure 116.  Pile  9B,  temperature at position 3  (upper  plot)  and
    position 4  (lower  plot).
                              197

-------
         80
         40
         20
                     100       200      300
                              TIME IN HOURS
                   400
          500
         60
       SL,

       g;
       e£,
       UJ
         20
                     100
200      300
TIME IN HOURS
400
500
Figure 117.  Pile  9B,  temperature at  position 5 (upper plot)  and
    position 6  (lower  plot).
                              198

-------
        60

      o


      I 40
      ui
      t-


        20
        80
        60
        40
      UJ
      h-
        20
                    100       200       300
                             TIME 'N HOURS
                   400
          500
                    100
200      300
TIME IN HOURS
400
                                                         500
Figure 118.  Pile  9B,  temperature at position 7  (upper plot)  and
    position 8  (lower  plot).
                            199

-------
         80
        60
      o
      UJ
      h-
         40
        20
                    100       200       300
                             TIME IN HOURS
                   400
          500
        60 -
      & 40
      UJ
      h-
         20
                    100
200      300
TIME IN HOURS
400
                                                         300
Figure 119.  Pile  9B,  temperature at position 9  (upper  plot)  and
    position 10  (lower plot).
                             200

-------
       80
       60
     u
     I 40
     LU
     t-
                   100      200       300
                            TIME IN  HOURS
                   400
          500
       80
       SO
     I 4°
     UJ
       20
          0
                   100
200       00
TIME IN HOURS
400
500
Figure 120.  Pile 9B, temperature  at position 11  (upper plot)  and
    position 12  (lower plot).
                              201

-------
         80
         60
      o
      Si  40
      UJ
      1-
         20
          0
         80






         60



      0__

      o:  40

      tu
      t~



         20
                    100      200       300
                             TIME IN  HOURS
                   400
          500
                    100
200       300
TIME IN HOURS
400
500
Figure 121.   Pile 9B,  temperature at position 13 (upper plot)  and

    position  14  (lower plot).
                              202

-------
       80
       60
     %' 40
       20
                  100      200       300       400
                           TIME IN HOURS
500
Figure 122. Pile 9B, temperature  at position 15
                             203

-------
                                                                        n>
                                                                        in
                                                                        -~4


                                                                        05
                                                     O
                                                     X
                                                     X
                                                     SQ
                                                     (D
                                                     3

                                                     O
                                                     CT
                                                     C/l
                                                     (D
                                                                            0)
                                                                            rt
•X)
m

o
i—i
x

>
o
t-o
                        [00
               200           300
                  TIME IW HOURS
                                                                 400
500
Figure  123.   Pile 7
     position 6.
concentration cf 07. The  gas sampling  probe was adjacent to

-------
o
Cn
                 20
                O
                >
                LJ IS
                O
                <£
                111
                0.
                 12
                HE
                I-
                O
                O
                O
                o
                at
                                100
2OO           300
   TIME IN HOURS
4OO
                                                                                       500
         Figure  124,  Pile 1, concentrations of 0_.  The  gas  sampling probe  was adjacent  to
             position 14.

-------
                      100
200           300
   TIME IN HOURS
400
                                                                            500
Figure 125. Pile  8,  concentrations  of 0-.   The gas sampling probe was  adjacent  to
    position 11.

-------
                      100
              200          300
                 TIME IN HOURS
                                                             400
                                                                          500
Figure 126. Pile
    position  26.
concentration of Op.  The  gas  sampling probe was adjacent to

-------
ro
a
oo
                                            200           300

                                               TIME IN HOURS
                                                                       400
                                                          5OO
         Figure 127.  Pile
             position 28.
8, concentration of  0_.   The  gas sampling probe was adjacent  to

-------
                             APPENDIX A-3


     MOISTURE CONTENT IN THE "WHOLE SAMPLE" AND THE "NON-WOODCHIP
                              FRACTION"

     When the material being composted was a mixture cf sewage sludge
and woodehips, two types o£ samples were usually processed for the
determination of moisture content.  These were the whole sample 'and the
non-woodchip fraction.  The non-v/oodchip fraction was prepared by re-
moving the woodehips by hand.  Except where not otherwise noted, the
data reported in the body of the report refer to the whole sample.
Both sets of data are reported in Appendix A-3, for comparative pur-
poses.

     At time-zero the moisture content of the non-v/oodchip fraction
was higher than that of the whole sample (Figures 128,  129, 130).  The
diffe^-iiee between the moisture content of the two kinds of samples at
time-zero was:   pile 8,  201; pile 9A, S%; pile 9B, 8%.   The difference
disappeared in the piles which dried extensively (Rutgers strategy),
but not in the pile which experienced only slight drying (Beltsville
process).
                                 209

-------
                                                                                       o
                                                                                      IjO
                                                                                      • 
-------
                           WHOLE SAMPLE
                           NON-WOODCHIP  FRACTION
                        100
                                      200            300
                                         TIME IN HOURS
                                                                   400
                                                                                 500
Figure  129.   Pile 9A, moisture r-ontent.

-------
t-o
I—I
ISJ
                  70
                 ui
                 K.
                 t-
                 
-------
                             APPENDIX A-4

                    REPRESENTATIVENESS OF  PILE 9B

       Since  the  Beltsville  Process  was  represented  herein by only one
pile  (9B), its representativeness was evaluated  in  reference to com-
parable  data reported  by  other  investigators.  Also,  the  grouped Belts°
vine  data was compared to  grouped  Rutgers  data.  The analysis  is
summarized in TABLE 21=

       The first  entry  for temperature (>70°C)  is  based on terminal
(day  21) observations  (38),  or  on mean  values  (9).  One of the  reports
(39)  does not provide  usable  temperature  data.   The second entry for
the Beltsville process  (80°C) js based  on the  extensively monitored
pile  9B, in  which  the  bulk  of the material  peaked at  78°C to 82°C.
The entry for the  Rutgers process is  based  on  the piles described in
Section  3, in which the bulk  of the  material  generally did not  exceed
60°C.

       The first  entry  concerning ventilation  (0,  level, 5-15%)  is
based  on one pile  at the end  of the  21  day  period.  The second  entry
OlOt) is an "average" value, the precise meaning of  which is diffi-
cult to  evaluate.   The third  entry  (12-214) is based  on pile 9B  over  a
21 day period, in which the  gas sampling probe -was centrally positioned.
The entry for the Rutgers process (16-21%)  is  from pile 9A at a  compar-
able position, and  represents a period  of 21 days.  On  average,  the
Beltsville piles v/ere provided  with  134 cfh/v/et ton during the  standard
21 day process period.   For the Rutgers piles  the blowers  provided, on
average, 1232 cfh/wet ton during the  12 day period of the  feedback
control.

      The first  entry concerning moisture content (7% decrease)  is
based on samples taken from four locations  (six replicates per location)
of one pile on the  terminal day of standard processing  (38).  Reference
9  does not include  suitable moisture  content data.  The second entry
(14$ decrease) represents  114 piles on the terminal day (processing was
usually for 21 days).  Th  .- third entry  (4% decrease)  is the  final value
of the series representing pile 9B.    The fourth entry  (.341 decrease)
represents the mean final  value of the indicated Rutgers piles.
                                213

-------
       TABLE  21.  COMPARISON OF FIELD DATA FOR RUTGERS AND BELTSVILLE PROCESSES.
Peak temperature Ventilation Decrease i
representing the requirement moisture
bulk of the pile to maintain an content
Process


Beltsville
Beltsville
Beltsville
Rutgers

( C)


>70
_.
80
60

oxygenated
02 level
(1)
5-15
10
12-21
16-21

condition (£)

cfh/ton
60-133 7
212* 14
132 4
12323 34

.n
Process
time Reference or
(days) pile


21 9, 38-39
21 39
21 Pile 9B
12 Files 4AiJ, 1]
8, 9A
Woodchips included in the sample, except that for pile 4A the saiaple was sieved before
the determination.

Mean delivery per ton (wet wt) of the initial iludge-woodchip mixture.  To convert  from
cffh to m^/Sj, multiply x 7,87 x 10" ; to convert from ton to tonne  (metric), multiply
The published value is 123 cfSi/yd .   This was converted to the common means of expression
by assuming that 1.74 yd3 weighs 1.0 ton.

Peak demand ^ 4800 cf Hi/ton

See Section 4.

-------
                            APPENDIX A-S

       FORCES DRIVING VAPORIZATION: METABOLICALLY GENERATED HEAT,
       AND UNSATURATION OF INLET AIR

     The experimentally derived value of 4.5% was calculated as  the
H-,0 removed per unit volume air from a metabolically  inert pile  o£
cored compost x lOO/HjO removed per unit volume air from an actively
composting pile,   (See Section 8 for observations on  a pile o£ well-
cured compost -- pile 12.)  The calculation leading to the theoretical
values (1.8% to 5.0%) assumes that ambient inlet air  at 20°C and 30%
relative humidity  (RH) is introduced into two physically identical
piles: one an actively composting pile; the other a metabolically
inert pile.  Upon exiting from the composting pile the air is at 60 C
and 100% RH, hence water removal amounts to 0.14S kg/kg dry air  (46).
For the metabolically inert pile two extreme cases are developed.
The first assumes that the air experiences the maximum theoretical
temperature decrease with passage through the pile.    Upon exiting
the air is at 13.3°C and 100% RH, hence water removal is 0.0026 kg/kg
dry air.   The second case assumes no temperature decrease, giving
exit air at 20°C and 100% RH and water removal of 0.0072 kg/kg'dry
air.  Thus, in this calculation non-biological air drying caused by
unsaturation of the air accounts for only 1.8% to 5.0$ of the water
removal from the composting pile.  The remainder is attributable to
uiierobial heat generation.
                                21S

-------
                             APPENDIX A-6

        RELATIVE SENSITIVITY OF THE MOISTURE  CONTENT AND VOLATILE
                             SOLIDS TESTS


     Consider the relationship between the decrements of organic
matter and water on a mass basis.  Assuming a release of 6000 cal
from the micTobial oxidation of 1 g organic matter to CO* and H^O,
and with the heat of vaporization of water at 20°C (the assumed
temperature of the inlet air and starting material) equal to 586
eal/g, 10.2 g water could be vaporized per g volatile matter de-
composed.  A small part of the water loss is made up through metabolic
water production (ca. , 0.8 g) .  Also, &A  (at  80°C exit temperature)  to
14% (at SO°C) of the heat removal is through dry air convection, and
another 21 (at SO C) to 41 (at 80°C) through raising the temperature
of the vaporized water.  Hence, the mass of water rsmoved provides
an indication of organic matter cecomposition that is 7.8 (at 50°C)
to 8.6 (at 80 C)-fold more sensitive than the mass of volatile solids
decomposed.

     The mass of water removed is rarely known, however, whereas the
$ moisture content is easily determined.  Under realistic conditions,
the ratio change in I moisture/change in % volatile solids exceeds -
unity at the outset of composting, and widens progressively due tt> the
different relative changes in the regaining masses of water and vola-
tile solids.   The widening of the ratio is mainly a function of the
initial moisture content and the initial volatile solids content, hence
no further generalization regarding its numerical value is possible.
In the ease of an initial moisture content of 75% (wet weight basis)
and volatile solids content of 7S% (dry weight basis), as the moisture
content traverses the 30% level the cumulative ratio is 3.S.   In
association with a decrease in moisture content from 31% to 30%, the
ratio is 9.8.
                               216

-------
                         APPENDIX  B

   ADVANTAGES  OF  FUEL  PRODUCTION THROUGH COMPOSTING  VS.
   DIRECT  COMBUSTION SEWAGE  SLUDGE CAKE
The  residue  of  the  composting  process  is  metabolically inert,
and  therefore relatively  easy  to  handle,  stockpile.,  and transport.

The  process  residue  is  dry  and granular,  or  can  be pulverized, to
granulate  it, making it relatively  easy  to  feed  into  the  combus-
tion chamber.

Separation of the drying  and combustion  functions  improves  over-
all  system reliability, in  that a breakdown  of the combustion
system does  not  interfere with the  sludge treatment  (ifuel  prepara-
tion) .

Thejnn£dyjriami c_A_dv an ta_ ge£

      Separation  of  the  drying  and combustion functions decreases
the  amount o£ energy spent  in  the vaporization o£ water.  This
stems from a comparison o£  biological  drying (composting) at 140°F
versus direct combustion  at 1SOO°F.  Note that for every  pound  o£
water vaporized  at  140°F  (via  biological  drying) , approximately
680  BTU's are saved  compared to combustion  at 1500°F.* This reflects
the  approximately O.S BTU's/lb/°F required  to raise  the temperature
of water vapor over  this  range.

      A further advantage, not  as  easily quantitated  in the  general
case, is that less  total  air is required  for the combustion.  For
every pound  of air not required,  approximately 370 BTU's  are saved,
reflecting the 0,26  BTU/lb/°F  specific heat  o£ air over this range.

      Energy must be  expended in biological drying  (composting) , of
course and this  decreases somewhat  the thermodynamic  advantage,
No exact accounting  is readily  accessible, but it 'is  evident that
biological drying is  advantageous compared to air drying  (see
TABLE 12.) _ ______        —
'~"~                     37th Edition, 1963.   The Babcock §
                 * New YorTc
                           217

-------
     a:
     UJ
     h-
        80
        60
        40
        20
          0
        60
     o
     e
     Q--  40
     UJ
     I-
        20
        APPENDIX C


  Temperature Observations


 for Piles 11A, 11B,  and  11C
IOO
200       300
TIME IN HOURS
400
500
                   100
         200      300
         TIME IN HOURS
                   400
         500
Figure 131.   Pile  11A,  temperature at position  1  (upper plot)
    and position  2 (lower plot).
                              218

-------
       eo
       60
     o
     o
     UJ

     H
       20
        0
                   100       200       300

                            TIME IN HOURS
400
                                     500
       80
     o
     SL,

     & 40

     UJ
          0
100       200      300

              IN HOURS
                                              400
         900
Figure 132.  Pile HA,  temperature at position 3  tupper plot)

    and position 4  (lower  plot).
                              219

-------
        80
        60
      o
        40
      UJ
      H
        20
           0

100       200      300
         TIME IN HOURS
                   400
          300
        80
        60
      u
        4°
      UJ
      t-
        20
           0
100
200      300
TIME IN HOURS
400
SOO
Figure 133.  Pile  11A,  temperature at position 5  (upper  plot)
    and position 6  (lower  plot).
                              220

-------
         80
         60
      o
      o
      UJ
         20

                    100       200       300
                             TIME IN HOURS
                   400
          500
        80
        60
      o
      0
        20
                    100
200      300
TIME IN HOURS
400
500
Figure 134.  Pile 11A, temperature  at  position 7 (upper plot)
    and position 8  (lower plot).
                              221

-------
       80
       60
       40
     UJ

     t-
       20
         0
100
200       300

TIME IN HOURS
400
500
       80
       60
     o
     o
       40
     UJ

     h-
       20
                  100
         200      300      400

         TIME IN HOURS
                            300
Figure 135.  Pile 11A, temperature  at  position 9 (upper plot)

    and  position 10 (lower plot).
                              222

-------
        80
        60
      o
      o
      •i™.'


      I 4°
      UJ
        20
                    100       200      300
                             TIME IN HOURS
                   400
          500
        80
        60
      o
      1 40
      UJ
        20
                    100
                                     __J=
200      300
TIME IN HOURS
400
500
Figure 136.  Pile 11 A,  temperature at position  11  (upper plot)
    and position 12  (lower  plot).
                             223

-------
                    100
200      300
TIME IN HOURS
400
500
                             200       300
                             TIME >N HOURS
                  400
         300
Figure 137.   Pile  11B,  temperature at position 1  (upper plot)
     and position 2  (lower  plot) „
                               224

-------
        80
      o
      o
      UJ
        40
        20
           0         100      200       300
                            TIME IN HOURS
                   400
          500
                   100
200      300
TIME IN HOURS
400
500
Figure 138.   Pile 11B, temperature at position  3  (upper plot)
     and  position 4 (lower plot).
                               225

-------
         6O
       o
       o
         40
       UJ
       j-
                    100
200      300
TIME IN HOURS
4OO
SOO
         80
       a.' 40
       LU
       H
         20
                    IOO
200      300
TIME IN HOURS
400
SOO
Figure 139.   Pile 11B, temperature  at position 5 (upper plot)
     and  position 6 (lower plot).
                                226

-------
       60
        40
     UJ
     s-
       20
                   100       200       300
                            TIME IN HOURS
400
500
       80
       SO
     u
     tu
                   100       200       300
                            TIME IN HOURS
400
BOO
Figure 140.  Pile 11B, temperature at position 7 (upper plot)
    and position 8  (lower plot).
                             227

-------
                   100
200      300
TIME IN HOURS
                                                        500
                   100
200
TIME IN HOURS
                                              400
300
Figure 141.  Pile 11B,  temperature at position 9 (upper plot)
    and position 10  (lower  plot).
                              228

-------
         60
       o
       e
          40
       UJ
       H
         20
                     100       200      300
                              TIME IN HOURS
400
500
          80
         60
       o
       SL,
         40
       UJ
          20
                     100       200      300
                              TIME IN HOURS
400
soo
Figure 142.   Pile  11B,  temperature at position  11  (upper plot)
    and position  12  (lower plot).
                              229

-------
         80
      o
      o
      UJ
      H
         40
         20
                    100
         200      300
         TIME IN HOURS
400
500
        80
      o
       ,
        40
      t-
        20
           0
100       200      300
         TIME IN HOURS
                                               400
         500
Figure 143.   Pile 11C,  temperature at position  1  (upper plot)
    and  position 2 (lower plot).
                              230

-------
       80
       SO


     o
     o^


     S! 40
     a
     LU
     t-



       20
                   100       200       300       400
                            TIME IN HOURS
                             500
       80
       60
    a: 40

    UJ
       20
                  100
200       300

TIME IN HOURS
400
500
Figure 144.  Pile  11C,  temperature at position  3  (upper plot)
    and position 4  (lower plot)„
                              231

-------
         80
        60
        40
      UJ
      I-
                    100       ZOO       300
                             TIME IN HOURS
400
                            500
      a:
        20
         0
                    100
200      300
TIME IN HOURS
                                               400
         iOO
Figure 145.  Pile 11C,  temperature at position 5  (upper plot)
    and position 6  (lower  plot).
                              232

-------
      o
        80
        60
        40
        20
         0
                    100
200      300
TIME IN HOURS
                                                400
                            500
        60
      cj
      o
      a: 40
      UJ
      t-
         20
                    100
200      300
TIME IN HOURS
400
300
Figure 146.  Pile  11C,  temperature at position  7  (upper plot)
    and position  8 (lower plot).
                               233

-------
       90
       60
     o
     o
       40
       20
                  100      200       300
                           TIME IN HOURS
400
500
       80
       60
    u
    o_

    a:
    §
    UJ
    h-
       20
                  100       200      300
                           TIME IN HOURS
400
300
Figure 147. Pile  11C,  temperature at position  9  (upper plot)
   and position  10  (lower plot).
                              !34

-------
         60
       o
       o
         40
         20
0
                     100
200      300

TIME IN HOURS
400
300
         80
         60
       o
       o
         40
       I-
         20
          0
                     100
200      300

TIME IN HOURS
400
500
Figure  148.  Pile  lie,  temperature at position 11  (upper plot)

    and position 12  (lower plot).
                              235

-------
                            APPENDIX D-l

             AIR NE^JED TO REMOVE HEAT AND SUPPLY OXYGEN


     A wide range of organic compounds shows little variation  in  the
energy released per mass of 0« used for complete oxidation  to  carbon
dioxide and water.  The mean value is approximately 14,000  kJ  released/
kg 0, utilized.  Using equation  (iv) and the values of h  t  from
TABLE 1 (for approach R) : Q^ *• 362 kJ/kg dry air; thus Ttutakes

                          m™
14000 = 38.7 kg dry air to remove 14,000 kJ o£ released energy.   Since
there is 0.232 kg 0,,/kg dry air, it takes 4.31 kg of dry air  to re-
plenish 1 kg of 02.  The ratio is 38.7 - 8.98.
                             APPENDIX D-2

UNSUCCESSFUL ATTEMPT BY THE BELTSVILLE GROUP TO IMPROVE DRYING,  IN  ISO-
LATION FROM CONSIDERATIONS OF PROCESS DYNAMICS


     In an attempt to improve drying, the "active stage" of the  Belts-
ville Process was extended from the standard 21 day period to 28 days,
and on day 14 ventilation was switched from the vacuum-induced to the
forced pressure direction and increased by approximately 4-fold  (10).
The modified process behaved as follows.  During the first 14 days  the
average peak interior temperature was 70°C.  The increase in ventilation
on day 14 marked the onset of a cooling trend, such that on day  21  the
average temperature was 42 C.  (In conventional Beltsville operation
the temperature remains at peak,  or near pejak, values throughout the
21 day period,)  This modification yielded only a modest improvement in
v/ater removal.   The initial moisture content of 63.6% decreased  to
52.1% in the 28 day period, compared to a terminal value of 56.3%
obtained in the usual 21 day period through conventional operation.

     This behavior is predictable based on expression iv, as con-
strained by the interaction between heat generation and temperature.
As is characteristic of the Beltsville Process, a sma31 jn resulted  in
a temperature ascent to biologically inhibitive levels.  Once peak
temperatures were reached (-vday S) h  £ was large but the resultant
0  was small, as evidsnced by the small moisture content decrease.  In-
creasing m on day 14 cooled the pile because the rate of biological
activity ~t=heat generation) could not sustain the peak temperature
against the increased rate of heat removal.  The decreasing outlet
temperatures were accompanied by decreasing values of h  t-   As pre=
dieted  the resultant .Q*  was small, judged by-the small Moisture
content decrement.
                                236

-------
       70
       50
     o
     o
     I 3°
     UJ
       -10
       70
       50
     I 3°
     UJ
     I-
        10
       APPENDIX E

Temperature Observations

        Pile 12
                   100      200       300
                            TIME IN  HOURS
                         400
          500
                            _L
                  100
       200       300
       TIME IN HOURS
400
                                                        500
Figure  149.  Pile 12,  temperature at position 1  (upper plot)
    and position 2  (lower  plot).
                             237

-------
       70




       50


     O


     I 30
     UJ
     t-


       10




      -10
       70
       50
       30
       10
      -10
                  100
                  100
200       300
TIME IN HOURS
200      300
TIME IN HOURS
400
400
500
500
Figure ISO.   Pile  12,  temperature at position  3  (upper plot)
    and position  4 (lower plot).
                              238

-------
       to
       80
     CJ
     I
     UJ
        10
       -10
                   100       200      300
                            TIME IN HOURS
                   400
          500
       70
       50
     o
     o
     LU
     h-
        10
       -10
                   100
200      300
TIME IN HOURS
400
SOO
Figure 151.    Pile 12, temperature  at  position 5 (upper plot)
    and  position 6 (lower plot).
                               239

-------
        70
       50
       30
        10
       -10
                   100
              200       300
              TIME IN  HOURS
400
500
Figure 152.
Ambient temperature during the experimental period
for pile 12.
                              240

-------
           APPENDIX F

Observations on Blower Operation,
Temperature, 02, C0?s pH, and
Moisture Content for Piles 6A, 6B,
            and 6C
Figure  153.Piles 6A,  6B,  and 6C,  cross-sectional  representation:   textured  area,  wood-
    chip cover and base;  clear area,  sludge-woodchip mixture;  circle,  perforated flex-
    hose.   The numbers indicate monitoring and control  positions:  thermocouples,
    positions 1 through 15;  7as sampling  probes,  adjacent to  position  13  (also adjacent
    to position 6 in pile  6A);  control  thermistor,  between positions  1 and 6.   The
    blower was operated in the forced-pressure mode.

-------
     100-ff
     o
     to)
     2
     LJ
     o
     
-------
      100-f
       80-
      o
      i-SO-ji
     o
     Q

     £C
     o
     _!
     IS.
       01.
                      JOO
                                   200
                                                3OO
                                           HOURS
                                                             400
Figure  155.  Pile  6B,  blower operation.  The baseline  represents operation as scheduled
    by  rimer,  and the area above the baseline represents operation through the temp-
    erature  feedback  control system.  The blower  was  operated in the forced pressure

    mode.

-------
       \oo-r
       80-!
      2
      O
      KSO-
      a
      a.
      O
      t- 40-
      «a
       20-
                                                             O
                                                                           500
                                   200           3(
                                      TIME IN HOURS
Figure 156.  Pile 6C,  blower operation.  The  baseline represents operation as  scheduled
    by  timer,  and the area above the baseline  represents operation through  the  temp-
    erature feedback  control system.  The blower was operated in the forced pressure
    mode.

-------
       80
       60
     u
     o
     tu
     h-
       20
        0
                   100       200      300
                            TIME IN HOURS
                   400
500
       80
       60


     O
     o^

     a: 40

     h-


       20
                   100
200      300
TIME IN HOURS
                                               400
Figure 157. Pile 6A, temperature  at  position 1 (upper plot) and
    position 2 (lower plot)„
                              245

-------
        00
        60
     u
     o
        40
     UJ
       20
          0
100      200       300      400
         TIME IN HOURS
                                                         500
       60
     u
     0
     a: 40
     UJ
     H
        20
          0
100
200      500
TIME IN HOURS
400
SOO
Figure 158,  Pile  6A,  temperature at position 3  (upper  plot)  and

    position  4  (lower plot).
                              246

-------
    o
    o
    UJ
       60
       40
       20
        0
         0

100
200       300

TIME IN HOURS
400
500
       80
       60
    o
    o
    Q.' 40
    UJ
    t-
       20
        0
         0
100
200       300

TIME IN HOURS
                                               400
                                      500
Figure 159.  Pile 6A,  temperature at position 5  (upper plot)  and

    position 6  (lower plot).
                              247

-------
        80
        40
      UJ
      H-
        20
                    100       200      300
                             TIME IN HOURS
                   400
          500
        80
        60
      Q-' 40
        20
                    100
200      300
TIME IN HOURS
400
500
Figure 160. Pile 6A, temperature  at  position 7 (upper plot) and
    position 8  (lower plot).
                             248

-------
        QO
        60
      o
      o
        40
      UJ
      t-
        20
         0
        80



        60


      JL
      a: 40
      in


        20
                    100
                    100
200       300
TIME IN HOURS
400
500
200      300
TIME IN HOURS
                                               400
         300
Figure 161. pile  6A,  temperature at position 9  (upper  plot)  and
    position 12  (lower  plot).
                              249

-------
       80
       80
     o
     o
       40
     UJ
     t-
       20
                   100       200       300
                            TIME IN HOURS
400
       80
       60
     o
     o
     Q-' 40
     UJ
     t-
       20
                   .00       200       300
                            TIME IN  HOURS
400
500
Figure 162.  Pile  6A,  temperature at position  13  (upper plot)
    and position 14  (lower plot).
                              250

-------
       80
      60
    UJ
    t-
      20
       0
                  100
200       300
TIME IN HOURS
400
500
Figure 163.   Pile 6A, temperature  at position  15,
                             251

-------
      o
      o
      UJ
      t-
        80
        60
        40
        20
                    100       200       300

                             TIME IN HOURS
                   400
                             500
        80
        60 -
      o
      p^


      a: 40
      S
      UJ
        20
         0
                   100
200       300

TIME IN HOURS
400
500
Figure 164.   Pile  fB,  temperature at position  1  (upper plot)

    and position 2  (lower plot).
                              152

-------
         60
       I 40
       UJ
         20
                     100
         200       300
         TIME IN HOURS
                   400
          500
         80
         60
       u
       o
       Q,'
       s
       UJ
         20
           0
100
200      300

TIME IN HOURS
400
500
Figure 365.   Pile 6B, temperature at position 5  (upper plot)

     and position 7  (lower  plot).
                             253

-------
        80
        60
      o
      o
      UJ
      I—
        20
                   100      200       300
                            TIME IN HOURS
                   400
          500
        80
        60


      cj
      o__
      o; 40
      UJ
      (-


        20
                   100
200      300
TIME IN HOURS
400
500
Figure  166.   Pile 6B, temperature  at position 9  (upper plot)
    and position 12  (lower plot).
                              254

-------
     u
     o
     UJ
     H
       80
       SO
       40
       20
                   100       200       300

                            TIME IN HOURS
400
500
Figure 167.    Pile 6B, temperature  at  position 13,
                               255

-------
        80
       60
        40
       20
        o t=~
                   100       200      300
                            TIME IN HOURS
                   400
          500
     o
     8-

     I 4°
       20
                   100
200      300
TIME IN HOURS
400
500
Figure 168.  Pile  6C,  temperature at position 2  (upper plot)
    and position  5  (lower plot).
                              256

-------
        00
        60
      o
      0
        40
        10
                    100      200       300      400
                             TIME IN  HOURS
                            300
        80
        60
      UJ
      t-
        40
        20
                    100
200      300
TIME IN HOURS
400
500
Figure 169.   Pile 6C, temperature at position 7  (upper plot)
    and position 9  (lower  plot).
                              257

-------
        80
        60
        40
      UJ
      t-
        20
                   100      200
                            TIME IN  HOURS
                   400
                   100
200      300
TIME IN HOURS
400
500
Figure 170.   Pile 6C, temperature  at position 12 (lower plot)
    and position 13  (upper plot).
                            258

-------
                100
                            200          300
                              TIME IN HOURS
                                                    400
                                                                500
Figure 171.    Pile  6A,  concentrations of 02 (upper plot) and C02
     (lower plot).
     position 6.
The gas sampling probe was  adjacent to
                               259

-------
               100
                          200         300
                             TIME IN HOURS
                           400
                                       500
Figure 172.   Pile 6A, concentrations of CU (upper plot)  and
    CO 2 (lower plot).
    position 13.
The gas sampling  probe  was adjacent to
                             260

-------
  20
  UJ
  3.
  :D
  _j

  §16
  z
  UJ
  UJ
  CL
  I-
  
-------
  20
 §16
 UJ
 o
 UJ
 Q.
 I-
 Z
 UJ
 o
 o
 O
 K
 UJ
               100
                           200          300
                             TIME IN HOURS
                               400
                                            500
Figure  174.   Pile 6C,  concentrations  of 02 (upper plot)  and C02
     (lower  plot).
    position  13.
The gas sampling probe was adjacent to
                              262

-------

                        PILE 6A —
                        PILE 68 	
                        PILE SC —
                                                       	o....
                                                                    ••©
                      100
                                   200           300
                                      TIME  IW HOURS
                                                               400
                                                                             500
Figure  175.   Piles 6Af 63,  and 6C, pH.  The  samples were  from interior  locations.

-------
      70
      SO
     I-
     OT
     o
      50
     o
     £E
                       PILE SA
                       PILE SB
                       PILE SC
      30
      20 U

                     100
200            300
   TIME IN HOURS
                                                                           soo
Figure 376.  Piles  6A,  6B,  and 6C, moisture content.  The  samples were from
    interior locations.

-------
                                       19.5 29.6 ---  64.3  62.9
                                      27.2 31.6 32.7 50.4 (
           -4.7 -1.8 +().()

       -0.8 +0.2 + 0.5 + 0.2


                1.1 +3.1


10.3 +18.0 +17.6 +53.8 (
Figure  177. Pile  6B,  special terminal  test of air penetration.  Temperature {  C)
    immediately  prior  to the introduction of heated air,  left hand cross-section;
    temperature  30 min after start  of introduction of  heated air, middle  cross-
    section; temperature differential,  right hand cross-section.

-------
       80
    a:  40 -
    UJ
    H-
       20 ~
      APPENDIX G

Temperature Observations

       Pile 13
                           200      300
                           TIME IK) HOURS
                           200      300
                           TIME  IN HOURS
                          L
                        400
eoo
Figure 178.   Pile  13,  temperature at position 1 (upper plot]
    and position 2  (lower  plot).
                             266

-------
        80
       SO
     o
     o
     UJ
     (-
        40
       2C
                   100
200       300

TIME IN HOURS
400
500
                   100
200      300

TIME IN HOURS
400
500
Figure 179.   Pile  13,  temperature at position 3  (upper plot)

    and position  4 (lower plot).

-------
        80
        60
      a:
      UJ
      H-
        40
        20
                    100
         200       300
         TIME IN HOURS
400
500
        80
        60


      o"
      o__

      a: 40
      UJ
      (-


        20
           0
100       200       300
         TIME IN  HOURS
                                                400
          500
Figure 180.    Pile 13,  temperature  at position 5  (upper  plot)
     and position 6 (lower plot)„
                                268

-------
                   100
200       300
TIME IN HOURS
400
500
         0 *•=
                   100
200
TIME IN HOURS
                                              400
         500
Figure 181.   Pile  13,  temperature at position  7  (upper  plot)
    and position  8 (lower plot;.
                              269

-------
        60




     |  40
     UJ



        20
        80
     e_^


     1 4°
     UJ
        20
                   100
200
TIME IN HOURS
400
                   100       200      300
                            TIME IN HOURS
400
500
500
Figure  182.   Pile 13,  temperature at position  9  (upper plot)
    and position 10 (lower plot).
                              270

-------
         80 ~
         60
       o
       e
       a:
         4O
         20
                    100      200       300
                             TIME IN HOURS
                   400
          500
         8O
         60
       o
         20
                    100
2OQ      300
TIME IN HOURS
400
500
Figure 183.   Pile  13,  temperature at position  11  (upper plot)
    and position 12  (lower plot).
                              1.71

-------
        80 -
      o
      o
       ; 40 -
         0 L
        80
      a: 40  -
      S
      UJ
        20 L.
         0
                    100
200       300

TIME IN HOURS
                             200      300

                             TIME  IN HOURS
                   400
500
300
Figure 184.   Pile 13,  temperature at position  13  (upper  plot)

    and  position 14 (lower plot).
                              272

-------
         §0
       u
       o
       a:
       LU
       {=
                     100
200       300

TIME IN HOURS
400
500
         80
         60
       o
       o
       a, 40
         20
                    100
200      300

TIME IN HOURS
400
Figure 185.    Pile 13, temperature  at  position 15 (upper plot)

     and position 16  (lower plot).
                               273

-------
        80
        60
      2_
      a:
      UJ
      t-
        20
                    IOO       200       300
                             TIME IN HOURS
                   400
          500
        80
        60
        20
                    IOO
200      300
TIME IN HOURS
400
500
Figure 186.   Pile  13,  temperature at position  17  (upper plot)
    and position 18  (lower plot).
                              274

-------
       80
       60  -
       40
       20
                  100       200       300
                           TIME IN HOURS
400      500
Figure 187.  Pile 13,  temperature at position  19,
                             275

-------