WATER POLLUTION CONTROL RESEARCH SERIES • ORD- 17O5OFIMO5/7O
          "OPTIMIZING LIPID
          BIOSTABILIZATION"
U.S. DEPARTMENT OF THE INTERIOR • FEDERAL, WATER QUALITY ADMINISTRATION

-------
             WATER POLLUTION CONTROL RESEARCH SERIES


The Water Pollution Control Research Reports describe the results
and progress in the control and abatement of pollution of our
Nation's waters.  They provide a central source of information on
the research, development, and demonstration activities of the
Federal Water Quality Administration, Department of the Interior,
through in-house research and grants and contracts with Federal,
State, and local agencies, research institutions, and industrial
organizations.

Water Pollution Control Research Reports will be distributed to
requesters as supplies permit.  Requests should be sent to the
Planning and Resources Office, Office of Research and Development,
Federal Water Quality Administration, Department of the Interior,
Washington, D. C. 20242.

-------
     OPTIMIZING LIPID BIOSTABILIZATION
                      by
                William Garner
          Midwest Research Institute
          Kansas City, Missouri 64110
                    for  the

    FEDERAL WATER QUALITY ADMINISTRATION

          DEPARTMENT OF  THE INTERIOR
              Program #17050 FIM
              Contract #14-12-198
     FWQA Project Officer,  R. C. Brenner
Advanced Waste Treatment Research Laboratory
               Cincinnati, Ohio
                   May,  1970
  For sale by the Superintendent of Documents, U.S. Government Printing Office
             Washington, D.C. 20402 - Price 60 cents

-------
           PWQA Review Notice
This report has been reviewed by the Federal
Water Quality Administration and approved for
publication.  Approval does not signify that
the contents necessarily reflect the views
and policies of the Federal Water Quality
Administration, nor does mention of trade
names or commercial products constitute
endorsement or recommendation for use.

-------
                            TABLE  OF CONTENTS

                                                                  Page
Table  of  Contents	     ii

Abstract	    ill

Acknowledgements	     iv

Introduction	      1

Experiment  I  - Performance of Digesters on Standard Loading and
  Mixing  Regimen	      5

Experiment  II - Effect of High-Shear Homogenization on the
  Digestion of Purina Dog Chow	      9

Experiment  III - Effect of Homogenization on Limed Digesters
  Heavily Loaded With Cottonseed Oil	     12

Experiment  IV - Effect of Homogenization on Soda-Ash Buffered
  Digesters When Dog Food Loading is Augmented With Cottonseed
  Oil	     17

Experiment  V  - Effect of Homogenization on Soda-Ash Buffered
  Digesters as Cottonseed Oil Gradually Replaces Purina
  Dog  Chow  as a Feed	     20

Experiment  VI - The Effect of Motor Oil on the Digestion of
  Purina  Dog  Chow and Cottonseed Oil	     23

Discussion	     26

Conclusions and Recommendations 	     30

Appendix A  - Apparatus and Methods	     32

Appendix  B  - Analytical Data	     49

Bibliography	     55
                                    ii

-------
                                ABSTRACT
          Laboratory-scale anaerobic digestion studies were carried out
to determine the effect of high-shear mixing (homogenization)  on the deg-
radation of lipids .   The studies showed that the intensity of  mixing must
be carefully tailored to the rate and type of feed if benefits are to be
realized.

          Dog food,  cottonseed  oil and motor oil were fed to  the digesters
daily as slug loads.  When properly operated, the 15 replicates digesters
gave reproducible results with good precision.

          High-shear mixing has little effect on degradation of dog food
and was deleterious  to lime-buffered solutions heavily loaded  with cotton-
seed oil.  When soda ash was used as a buffering agent, heavy  loads of oil
caused the digester to "go sour" regardless of mixing system,  although
homogenization gave somewhat better results.  Under some conditions, homo-
genization led to serious foaming in the digesters .  Kinetic data and
chemical analysis of drawdown samples confirmed the hypothesis that homo-
genization or soda ash buffering can accelerate the hydrolysis of a heavy
load of fat to a point where the saponif ication products overload the
methane fermentation or affect it by surface effects.  Motor oil was not
readily digested, but did not appear to influence the digestion of cotton-
seed oil or dog food.

          Other observations were:  more rapid gasification occurs as the
feed rate is increased; lime and. soda ash cannot be used interchangeably
as buffers; feed COD cannot be used to predict methane yield;  and the tem-
perature response of a system greatly exceeds Q^Q = 2.
          This report was submitted in fulfillment of Program No. 17050 FIM,
Contract No. 14-12-198, between the Federal Water Quality Administration and
Midwest Research Institute.
                                   iii

-------
                            ACKNOWLEDGEMENTS
          This report was written by Dr. William Garner,  who was  also the
project leader.  Other Midwest Research Institute personnel who contributed
to the project were Messrs. Eugene Sallee, Tom Shearon,  John Morrison,
John Maberry, Earl White, Aubrey Cloud, John Unrau, Merrill Nissen,
Paul Garrett, Jack Honts, Mrs. Jeanne Robertson and Dr. Walter Langston.
Dr. Ross E. McKinney, Parker Professor of Civil Engineering, University of
Kansas, and Dr. Raymond C. Loehr, Professor of Civil and  Agricultural
Engineering, Cornell University, were consultants to the  research program.
Dr. J. Earl Barney II, former Head of the Analytical and  Environmental
Sciences Section, provided technical supervision for the  project.
                                   iv

-------
                              INTRODUCTION
          Wastewater must be renovated before it is returned to the  envi-
ronment or reused.  At present, domestic wastewater is processed in  the
United States in order to remove pathogens and reduce the organic matter
which, in serving as growth media for common soil and aquatic bacteria,
would deplete the receiving stream of dissolved oxygen.   The two unit
processes of biosorption and settling are employed for this purpose.   Set-
tled organic matter consists of raw wastes as well a^ the flocculated
bacteria that have adsorbed and absorbed the non-settleable organic  matter
in the stream.  This settled organic matter is highly putrescible, and it
must be stabilized before it is spread on agricultural land, used as land-
fill, or disposed of at sea.

          Sludge can be stabilized by chemical oxidation in the processes
of incineration or wet oxidation.  It can also be stabilized by microbio-
logical oxidation and by the process of methane fermentation (or anaerobic
digestion).  This latter process has been included in the design of  most
of the larger plants that have been constructed in the United States during
the past several decades.!./  Digestion has many attractive features, but
the trade-offs are such that abandonment of this process in favor of oxi-
dation processes is gaining favor.  Table I presents the advantages  and
disadvantages of anaerobic digestion.

          The wide variety of natural substances of biological origin that
are insoluble in water and soluble in such non-polar solvents as ether,
chloroform, and benzene are classified by the  technologist as "lipids" or
"lipides," which is derived from the Greek HTTOS for fat.  "Grease"  is a
classification that is designated by the feel  of the substance and the
ability of the material to make transluscent paper transparent.  The term
"fat" implies a biological source, but a chemist would restrict the term
to describe the variety of liquids and solids  that are formed naturally
by the esterification of long chain carboxylic acids and glycerol.

          Loehr and Kukar=-' have shown that the complete spectrum of natural
lipids is to be found in wastewater.  Hunter and Heukelekiaoi' place the
quantity of grease in Highland Park, N. J., sewage at 25%.  A variety of
industrial wastes, especially from packing houses, dairies, and canning
plants, would increase the level of fat, while petroleum products, espe-
cially automotive motor drain oil and shop floor scrubbings, would con-
tribute a nonglyceride fraction.

-------
                                 TABLE I

               ADVANTAGES AND DISADVANTAGES IN ANAEROBIC
                  DIGESTION OF MUNICIPAL WASTE SLUDGES
         Advantages

Approximately 2/3 of the BOD
  in wastewater stream can be
  processed without cost of
  aeration

Can process hydrophilic sludges
  of low solids content

Digested sludge more filterable
  than fresh sludge

Low ORP inhibits corrosion

Some aerobically bioresistant
  materials are digested

Produced methane can provide
  power for plant
          Disadvantages

Equipment is more expensive than
  aeration equipment

Malodor from H2S and other
  anaerobic metabolities

Process variables of pH, feed rate,
  N&P nutrition, temperature and
  inorganic concentration must be
  carefully controlled

Air-oxidized H2S is highly cor-
  rosive

Difficult to initiate fermentation

Grease is concentrated in digesters
  as scum layers

Oxygen is lethal to process

Produces a strong supernatant liquor
  which constitutes a heavy organic
  and nutrient loading to main bio-
  logical treatment process
          A biochemical analysis of lipid digestion anticipates two sources
of difficulty.  First, normally occurring wild bacteria will not be able
to metabolize certain synthetic molecular structures and some petroleum
derivatives.  There is no definitive study in the sanitary engineering
literature on the relative rates of degradation of the various molecular
species of hydrocarbons although Davis^t/ presents sufficient evidence to
establish that different rates do indeed exist.  Some of Davis' own re-
search^.' indicates that the normal paraffins are more rapidly oxidized by
some aerobic microorganisms.  Beerstecher,2/ on the other hand, states
that the branched chain hydrocarbons are more susceptible to oxidation.

-------
          Jeris and McCartyZ/ have shown that microbial mechanisms  are
available in a methane fermenter to convert palmitic acid and water to
methane and carbon dioxide.  The same order of thermodynamic potential
exists for decane:

     CH3 - (CH2)s - CH3 + 6H20	> 7CH4 + 3C02 (AF - -37 kcal/mole)


          Davisi' cites an abundance of evidence that the sulfate-reducing
bacteria can oxidize hydrocarbons.  Anaerobic breakdown of hydrocarbons by
other bacteria has not been investigated, except for a report by Chouteau
et al.—  that Pseudomonas aeruginosa could dehydrogenate n-heptane.

          The second criterion for lipid digestion is the actual contact of
the microorganism and the surface of the lipid.  While in the metabolism
of soluble materials, the rate of nutrient utilization is a function of the
concentration of nutrient, the rate of metabolism in a heterogeneous sys-
tem of lipid-water-bacteria is related to the actual interfacial area rather
than bulk amounts of material.  Desnuelle—' has shown that surface area is
rate-limiting in the lipase-catalyzed hydrolysis of triglycerides.  This
concept is easily extrapolated to the problem of scum-layer formation in
a sludge digester.  Lipids, even though degradable, separate and float to
the top of the digester tank, forming a scum layer.  Once this material has
separated, it is unavailable as a medium for bacterial growth unless me-
chanical energy is introduced to redisperse the lipids in the water.
McKinneylx' has recognized surface area as a possible rate-limiting factor
and suggested a high-speed dispersing nozzle in the sludge recycle system
as a solution to the grease scum problem.

          The research described in this report was designed to evaluate
this redispersal of scum layers as the rate-limiting process in digestion
of a typical lipid that is recognized as biodegradable.  The usual anaerobic
digestion system was modified to permit high-shear homogenization of any
lipids that separated as a supernatant layer.  The protocol for the experi-
ments was designed to answer the following questions:

          1.  Does high-shear homogenization have any observable effect on
an established digestion pattern where lipids are present only in very
small concentrations?

          2.  Does high-shear homogenization have any effect on the diges-
tion process when high concentrations of a biodegradable lipid (cottonseed
oil) are present?

-------
          3.  Does the presence of a non-biodegradable lipid (motor oil)
influence the effect of high-shear homogenization on the biodegradation of
a biodegradable lipid (cottonseed oil)?

          The above questions have been answered by the following series
of experiments:

          Experiment I - Performance of 15 replicate laboratory digesters
on a standard loading and mixing regimen of Purina Dog Chow.  (Question 1)

          Experiment II - Effect of high-shear homogenization on the diges-
tion of Purina Dog Chow.  (Question 1)

          Experiment III - Effect of homogenization on limed digesters
heavily loaded with cottonseed oil.   (Question 2)

          Experiment IV - Effect of homogenization on soda-ash-buffered
digesters when dog food loading is augmented with cottonseed oil.  (Ques-
tion 2)

          Experiment V - Effect of homogenization on soda-ash-buffered
digesters when cottonseed oil gradually replaces Purina Dog Chow as a
feed.  (Question 2)

          Experiment VI - Effect of motor oil on the digestion of Purina
Dog Chow and cottonseed oil.  (Question 3)

          The data from these studies also have been analyzed for any
other possible information that might be useful for improving the anaerobic
digestion of wastes.

-------
                              EXPERIMENT I

                 PERFORMANCE OF DIGESTERS ON STANDARD
                      LOADING AND MIXING REGIMEN
          The data for the baseline studies must be derived  from several
different time periods.  The overall experimental design was such that the
digesters were "fed" to a level that produced failure.  Thus, after each
experiment  that involved an overload "feed," there was a period of re-
adaptation  or re-acclimation to establish the bacterial population.  This
was  followed by a period of several days' operation at a constant feed
level.

          Initially the digesters were  filled to the  1 gal.  (2,785 ml)
level.  The daily drawdown was 175 ml;  the detention  time, 21.6 days.  The
digesters were inoculated with screened digester sludge and  fed at a rate
of 1.32 g/^/day of dog food for 17 days.  The feed level was then increased
incrementally to 3.96 g/^/day.  During  the period 10-9-68 to 10-15-68, all
15 digesters were stirred by head gas recirculation only.  This period can
be compared to similar "blank" periods  of 11-5-68 to  11-11-68, 11-21-68 to
12-5-68, and 3-9-69 to 3-14-69.  The average gas production  of the 15 di-
gesters during the period 10-9-68 to 10-15-68 is plotted as Curve B in
Figure 2, p. 10.  This curve is quite linear.

          The data from each of the four "blank" periods are presented in
Table II.  The most significant information to be derived from the data are
the variability of the digesters in each period.  This is presented as the
standard deviation divided by the average.  Thus the  four periods can be
compared.  The value of 31.47. for the standard deviation for the period
10-9-68 to  10-15-68 is well within the average for microbiological systems
while the other three sets of data show that the variability during these
"blank" periods was very low for such a system.Ii/  One approach to the
overall variability of the system is to average the four values of the
standard deviations after weighting for the time period.  This overall
standard deviation was 12.47».  During the entire period of 10-24-68 to
1-7-69, the average gas production was 277 I with a standard deviation of
only 6.67..  The literature does not have comparable values because most
previous studies with anaerobic digesters have been performed with a single
unit.

          During the nine-day period, 11-12-68 to 11-20-68,  inclusive, the
digesters were operated without head gas recirculation or liquid recircula-
tion.  The digesters were stirred only by occasional shaking.  During this
period, the 15 digesters produced an average of 33.75 SL of gas from a
loading rate of 3.96 g/je/day.   The gas yield was 0.250 4/g feed, but the

-------
                                TABLE II
             PERFORMANCE OF DIGESTERS DURING "BIANK" PERIODS
                             Cumulative Gas Production in Liters
Digester

   1A
   IB
   1C
   2A
   2B
   2C
   3A
   3B
   3C
   4A
   4B
   4C
   5A
   5B
   5C

Average

Standard deviation
  as % of average

Feed rate g/4/day

Gas yield //g feed

Average methane con-
  tent as %
10-9-68 to
10-15-68
(6 days)
17.55
30.70
22.74
29.76
30.73
45.45
22.82
15.93
19.11
30.00
39.14
30.01
45.51
3 9. .64
28.56
11-5-68 to
11-11-68
(6 days)
30.52
30.89
29.72
29.81
30.63
30.82
25.30
28.86
30.10
28.14
27.02
28.34
23.36
26.99
28.14
11-21-68 to
12-5-68
(15 days)
66.54
65.07
67.32
70.09
72.05
73.11
58.94
70.48
67.37
57.73
62.64
65.29
52.44
63.19
63.68
3-9-69 to
3-14-69
(5 days)
16.88
17.34
19.42
16.51
18.58
19.73
17.58
17.71
17.94
17.85
16.97
15.79
16.22
16.29
18.27
29.85
28.58
65.06
17.54
31.4
3.96
0.331
7.7
3.96
0.318
8.6
3.96
0.289
6.6
5.56
0.292
              54
              54
               49

-------
nine-day period includes one day during which the bath temperature dropped
to 20°.C.  The decrease in gas yield of this period when compared with the
preceding and succeeding periods does not appear to be significant.

          The baseline level of feed was to have been established at 75%
of the feed level that would cause one-third of the digesters to malfunc-
tion either by becoming sour or passing through the feedstock undigested.
It was not possible to produce biochemical malfunction as the feed level
of dog food could be raised to only 20 g/gal. before the liquid pumping
system became clogged.  This is comparable to a loading of 0.28 Ib volatile
solid/ft3/day (4.45
          The data from this build-up period, 12-9-68 to 12-16-68, are
presented in Figure 1.  When computer based curve fitting techniques were
applied to these data by Dr. James J. Downs, a linear equation,

          GAS PRODUCTION = -0.032 + 0.072 FEED RATE

was obtained that accounted for 52% of the variance in the data.  The in-
clusion of the temperature data gave the equation:

          GAS PRODUCTION = -23.3 + 0.023 FEED RATE +4.1 In T°K

that accounted for 78% of the variance.  The negative constants are dis-
turbing until it is realized that these data represent only a small portion
of the full range of possible relationships that would have the shape of
a typical growth curve.  This analysis of these data leads to the inference
that digestion rate per unit waste load would increase with loading rate
at some levels of loading.

-------
   37
I  35
   34
0.33

>
» 6 0.31
X
u
a£
«/> f\
go
°2 0.29
»i
85 Ik
5*
i °-27
o

n 
-------
                              EXPERIMENT II

                 EFFECT OF HIGH-SHEAR HOMOGENIZATION ON
                    THE DIGESTION OF PURINA. DOG CHOW
          The digesters were operated during the seven-day period, 10-16-68
to 10-23-68, with a constant dog food loading.  The 15 digesters were di-
vided into five groups of three:*

          Group 1 - gas-stirred only;

          Group 2 - Gas plus liquid stirring with spring-loaded ball
                    valve operating at 50 psig;

          Group 3 - Gas plus liquid stirring plus homogenization with
                    relief valve set at 100 psig;

          Group 4 - Gas plus liquid stirring plus homogenization with
                    relief valve set at 300 psig; and

          Group 5 - Gas plus liquid stirring with relief valve set at
                    900 psig.

          The digesters were filled to 3,785 ml and were maintained on a
daily feed of 15 g (3.96 g/f/day), 0.23 Ib volatile solids/ft3/day.  The
detention time was 21.6 days.  The pH was determined each day before
feeding and the system  was buffered with lime (Ca(OH)2), and sodium
bicarbonate (NaHC03) so that 1 g of lime was added with the feed for each
0.10 pH unit below 7.00 to a maximum of 5 g and 0.1 g of NaHC03 was added
with the feed for each 0.10 pH unit below 6.50.  Head gas was recycled at
the rate of 100 ml/min.  Liquid recirculation rate was 100 ml/hr.

          The results of the experiment are presented in Figure 2.  There
is a greater variability in this group of data than in the data from the
later studies.  Part of this is the result of averaging three rather than
five data.  There appears, however, to be no significant effect in the
digestion process due to the homogenization process.  Group 2 had the
greatest rate of gas production of the five groups but the rate is only a
few percent greater than the previous six-day "blank" period.  Group 3 is
definitely lower in the early part of the study, but the line joining the
data of the last four days has the same slope as lines 1, 4, and 5.
   In later experiments, the digesters were divided into three groups of
     five.  The coding of the digesters, however, as 1A, IB, 1C, 2A, 2B,
     etc., was maintained throughout the entire research and has been kept
     for this report.

-------
          LOADING RATE » 3.96 g/fl/day PURINA DOG CHOW
    35
1 = GAS STIRRING ONLY
2= GAS STIRRING PLUS LOW PRESSURE LIQUID RECIRCULATION
3 = GAS STIRRING PLUS LIQUID HOMOGENIZATION AT  100 LB/IN2
4= GAS STIRRING PLUS LIQUID HOMOGENIZATION AT 300 LB/IN2
5= GAS STIRRING PLUS LIQUID HOMOGENIZATION AT  900 LB/IN2
B = PREVIOUS SIX DAYS AVERAGE OF FIFTEEN DIGESTERS
     STIRRED BY HEAD GAS RECIRCULATION  ONLY
    30
(/>
oc
I   25
Z
O
8
    20
t/>
3   15
!S!
U   10
                                                                                 J
                                                                                 7
                             3          4
                                 DAYS
                    Figure 2 - Effect of Homogenization on Digestion
                                        10

-------
The experiment was discontinued after the seventh day because of foaming
problems.  The problem of foaming plagued the entire research program.
This was the only pronounced effect on the digestion process due to high
shear homogenization when lipid levels were low.
                                     11

-------
                              EXPERIMENT III

                EFFECT OF HOMOGENIZATION ON LIMED DIGESTERS
                    HEAVILY LOADED WITH COTTONSEED OIL
           After Experiment II,  the group of 15  digesters  was  divided  into
 three groups of five:

           Group A - Gas-stirred only;

           Group B - Gas-  and  liquid-stirred with  spring-loaded ball valve
                     operating at 50 psig;  and

           Group C - Gas-  and  liquid-stirred plus  homogenization with  relief
                     valve set at 300 psig.

           Prior to  Experiment III,  the  digesters  were all functioning
 satisfactorily  except  for plugging of the  diaphragm pumps when the feed
 level was  raised to 20  g/gal./day.   The digester  contents had been pooled
 and  redistributed.   This  slight exposure to the atmosphere had caused
 problems in  the past,  so  the  digesters  were brought up  to a feed  level of
 3.96  g/i/day dog food  over a  period of  a few days.   The digesters were all
 performing satisfactorily and uniformly before  the  addition of cottonseed
 oil,  as can  be  seen from  the  data  presented in Table III.
                                TABLE  III

         PERFORMANCE OF DIGESTERS PRIOR TO ADDING COTTONSEED OIL
   Date

26 December
27 December
28 December
29 December
30 December
31 December
 1 January
 2 January
                 Average Cumulative Gas
                  Production in Liters
Group A
3.84
4.32
6.42
8.52
11.75
15.70
21.00
25.60
Group B
4.21
5.36
7.73
9.63
12.02
14.81
18.75
22.30
Group C
4.18
4.74
7.24
9.29
12.39
15.38
18.95
24.32
           Remarks

3.96 g/£/day Purina Dog Chow
Digester contents pooled
Raw sludge only added
1.32 g/^/day Purina Dog Chow
2.64 g/^/day Purina Dog Chow
3.96 g/^/day Purina Dog Chow
3.96'g/jg/day Purina Dog Chow
3.96 g/4/day Purina Dog Chow
          In Experiment III, the digester volume was 3,785 ml; the deten-
tion time, 21.6 days.  Head gas was recycled at the rate of 100 tnl/min;
liquid recirculation was at the rate of 100 ml/hr.  Lime was added with the
feed as a buffer to each individual digester at the rate of 1 g for each
                                     12

-------
0.10 pH unit below 7 until 1-11-69, after which 0.1 g of NaHCOa was added
with the feed for each 0.10 pH unit below 7.  The loading protocol (see
Table IV) was such that the dog food was gradually replaced by cottonseed
oil based on the approximate COD equivalence of 1 ml oil = 3 g dog food.
                                TABLE IV
               PROTOCOL FOR DIGESTER LOADING IN EXPERIMENT III
   Date

 3 January
 4 January
 5 January
 6 January
 7 January
 8 January
 9 January
10 January
11 January
12 January
13 January
14 January
15 January
16 January
17 January
  Dog Food
(g COD/A/day)

    3.73
    0.00
    2
    2.
    3
    3
    3
  49
  49
  73
  73
  73
    2.99
    2.24
    2.24
     .24
     .49
     .49
     .49
2.
1.
1.
1.
Cottonseed Oil
(g COD/l/day)

    0.70
    0.70
    0.70
    0.70
    0.70
    0.70
      .40
      .10
      .80
      .80
      .80
      .50
      .50
                                          Total
                                      (g COD/A/day)
    1.49
1.
2,
2,
2,
2.
3.
3.
3.50
3.50
4.43
0.70
  ,19
  ,19
4.43
4.43
6.13
  .09
  .04
  .04
  ,04
  ,99
  .99
  .99
                     3.
                     3.
5.
5.
5.
5,
4.
4.
4.
                                          4.99
          The data on gas production are presented in Figure 3.  These data,
as well  as analytical data  presented in the Appendix,  show'that  the three
groups of digesters  can be  distinguished on every basis  of comparison.
Furthermore, Group A, which was  the standard  to  which  the  other  groups
were  to  be compared, showed a  unique behavior.

          The following significant phenomena were observed during  the
15-day period:

          1.  Group A produced more gas than  Group B,  which produced more
gas than Group  C.

          2.  The methane content  of gas produced by Group A was higher
than  that produced by Group C  as seen  in Table V:
                                    13

-------
100


 90
to
Q
O
    70
    60
    50
a.
uo  40
O
 30


 20


 10


  0
               A = GAS STIRRED ONLY
               B = GAS +  LIQUID STIRRED (50 PSIG)                                           C

               C= GAS +  LIQUID STIRRED + HOMOGENIZATION  (300 PSIG)                 /
I      I      I  	I
                                                                      I      I     I      I
I
                             6     78     9     10    11     12     13    14    15    16    17
                                           JANUARY
         Figure 3 - Performance of Digesters Fed Dog Food and Cottonseed Oil Buffered with Ca(OH)2

-------
                                 TABLE  V

                     PERCENT METHANE IN DIGESTER GAS

                                 1-10-69              1-17-69

               Group A            57.4                  63.4

               Group B            41.8                  50.8

               Group C            39.8                  40.2
          3.   The data from the neutral lipid extraction, when examined
for a single digester over a period of time, appear to reflect an adaptation
of the bacterial population to the added cottonseed oil.   There also seems
to be a cyclic phenomenon in the individual lipid levels.  When examined
in relation to time, the average lipid levels in the various groups that
are presented in Table VI imply the development of a bacterial flora more
capable of hydrolyzing the neutral oil.  In the case of Group C, however,
the initial step of the breakdown of the oil to gas is too rapid, leading
to a build-up of acid intermediates.  Methane digestion is inhibited by
the pH drop, and the slight increase in lipid levels of Group C that occurs
between 1-13-69 and 1-16-69 may also be the result of pH drop affecting the
hydrolyzing microorganisms.
                                TABLE VI

              AVERAGE LIPID LEVEL IN EXPERIMENT III (g/I)

               Group A        Group B        Group C       Total Added*

                 1.21           0.81           0.76             2.91

                 0.74           0.52           0.62             7.40

                 0.59           0.40           0.96            11.36
   Total added cottonseed oil from 1-3-69 to given date in grains per liter.
                                     15

-------
          The soluble COD and MLVSS data also confirm the premise of dif-
ferential adaptation.  The MLVSS, which would measure both undigested dog
food and bacterial mass, decreased as the dog food load decreased, but
also decreased with the amount of shear.  Soluble COD in Group A decreased
from 1-7-69 to 1-14-69 as the dog food load was decreased but Group C
maintained the same level.  The decrease of MLVSS in the Group C digesters
could be the consequence of the formation of calcium soaps, these coating
the dog food particles and making them coalesce at the bottom of the
digester.

          The malfunctioning of the digesters in Groups B and C can be
discerned as early as 1-8-69 when the pH values are examined.  All pH's
were above 6.80, but Group A had values all above 7.00.  Since lime was
administered to each individual digester at the rate of 1.0 g of Ca(OH>2
for each pH unit below 7.0, the Group C digesters were most heavily limed.
The lime maintained the pH in the normal operating range until 1-14-69
when all five digesters in this group had samples with pH readings 6.75
and below.
                                    16

-------
                              EXPERIMENT IV
      EFFECT OF HOMOGENIZATION ON SODA.-ASH BUFFERED DIGESTERS WHEN
           DOG FOOD LOADING IS AUGMENTED WITH COTTONSEED OIL
          Prior to beginning this experiment, the liquid volume in all the
digesters was reduced to 2,160 ml to allow the foam room for expansion.
The detention time remained at 21.6 days.  Head gas was recycled at the
rate of 100 ml/min and liquid was recycled at the rate of 50 ml/hr.  The
contents of the Group A digesters were pooled and diluted with fresh sludge
and this mixture was distributed evenly among all 15 digesters.

          The digesters were then brought incrementally up to a daily dog
food feed level of 5.55 g/jfc (0.32 lb/ft3).  Soda ash, Na2C03, was used to
buffer at the rate of 0.2 g added with the feed to each digester for each
0.10 pH and below 7.00.

          During the break-in period from 2-6-69 to 2-12-69, the 15 digesters
produced 29.63 a of gas with standard deviation = 2.42 or 11.57. of the mean.
The dog food level was then maintained at 5.55 g/£/day, while an additional
1 ml of cottonseed oil was added from 2-14-69 to 2-17-69 and 2 ml was added
from 2-18-69 to 2-20-69.  At the end of the experiment, the three groups
of digesters could not be differentiated on the basis of cumulative gas
production.  These cumulative gas values, obtained with all pumps operating,
were 23.71 I, 23.98 it and 24.68 I, respectively, for Groups A, B, and C.
These data do not differentiate the groups.  When daily gas production is
plotted, as in Figure 4, some slight differences in the three groups may
be discerned.

          Table VII shows the results of lipid analyses during the period
of Experiment IV.  As in the case of Experiment III, where lime was used
to buffer, the lipid level appears  to be inversely related to the amount of
shear.
                                TABLE VII

               AVERAGE LIPID LEVEL IN EXPERIMENT VI
 Date

2-10-69
2-13-69
2-17-69
2-20-69
2-24-69
Group A

 1.12
 0.89
 2.15
 2.65
 1.46
Group B

 0.81
 0.75
 1.60
 1.83
 1.22
Group C

 0.69
 0.56
 1.22
 1.84
 0.705
Total Added*

    0.00
    0.00
    1.69
    4.22
    4.22
*  Total added cottonseed oil from 2-10-69 to given date.
                                    17

-------
   5.00
   4.00
 . 3.00
z
o
O
  2.00
  1.00
                                               O  • GAS STIRRING ONLY

                                               A  • GAS + LIQUID STIRRING (50 PSIG)


                                               D  • GAS + LIQUID STIRRING +

                                                       HOMOGENIZATION (300 PSIG)
                                      J0.844g/f/doy
                 [0.422 g/0/doy cottonseed oil feed
                 5.55 g/J/doy Purina Dog Chow
             T3    U1516    17    18    1?"
20
                        FEBRUARY


   Figure 4 - Digester Performance with Soda-Ash Buffering


                                       18

-------
          However, none of the three groups of digesters have adapted well
to the slugs of cottonseed oil.  The premise that the functioning of the
hydrolyzing bacteria is affected by pH as well as the methanogenic bacteria
is reinforced here.  The Group A digesters, which initially appear to re-
spond to the added oil, were the most rapid to fail.  This can be seen in
the volume of gas produced as shown in Figure 4 and in the methane content
of the gas as well.  On 2-14-69, all three groups of digesters were produc-
ing 53% methane while by 2-21-69 the average methane content of the gas
produced by the three groups  was 47, 51, and 527,, for Groups A, B, and C,
respectively.  The results of this experiment indicate that homogenization
had a slight positive action in overcoming the malfunction caused by heavy
oil loading accompanied by soda-ash buffering.
                                    19

-------
                               EXPERIMENT  V

        EFFECT  OF  HOMOGENIZATION ON SODA-ASH  BUFFERED DIGESTERS AS
                     COTTONSEED OIL GRADUALLY REPLACES
                         PURINA DOG CHOW AS A FEED
          This  experiment was  similar  to  Experiment  IV except that the
 level  of  dog  food  loading was  decreased while  the  level of cottonseed oil
 was  increased.  The volume  of  the  liquid  in  the digesters was 2,160 ml;
 the  detention time, 21.6 days.   Soda ash  was used  to buffer  the individual
 digesters at  the rate  of 0.2 g added with the  feed for each  0.10 pH unit
 below  7.00.   Protocol  for loading  as well as gas production  and pH data are
 presented in  Table VIII.

          The major conclusion of  this experiment  is that heavy loads of
 cottonseed oil  are deleterious  to  the  performance  of soda-ash buffered
 digesters.  All three  groups of digesters showed malfunctions as the per-
 centage of the  COD load that was due to cottonseed oil was increased.  The
 three  groups  of digesters showed some  differences  in performance and, if
 rated  by  the  usual criteria, the digesters subjected to gas  and liquid
 stirring  (Group B) gave a slightly better performance than the other two
 groups.   The  slight difference  (4%) in the cumulative gas production be-
 tween  Groups  B  and C is probably not significant.  If the difference in
 performance of  the liquid stirred digesters was due  to shear, it appears
 that the  spring-loaded ball valve, operating at 50 psig, could provide
 adequate  shear, if, indeed,  the  observed differences  were due to shear.

          The pH data  from  this experiment are presented in  Figure 5.
Groups B  and  C  appear  to sour a day later than Group A.  The MLVSS data
 also appear to  differentiate Group A from the other  two groups as can be
 seen in Table IX.  The soluble  COD values  for 4-1-69 that were 24.2, 17.0,
 and 15.8  g/4  for Groups A,  B, and C, respectively, also appear to dif-
 ferentiate.

                                TABLE IX

                 MIXED LIQUOR VOLATILE SUSPENDED SOLIDS
                          IN EXPERIMENT V  (g/4)

 Date               Group A             Group B             Group C

3-18-69              22.1                23.1                17.8
3-25-69              14.2                 16.2                12.5
4-1-69               16.0                 16.1                13.5
                                    20

-------
N)
                                                    TABLE VIII

                            EFFECT OF HOMOGENIZATION ON THE DIGESTION  OF  COTTONSEED OIL
WHEN FEED RATE IS KEPT CONSTANT IN COD
Feed Rate in Grams
(COD/ i/day)
Date Purina Dog Chow
3-15-69
3-16-69
3-17-69
3-18-69
3-19-69
3-20-69
3-21-69
3-22-69
3-23-69
3-24-69
3-25-69
3-26-69
3-27-69
3-28-69
3-29-69
3-30-69
3-31-69
4-1-69
4-2-69
4-3-69
5.23
5.23
5.23
5.23
5.23
4.58
4.58
4.58
3.94
3.94
3.94
3.29
3.29
3.29
2.62
2.62
2.62
3.29
3.29
3.29
Cumulative Gas Production
in Liters
Cottonseed Oil Group A
0
0
0
0
0
0.61
0.61
0.61
1.22
1.22
1.22
1.83
1.83
1.83
2.44
2.44
2.44
3.05
3.05
3.05
3.73
8.14
11.45
15.93
17.93
21.06
23.85
27.23
31.17
33.14
35.80
37.97
39.92
42.01
43.24
45.08
46.11
47.48
48.31
49.16
Group B
3.80
8.16
11.95
16.14
19.75
23.11
26.30
29.78
33.87
36.23
39.40
42.02
44.50
46.91
48.23
49.90
51.62
53.49
54.41
55.90
Group C
4.19
8.49
12.08
15.33
19.25
22.58
25.67
29.09
32.98
34.83
37.86
40.39
42.52
44.89
45.86
47.74
49.30
51.12
52.24
54.07
Average
pH in 15
Digesters
6.95
6.97
7.06
6.93
6.93
6.91
6.87
6.97
6.96
6.92
6.84
6.85
6.80
6.80
6.66
6.57
6.51
6.47
6.46
6.36
pH Range in
14 Digesters*
6.80-7.30
6.85-7.30
6.95-7.25
6.85-7.20
6.85-7.25
6.70-7.15
6.70-7.10
6.85-7.05
6.90-7.10
6.80-7.15
6.65-7.10
6.70-7.05
6.60-7.00
6.60-7.05
6.50-6.80
6.40-6.90
6.25-7.05
6.30-6.90
6.30-6.90
6.30-6.75
         Digester 1A was malfunctioning during most of this study, and its pH has not been included in the
           range.

-------
                                  GROUP A - GAS STIRRING ONLY
      4


      7


   6  8|


   7


   8  6


      7



   6  8


   7


   8  «



      7
         I    I    I    I    I    I    I
                                                    I    I    I
                                                                           J	!
           j	  I    i
                                                                              j	i
                                                                                    IA
                                                                                    1C
                                                                                    2A
                                GROUP B - GAS + LIQUID STIRRING (50 PSIG)
   6


   7


6  e


7


8  6


   7


6  8


7


8  6
           J _ I _ I..  I _ I _ I  .  I _ I _ I _ I _ I _ I _ I - 1 - 1 _ I _ I _ I _ L
                       _ _  .  _    _
                       GROUP C - GAS + LIQUID STIRRING + HOMOGENIZATION (300 PSIG)
                                                                                       K
                                                                                       3A
                                                                                       3.
                                                                                      3C
           iiriiiTiii   ITIIII    i   i    r
           I    I    I    I    I    I    I    1    1   I   1   I    I    1    I    I    1    I    I
                                                                                       5A
                                                                                       5C
        15  14   17   18   19   20   21   22  23  24  25   26   27   »   »  30   31   I    2   3   4
        MARCH
                                                                           APRIL
Figure 5  - Variation of pH in Individual Digesters  During  Experiment  V
                                              22

-------
                              EXPERIMENT VI

               THE EFFECT OF MOTOR OIL ON THE DIGESTION OF
                   PURINA DOG CHCW AND COTTONSEED OIL
          The evaluation of the effect of a non-biodegradable lipid on the
digestion process was carried out with the digesters filled with 2,160 ml
of mixed liquor.  The detention time was 21.6 days.  There were some dif-
ficulties with the break-in period and the Group C digesters were not yet
in full balance when the experiment began.  The loading protocol is shown on
Figure 6 along with the performance data.  It is apparent that motor oil
had little effect on the digestion of the dog food and cottonseed oil, as
neither the gas production rate, the methane content of the gas, nor the
pH was affected by its addition.

          Since the poorer performance of Group C digesters could have been
due to the action of the homogenizers, the liquid pumps were turned off
for three days to evaluate this idea.  Gas-volume data from each of the
three groups of five digesters are averaged and presented below as ratios
of the averages with Group A averages  the common denominators for all
three groups of ratios.
          Condition

               1
               2
               3
               4
Group A

  1.00
  1.00
  1.00
  1.00
Group B

  1.06
  0.99
  1.00
  0.97
Group C

  0.68
  0.81
  0.89
  0.70
where Group A = gas recirculation only.
            B = gas -I- liquid recirculation  (50 psig) .
            C = gas + liquid recirculation + homogenization  (300 psig).

          Condition 1 existed on the fifth day (5-4-69) after operation
at a daily feed rate of 9 g of dog food and 1 ml of cottonseed oil with
both liquid and gas pumps operating.

          Condition 2 existed on the fourth day  (5-9-69) after 1/2 ml of
motor oil was fed daily in addition to the dog food and cottonseed oil.
All pumps were operating.
                                     23

-------
         F
         E
         E
         D
                                                                      GAS STIRRING ONLY
    I   0.462 g/ fl cottonseed oi I
                               1  0.231 g/1 motor oil
9/0 Purina Dog Chow
                                                      / /
                                          /  / /  Y 7 7~
                                                      / /
                                                                       /  7  7 / / /
   5  '
                                               GROUP A = GAS STIRRING ONLY
   §3
8  5
   o
                           GROUP B = GAS + LIQUID STIRRING (50 PSIG)
                                                     I
3

2

1

0
                  I
      I
I
GROUP C = GAS +  LIQUID STIRRING + HOMOGENIZATION
I      I     I      I     I      I     I   I   I     I      I
                                                                                        (300 PSIG)
            29    30
           APRIL
      1     2
      MAY
     3456     7     8    9     10    11    12    13
     Figure 6 - Effect  of Motor Oil on Lipid Digestion
                                                            14
                                                                                               15
16

-------
          Condition 3 existed at the end of three days (5-12-69) with only
the gas pumps operating.  The feed rate was the same as in condition 2.

          Condition 4 existed after the liquid recirculation pumps had
been turned back on and had been operating for four more days (5-16-69).
The feed rate was the same as in condition 2.

          The results of the above analysis seem to again implicate high
shear homogenization as a negative factor in lipid digestion.
                                    25

-------
                               DISCUSSION
The Effect of Shear on  the  Digestion of Lipids

          The results of  the experimental program described in this report
present a strong caution  against  the use of high-shear homogenization as a
remedy for scum layer formation in  full-scale anaerobic digestion.  Ex-
periment III produced data  to show  that lipid digestion in lime-buffered
digesters can be slowed by  a low-shear homogenization, while lipid digestion
is upset by high-shear homogenization equivalent to  that produced by a
Waring Blendor.  Experiments IV and V present data that show a slight
process improvement due to  high-shear homogenization when soda ash is used
as a buffer in a system heavily loaded with lipid.  This reversal, however,
occurs when the digesters were malfunctioning for another reason.

          The anaerobic digestion process like all biological processes is
described by the term "steady-state dynamics."  The hydraulic analogy is a
watershed in which flow into a portion of the system is matched by flow
out.  If flows are not matched, the result is either flooding or draining
of an impoundment.  A simplified description of the digestion process is
                                 SLUDGE

                        (Insoluble Organic Matter)

                            R!     Extracellular Enzymes

                         (Soluble Organic Matter)

                            R2     Acid Bacteria
                             Volatile Acids
                            R.
                             •3
Methanogenic Bacteria
                               CH4 + C02
          In a well-operating digester, the rates of reaction in units of
material altered per unit time must be equal, R^ * R£ • R^.  The ability
of the organisms to multiply and adapt to increased food levels, however,

varies greatly, with —ij4 —2. jt —2. .   All of the technical data available
                     dt   dt    dt

                                    26

-------
indicate  that -^. > —2. .   It appears that in the case of grease, —-- >
               dt    dt                                           dt
dR2 > dR3  when adequate surface is created between the grease and the
dt    dt
digester liquor.  The soluble organic matter,  in this case fatty acid salts
of calcium or sodium or soaps, can have two modes of actions.   It can ac-
celerate R2 so that the volatile acids build up, lower the pH, and  inhibit
methane fermentation.  The  result is a "stuck" digester.  The  other  action
of the soaps is to directly inhibit methane fermentation by affecting the
bacterial surface.  This, too, would lead to a "stuck" digester.

          Soda ash, which supplanted lime as a buffer for the  purpose of
reducing foaming, proved harmful because it, also, accelerated hydrolysis
too much.  The data are consistent with, but do not prove, the following
hypothesis:  Sodium soaps (sodium salts of long chain fatty acids)  would
inhibit methane fermentation. Sodium soaps are much more soluble than
calcium soaps, but they are still very sparingly soluble.  When sodium
soaps reach a certain level of concentration in water, the soaps form
emulsions.  Most of the soap  is present as the oil phase of an emulsion,
a portion acting as its own emulsifying agent.  High-shear now functions
to provide interface between  the soap micelle and bacteria that degrade
the soap to simpler metabolites that can be fermented to methane.  McCartylr.'
has suggested the use of calcium chloride  to reverse  sodium oleate poisoning
in the same series of articles that advocate the use  of sodium bicarbonate.

          The results of the  study, nonetheless, should not be  interpreted
as condemning the incorporation of  surface mixers  in  the design of digestion
tanks.  There is a report—'  of the installation of  four 10 hp Lightning
mixers on a 100,000  ft3  tank  in the town of Tonawanda, New York, with the
very rewarding  result of eliminating  scum  formation.  Since most of  the
scum from  the settling  tanks  is burned,  this tank  is  not subjected to a
heavy grease  load.  The  results of  this  research  led  to  the prediction that
the operator  of the Tonawanda Plant would  be in  serious  trouble should the
situation  arise when the grease burners  are inoperative at the  same  time  a
heavy slug of grease is  fed  to  the  digesters as  a  shock  load.  While me-
chanical mixing in addition  to  that caused by rising  gas bubbles is  es-
sential  for high-rate digester  performance, the  hydrolysis of heavy  grease
loads could proceed  too  rapidly as  a  result of  creating  grease-digester
liquor  interface.  In  such a  situation there would be a  build-up of  soaps,
either  calcium  soaps from  lime, or sodium  soaps  from soda  ash,  that  would
affect  the methane bacteria  and cause digester  upset.  Foam production,
which  is  especially  severe when  lime  is  used  as  a buffering agent, may also
negate  any process advantage  due  to high-shear  mixing.
                                     27

-------
Ability  of  Digesters  to  Degrade Heavy Loads  of Grease

          It  is  reported by  BasuM/  that heavy grease  loads cannot be di-
gested.   In Experiment III,  during  the  last  three days of fermentation,
the Group A digesters were loaded with  a feed stock that was at least 43%
fat by weight  and  70% fat on the basis  of COD.  This feed produced the most
rapid gasification and the highest methane content gas of any sample of
feed stock  used  in the study.  High  shear that was equivalent to that of
a Waring Blendor eventually  caused a matched group of digesters, Group C,
to go sour.   Even  the lower  level emulsification of the Group B digesters
caused malfunction.   The level of gas stirring in Group A, however, was
such that about  two volumes  of gas equal to  the liquid volume in the di-
gester were pumped to the bottom of  the digester each hour.  This type of
stirring cannot  be quantitated, but  the digester contents were undergoing
a lively bubbling  due to gas recirculation.
Optimum Operating  Conditions  for Sludge Digesters

          Loading  rate:  Manual of Practice  16—' differentiates "high
rate" as opposed to "standard rate" digestion on the basis of loading,
0.15 to 0.40  Ib volatile solids/ft3/day compared to 0.04 to 0.10 Ib. vola-
tile solids/ft^/day.  The article goes on, however, to consider mixing as
an essential  feature of the high rate system.  In this study, the digesters
were able to accommodate 0.29 Ib volatile solids/ft-^/day of the highly
putrescible dog food.  Since the experimental design was such that most of
the time constant  feed levels were maintained, not much kinetic data can
be derived from the experiments to establish a digestion rate.  Still,
during the periods of startup, the digesters responded in one day to any
load increase that was not an overload.  The overload rate was not reached
by slow buildup at any time, so a maximum load cannot be set for the ex-
perimental system either with a dog food or a combination dog food and
cottonseed oil feed.

          Temperature:  The extreme temperature sensitivity of the process
can be seen in Figure 1.  The of ten-published!^.' diagram of performance vs
temperature of the digester places the optimum temperature of the mesophilic
bacteria at 38°C.  Occasional shifts in temperature in the digesters used
in this study showed that activity increased with temperature from 34° to
41°C.  In contrast to our observations, Lyman, McDonnell, and KrupiJ./ cite
experience with full-scale digesters where a 5°C variation had no effect on
gas production.

          pH;   The normal process range recommended for pH is 6.80 to 7.20.
The results of the pH measurements in this study imply that the range may
be much more narrow, as any drop below pH 7.00 seemed to portend process
difficulties.

                                    28

-------
          Microbial ecology.  Several hundred microbial smears were ex-
amined with the view to characterizing the dominant organisms  in the di-
gestion process.  No pattern developed that could be discerned by the
simple test except that the tetrad packets of cocci reported by CooksonJJL/
appear to increase during cottonseed oil feeding.

          Buffering agent;  This research program has also produced results
that show that soda ash and lime cannot be used interchangeably as buffering
agents.  Unless overmixed, lime buffered digesters can tolerate heavy loads
of biodegradable oil, while under the same conditions soda-ash buffered di-
gesters go sour.  Filbert—' has analyzed digester startup problems.  He
quotes both McCarty and McKinney as advocating the use of sodium bicarbonate
as a buffer.  Soda ash at a pH below 8.3 would be converted from sodium
carbonate to sodium bicarbonate.
Statistical Basis for Plant Design

          This research program is innovative in the use of replicate di-
gesters and elementary statistical techniques for evaluation of the data.
Few other recently published research papers report the use of more than
a single digester to establish the design principles and parameters that
are used for the construction of full-scale plants.  The study of the ef-
fects of radioactivity on digestion by Grune, Church, and KaplanZO./ used
statistical analysis to evaluate data from a bank of 18 digesters, but the
data from that study have little value for design.  While the overall re-
search program was conducted with impressive precision, many individual
data varied widely from the mean.  A design parameter without a measure of
variability is valueless.  With this in mind, the difficulties encountered
in practice with anaerobic digestion are easily understood.
                                    29

-------
                     CONCLUSIONS  AND RECOMMENDATIONS
           The  program described in this  report was  initiated because  sound
 biochemical  rationale predicted that  the process  of digesting waste fats
 and  oils  would be accelerated by creating more favorable  conditions for
 bacterial hydrolysis of these water-insoluble  materials.   The results of
 the  study have not precluded the use  of  surface mixers  or other devices to
 incorporate  scum  layers into the main mass  of digesting  sludge.  What has
 been shown,  however, is that if such  a process modification is effected,
 the  intensity  of the turbulence must  be  adjusted  to the nature of the feed-
 stock.

           All  waste treatment processes  must be highly  adaptable to the
 variable  quantity and composition of  the feedstock.  The  anaerobic di-
 gestion process is particularly susceptible  to upset when any of the  process
 variables are  altered.   There is no question that the process is far  from
 a state of maximum optimization.  There  is,  however, the  question of  the
 possibility  of further optimization while retaining the necessary degree
 of adaptability to feedstock.   Real time feedback control techniques  that
 have been adaptable to other chemical processes cannot  be used for diges-
 tion.   Few such techniques  are available for characterization of the  feed-
 stock and we do not yet fully understand the kinetics of  methane fermenta-
 tion.

           Current ideas in  sanitary engineering design  are that the dis-
 advantages of  digestion outweigh the  advantages.   (See  Table I.)  Still
 the  majority of sludge in this country is probably  stabilized by anaerobic
 digestion and  operators of  these digesters are continually faced with the
 problems  of  grease overload and scum  layer formation.

          A  continuation  of the research  program  described in this report
 would surely yield  better design parameters  for digester mixing.   Such a
 program should  include  laboratory studies with  replicate  1-gal.  digesters.
 Since gas  stirring  of  the type  employed  in the  present  study creates a
 fair degree  of  turbulence,  the  baseline  process should  be digesters that
 are  stirred  only by the effervesce of  the gas or  briefly stirred before
 sampling.  With  the  quiescent  digesters,  the effect  of  increased turbulence
 on the  performance  of the digesters in terms of conversion of volatile solids
 to gas, and  stability to  shock  loads of  grease would be evaluated.   Natural
municipal  sludge would be used  as a feedstock with authentic sewage grease
used as the  lipid  for shock  loading.

          Simultaneously with a  laboratory program,  operators in a selected
 region, possibly the Missouri River Basin District  (because of its  many
animal-processing  facilities),  should be asked about their experiences with
malfunction due to  grease overload.
                                    30

-------
          Information from these two programs will provide design criteria
for mixers that will help cure "scum" layers, and can be operated in a
manner that will not cause upset by a too rapid saponification of a shock
load of triglyceride.
                                     31

-------
     APPENDIX A
APPARATUS AND METHODS
         32

-------
                                APPARATUS
          A schematic drawing of a single digestion unit is shown in
Figure 7, while Figure 8 shows the actual components.   A later modification
was the addition of a 1-1 foam trap between the digester and the gas pump.
Fifteen such digesters were assembled into a unit that included a circulating
water bath.  The bath was maintained at 36°C ± 1° during most of the ex-
perimental period.  The complete assembly is shown in Figure 9.

          The most important single aspect of the project was the homogeni-
zation units because the entire program has been based on the premise that
scum-layer digestion might be accelerated as the material is homogenized
into the main body of the aqueous fermenting mass.  Homogenization was to
be effected by pumping the lipid and aqueous phases together through an
orifice under pressure.

          A Hoke  No. 6528L4B pressure relief valve when  fed with a Milton
Roy mRoy-110-A diaphragm pump has been shown to  incorporate a considerable
amount of vegetable oil into water and produces  a satisfactory emulsion
when the valve is set at less than 500 Ib.  This finding overcame the
greatest potential engineering bottleneck in the project.

          A Hoke  pressure relief valve was constructed  of brass and Teflon.
Although the brass was a possible source of copper contamination which
would have  inhibited bacterial growth, this deleterious  effect was  elim-
inated by maintaining an adequate level  of sulfide ion  in  the  digester
liquor as proposed by Lawrence and McCarty.££/   Another potential problem
was a continual change  in orifice geometry, due  to normal wear or abrasion
by grit.  A four-day  trial with an oil and water system showed  that no
unusual  wear damage  to  the orifice occurred under  these conditions.  When
the system  was used  on  a sludge that had not been  settled,  some  abrasion
of the valve orifice  was observed.   It was necessary  to replace  these
relief valves  several  times  during the study.

           It was  not  possible to measure the homogenizing  pressure  with a
gauge while the system was  in operation.  The hydraulic "hammer"  from the
diaphragm pump would  have  rapidly destroyed a  gauge.  An auxiliary  hy-
draulic  system with  a  pressure ballast tank and  a  snubber  on the  test gauge
was used to set and  periodically  check pump and  relief valve pressures.

           Gas  recirculation was effected with  a  small Presso-Vac pump  in
which  the  inlet stream was  reduced by means  of an orifice  which consists
of glass capillary  tubing  in which was  inserted  wire.  Frequent cleaning
and replacement of  the diaphragm  and Viton valves were necessary for con-
 tinuous  operation.

                                     33

-------
OJ
-p-
                            * No. 4  stopper
                           ** Na 13 stopper
                            * Tefbn Gasket
                          Fill
                          Portx

                        250 ml
                       Leveling
   Hydraulic
         Hoke Pressure
            Relief
                                          Bottle
Diaphragm
Pump mRoy
R-110-A
                  „.
                  fester
                   Liquor
                                            105?
                                           NaCI
                                         atpH3
                                           I—J
                                           J Neptune
                                           Gas Pump
                                                                                5 gal
                                                                               Carboys
                                                                               Calibration
                                                                               Scale
                        Figure  7  - Schematic Diagram of Digestion Apparatus

-------
.
u
                                Figure 8 - Photograph of Digestion Apparatus

-------


Figure 9 - Fifteen Digester Unit Assembly

-------
Gas was collected by displacement of an acidified 1070 NaCl solution from
one polyethylene carboy into another carboy which was under a slight pres-
sure.  The larger the volume of gas produced, the larger was the pressure
on that gas due to hydraulic head.  Appropriate conversion factors for
pressure and temperatures were used in calculating the amount of gas that
was produced.

          Gas samples were collected from the system by inserting a glass,
gas-collection bulb in the vent line.

          Liquid samples were obtained from  the pressurized system by
opening the sample port.  Stainless-steel tubing was used  to connect the
liquid and gas systems to the glass bottle which served as the  digester.
Nylon Swagelok fittings were employed where  nec-essary.  The liquid exhaust
tube was  1/4 in. O.D. which passed  through a 3/8 in. O.D.  tube  placed  in
the  stopper.  A Nylon Swagelok 3/8  in. to 1/4 in. union has been bored
through,  and the 1/4 in. end ferrules were replaced  with Teflon ferrules.
In this manner a gas-tight seal was maintained while  the exhaust  tube
could be  placed  to skim  the surface or sample the main body of  liquid.

          Material was introduced into the digester  through the liquid
and  gas return  line.  Thus, there was no holdup.
                                     37

-------
                           ANALYTICAL METHODS
                CONVERSION OF DIGESTER GAS VOLUME DATA TO
                            STANDARD CONDITIONS
          Several variables were considered in the conversion of the daily
gas readings to standard conditions.  These are:

          1.  The time span over which the volume change was observed,

          2.  Reservoir temperature,

          3.  Barometric pressure,

          4.  Head of liquid in the reservoir,

          5.  Variation in diameter among the reservoirs, and

          6.  Vapor pressure of the brine solution.

          Gas was collected in a polyethylene bottle that was filled with
a 10% brine solution acidified to pH 3.  Liquid in this reservoir was dis-
placed by the gas produced during digestion into another bottle that was
placed superior to the first, as shown in Figure 10.  Thus, the gas in
the reservoir was under a slight pressure at all times.  The datum that
was recorded by the operator was the volume of liquid remaining in the
lower reservoir.  Since digester gas could be vented or C02 added during
the liquid sampling and feeding operations, this datum was recorded both
at the beginning of the day, and after all the daily operations were com-
pleted.  An adjustment for this variable time period had to be made.  The
equation for calculating this adjustment is:


                            va = ^ (v1 - v2)                         (i)


where     Va = gas produced during elapsed 24 hr

          V| = previous afternoon reading of liquid content of the lower
                 reservoir in liters

          V£ = morning reading

           t = elapsed time in hours between Vi and V?.
                                    38

-------
                        VENT
Figure 10 - Diagram of Location of Gas Reservoirs
                        39

-------
          The equation for calculating the adjustment for reservoir tem-
perature is:

                          Vb = Va x   273'2    ,                      (2)
                                    273.2 + T

where     T = temperature in °C, and Vfe is temperature and time corrected
                gas volume.

          If a table of adjustment factors is calculated, Eq. (2) can
be simplified to

                             Vb = Va x F                              (3)

A set of factors is shown in Table X.
                                TABLE X

            FACTORS FOR TEMPERATURE ADJUSTMENT OF GAS VOLUMES

                 Temperature
                l£         H.                    Factor

                20         68                    0.932
                21         70                    0.929
                22         72                    0.926
                23         73                    0.922
                24         75                    0.919
                25         77                    0.916
                26         79                    0.913
                27         81                    0.910
                28         82                    0.907
                29         84                    0.904
                30         86                    0.901
          The adjustment for pressure must recognize the barometric pres-
sure, the liquid pressure head, and the vapor pressure of the brine solu-
tion in the reservoirs.  The equation for this adjustment is:
                                                                      (4)
                                    40

-------
where     Vc = gas volume at STP,

           B = barometric reading,

           H = liquid head,

           W = vapor pressure,

each as millimeters of mercury.

          In Figure 10,

           h = height of  liquid  exerting pressure,

           d = 635 mm, the  distance one reservoir was mounted above  the
                  other,

and

   R1  and R" = height of liquid  in the respective  reservoirs.

These  are related as  follows:

                              h = d +  R"  -  R1                           (5)

 but                             R" + R1 = C                            (6)

 so                         h = d + c  - 2R1                            (7)


 "C"  was actually the height of the liquid when it was all in the lower
 carboy and  this  amounted to 360 mm.  Since the lower carboys were calibrated
 in liters,  the datum that was recorded daily by the operator is liters of
 liquid.  R1   is  related to  VL  (the  volume of the lower carboy) by the
 empirical relationship

                                R1 = 18.4 VL                           (8)

 where  R'  was in units of millimeters height per liter contained.

           The variation in diameter among the 30 carboys that were used as
 reservoirs did not exceed a maximum of 2% of the average.   Since the error
 in reading the volume of liquid  in the carboy exceeded the  variation in
 diameter, the latter variation, while recognized, was not significant.
                                      41

-------
          Since we were interested  in  the pressure  change  over  the  time,
t , an average value for  R'  was obtained by  use of  the relation,
                         2R1 =  18.4  (V1 + V2)                          (9)

where  V^  and  ¥2  are liquid  volumes in the  lower carboy at  times   t  ,
and t2 , respectively.

          To convert the height, H  ,  into pressure head  in millimeters  of
mercury, H , we multipled by the specific gravity of  the brine which  is
1.07, and divided by the specific gravity of mercury  which is  13.8.   The
correction for change in density due  to temperature was  too  small  to  con-
sider.
                              H = 1^2211                                (10)
                                  13.8

          By combining the various expressions, we obtain


                H = i^|-  [635 + 360  - 18.4  (VT + V2)]                  (11)
          The third pressure correction,  the vapor  pressure  of  the brine,
is 18 mm Hg at 25°C.  The variation of  the vapor pressure with  temperature
in the operating range is too small to  be considered.  Thus,  W * 18   in
Eq. (4).

          A general expression  for standardizing the gas volume can now
be obtained by combining Eqs . (1), (3),  (4), and (11).
                             360 . lg<4  (v  + v
                 13.8                     1      -            76Q
which reduces to
                (Vj - V2) 0.0316F  r                          -i
           Vc = —i	^	  [59.15 + B - 1.43  (Vx + V2)J         (13)
where  Vc  is liters of gas at STP that would be produced  in a 24-hr period.
The actual calculation was performed by an IBM 360 computer.
                                    42

-------
               CHEMICAL OXYGEN DEMAND (COD)  DETERMINATIONS
          COD determinations were performed as follows:

          There was added in sequence to a sample flask, a 1-ml sample of
MLSS or MLSS filtrate, 20 ml distilled H20, 25 ml of 0.250 N K2Cr207, 30
ml of H2S04, doped with 22 g silver sulfate, AgS04, per 9-lb bottle of
acid, and 0.4 g HgS04.  The mixture was refluxed for 2 hr and cooled to room
temperature.  (Cooling is essential if the indicator is not to be oxidized.)
Two drops of ferroin indicator were added and the solution titrated to a
red-brown end-point with 0.1 N Fe(NH^)2(80^)2.

          Seven replicate COD determinations  on a MLSS  filtrate gave re-
sults with a standard deviation of 470.

          A small portion of the mixed liquor appeared to resist complete
digestion.  A trace of white oil appears  during  the digestion which re-
fluxes  as a steam distillate.  When the mixture  is cooled,  it solidifies
and  floats on the  top of the aqueous  layer.

          During the  experimental  period, COD was  determined  in both  the
mixed  liquor and the  mixed  liquor  filtrate.   Data  from the  latter  deter-
mination are presented in Appendix B  as  soluble  COD.
                           SOLID DETERMINATIONS
           The mixed liquor volatile suspended solid (MLVSS) was determined
 by a slight modification of the Standard MethodsjJL/ procedures.  Two
 changes were made.   The usual asbestos pad in the Gooch crucibles was pre-
 coated with fine sand (Fisher S-151) and 1 ml of a 2% solution of Rohm and
 Hass1  C-7 cationic  polyelectrolyte was added to the 50-ml mixed liquor
 sample prior to filtration.

           Number 3  Gooch crucibles fitted with "F" covers were used.  MLSS
 was determined by heating at 110°C and "Ash" was determined by firing at
 600°C in a Blue M-M-25A-1A muffle furnace.  In each case, the crucibles
 were allowed to cool in a desiccator loaded with Drierite.  Heating was
 carried out for at  least 16 hr and it was demonstrated that "constant
 weight" was achieved with both MLSS and ash in this period.  Since the
 amount of inorganic buffer added along with the feed was highly variable,
 the MLSS and ash are not significant variables.  The data  from the MLSS-
 ash are the MLVSS which are presented in Appendix B.
                                      43

-------
          The added polyelectrolytc introduces an error.  Since more than
1.0 mg of polyelectrolyte was needed to produce a satisfactory precipitate
(20 mg was used), it was assumed that most of this added material remained
absorbed on the MLSS.   This would cause a maximum error of the order of
+ 1.57, with the majority of samples that were examined.  Some portion of
the polyelectrolyte could also affect the COD determination on the filtrate
very slightly.
                              NEUTRAL LIPID
          Neutral lipid was determined by a modification of the method of
Loehr and Rohlich.22/  Fifty milliltters of the MLSS was placed in a 250-ml
separatory funnel equipped with a Teflon stopcock.  One hundred milliliters
of acetone was added and the mixture shaken vigorously.  Twenty-five milli-
liters of chloroform was then added and the mixture was shaken again.  An
additional 75 ml of chloroform was thcr added and the mixture was shaken
gently.  Two layers were allowed to separate.  The lower layer was removed
and filtered through Whatman No. 1 paper.  One hundred milliliters of the
filtrate was evaporated to a small volume in the hood and this small volume
was then transferred to a weighed 50-ml beaker.  This smaller beaker was
then placed in the oven  for at least 16 hr at 110°C before cooling and
weighing.  Replicate lipid extractions showed excellent precision.  Re-
covery of portions of cottonseed oil added to a composite sludge was 53.5,
55.5, and 56.57, for three determinations.  The weight of lipid recovered
from the extract was multiplied by 36.5 (20/0.55) to give "LIPID" in grams
per liter, which is reported in Appendix B.
                     METHANE CONTENT OF DIGESTER GAS
          The digester gas was analyzed by gas chromatography.  A Perkin-
Elmer Model 154B instrument equipped with thermister detectors was used.
The separation column was packed with 80 to 100 mesh Porapak Q which gave
excellent separation of C02, CH^ and ^.  The 4 ft x 0.25 in. column was
maintained at 40°C.  Helium was the carrier gas.  The relative detector
sensitivity as determined with analytical grade reagent gases was 1.37 for
the CH4/C02 ratio of peak heights.

          Figure 11 shows the system that was used to introduce the sample
into the instrument.  The 250-ml glass gas sampling bulb was equipped with
glass stopcocks that were sealed with silicone grease.  The Teflon gas
sampling valve (large black knob) permitted the evacuation of the sample
bulb as well as the stainless-steel sample loop that can be seen behind
                                    44

-------
Figure 11 - Sampling Apparatus for Gas Chromatograph



                      45

-------
the Teflon valve.  A sample loop with a volume of 0.474 ml was used.  It
is critical that the sample loop not be too large.  When we attempted to
calibrate the column with a 0.9 ml loop, the column was overloaded with
pure C02-  With the smaller loop, the calibration curve showed complete
linearity with all mixtures of C02 and
          The sample system could be completely evacuated.  A trace of air
was introduced into the sample from the stopcock extension when the evac-
uated bulb was attached to the gas reservoir exhaust line.  This air did
not interfere.  The column did not separate 02 and N2 .  This was not
necessary since, prior to an experiment, the gas reservoirs and digesters
were thoroughly flushed with C02 •  The gas reservoir samples were saturated
with H20 which did not produce any response with the fractometer under
the prevalent conditions.

          Peak height above the  baseline was measured on the readout chart
for CH4 and C02 •

                         CH  peak x 100
                                               "
                    CH4 peak + 1.37 C02 pealc
                       HYDROGEN ION CONCENTRATION
          A Corning Model-7 pH meter equipped with standard size glass
and calomel electrodes was used to determine the hydrogen ion concentra-
tion of the sample.  The electrodes were calibrated daily with pH 7 buffer.
A  100-  to  175-ml  sample  that was  freshly withdrawn from  the digester was
stirred with a Teflon coated magnet that was rotated by a Mag Mix while
pH was being determined.  The pH varied continuously with time, but con-
trary to the caution of Standard Methods that the pH would rise due to
loss of C0£, the  pH rapidly decreased  for a few seconds  then decreased
continuously but  much more slowly.  The pH at the apparent point of sta-
bilization was recorded.
                              MICROBIOLOGY
          Samples of the digester contents were taken for microblal assay
two to  three times each week.  These were diluted with one or two volumes
of water and spread on glass slides that were fixed by warming.  The smears
were then stained with Gram's stain and observed with an oil immersion lens
at a magnification of 980.
                                    46

-------
                                FEEDSTOCK
          The digesters were fed Purina Dog Chow that was ground with a
Waring Blendor until it passed through a 20-mesh sieve.  This material
readily wetted and could be transferred with a jet of water.  The Purina
Dog Chow gave the following analysis:

               Carbon        = 48.90%
               Hydrogen      =  6.25%
               Nitrogen      =  4.49%
               Moisture      -  8.52%
               Ash           =  8.03%
               Lipid         =  7.87%
               COD           =  0,943  g/g  solid
               Maximum
               Biodegradable =  84.45%

           The cottonseed  oil that  was  used as  a  typical  biodegradable  lipid
 gave  the  elemental  analysis:

               Carbon        »  77.71%
               Hydrogen     -  11.57%
               Nitrogen      =  00.00%
               Oxygen         =  10.72%  (by  difference)

           The  calculated COD from these results  was 2.891 g/g oil.   This
 is slightly higher  than the value of 2.785 that can be calculated for
 pure glycerol  trioleate.   Based on a density of 0.9140 g/ml at 25°, the
 oil has a value  of  2.64 g COD/ml oil.   Thus, in the reported experiments,
 0.357 ml of cottonseed oil was  the COD equivalent of 1 g of dog food.

           The motor oil that was used as a typical non-biodegradable motor
 oil was Philube  Gear.Oil - all  purpose 80 - MIL-L-2105 which is distributed
 by Phillips Petroleum.  It was  not analyzed.

           Tapwater originating in the Missouri River was used as liquid
 makeup.

           Sludge specimens were obtained most frequently from the first
 stage digester of  the two-stage digestion system of  the southern Johnson
 County (Kansas) Treatment Plant.

           Ca(OH)2> NaHCO^ and Na2COo  were used  in  various  combinations  as
 buffering agents at various time  periods  during the  study.   Exact  details
 are given in the sections on Experimental Protocol.

                                     47

-------
                       EVALUATION OF HOMOGENIZERS
          The combination of diaphragm pump and pressure relief valve was
chosen to simulate the performance of the type of homogenizer that is
widely used to homogenize milk and a multitude of other polyphasic mate-
rials.  A typical commercially available homogenizer is manufactured by
the Manton-Gaulin Company.  The mRoy pump and Hoke pressure relief valve
were picked because they were adequate for the job, cheap in relation to
available commercial laboratory scale units, and not needed for long-term
service.

          The homogenizing system was checked for two requirements.  First,
the homogenizer had to produce a reasonably stable emulsion.  A permanent
emulsion was not required as the surface of the digester was continually
skimmed and any surface lipid film (scum) would be re-emulsified rapidly.
The liquid pumps were set to circulate about 3 to 5% of the digester
volume per hour.  The other requirement was that the shear that produced
the emulsion was not so high that it would rupture a significant number
of bacteria.

          The ability of the system to emulsify cottonseed oil and dis-
tilled water was quantified by two rather subjective methods.  The emul-
sions produced by the pump-relief valve system were examined by phase
microscopy of a hanging drop.  The smaller micelles (globules of oil)
could not be counted because of the Brownian movement.  Thus the absence
of large micelles was the criterion of emulsification.  The second criterion
was "creaming time," the period required for oil to coalesce and form an
observable layer on the surface of the liquid.

          Oil:water mixtures, 1:9 and 1:49,  were examined in the system
with the relief pressure settings of 100, 200, 400, and 800 psig.  All
pressure settings produced emulsions.  This is a relevant fact as the
diaphragm pump is equipped with a spring-loaded ball valve that requires
50 psig pressure to open.  Therefore, the digesters which were fitted
with pumps designed for liquid recirculation only, had some small func-
tion as a homogenizer.  The pressure setting of 200 to 400 psig produced
an emulsion that was comparable to emulsions produced by a Waring Blendor
and also by treatment of the mixture with 20 khz sonic energy.  The 800
psig setting caused heating of the sample without much observable benefit.

          A sample of sludge was circulated through the pump-relief valve
system for 10 cycles.  The sludge was examined at 970 magnification by
phase microscopy and by stained smears.  No rupture of cells nor decrease
in population could be observed.
                                    48

-------
  APPENDIX B
ANALYTICAL DATA
     49

-------
                                   TABLE XI

                   RESULTS OF EXTRACTION OF NEUTRAL LIPID
                                                                     1-16-69

   1A      0.473      0.633      0.844    0.440    1.525    0.917     0.967
   IB      0.618      0.720      1.361    0.571    1.252    0.774     0.710
   1C      1.345      0.644      0.997    0.760    1.212    0.804     0.775
   2A      0.655      0.676      1.205    0.830    1.117    0.808     0.589
   2B      0.582      0.898      1.227    1.127    0.957    0.760     0.564
   2C      0.655      0.473      1.398    1.062    0.688    0.601     0.382
   3A      0.473        --       1.270    1.247    0.495    0.866     0.524
   3B        --       0.847      1.514    0.716    0.604    0.389     0.840
   3C      0.618      0.916      1.154    0.513    0.950    0.484     0.494
   4A      0.473      1.029      1.121    0.422    0.976    0.264     0.884
   4B      0.510      1.222      1.121    0.695    0.772    0.204     0.884
   4C      0.582      0.804      1.147    0.589    0.491    0.324     1.116
   5A      0.364      1.534      1.325    0.640    0.772    0.633     0.909
   5B      0.587      0.796      1.412    1.062    0.612    0.316     0.877
   5C        --       0.665      0.830    1.069    0.299    0.524     1.040
(grams/liter)
1-26-68
0.473
0.618
1.345
0.655
0.582
0.655
0.473
--
0.618
0.473










0.510
0.582
0.364
0.587
~ ™




12-27-68
0.
0.
0.
0.
0.
0.

0.
0.
1.
1.
0.
1.
0.
0.
2-10-69
1.
1.
1.
1.
0.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
274
110
198
106
890
085
688
990
579
717
968
874
535
568
491















633
720
644
676
898
473
--
847
916
029
222
804
534
796
665















1-3-69 1-6-69
0.
1.
0.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0.
2-13-69
0.
0.
0.
0.
1.
0.
0.
0.
1.
0.
0.
0.
0.
0.
0.
855
921
688
855
136
895
855
444
147
430
579
612
597
510
622















844
361
997
205
227
398
270
514
154
121
121
147
325
412
830
2-17-69
1.740
1.915
2.319
2.115
2.646
1.445
1.951
2.341
1.410
0.870
1.023
1.227
1.219
1.390
1.241
0.440
0.571
0.760
0.830
1.127
1.062
1.247
0.716
0.513
0.422
0.695
0.589
0.640
1.062
1.069
1-9-69
1.
1.
1.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
525
252
212
117
957
688
495
604
950
976
772
491
772
612
299
2-20-69
2
2
2
2
3
1
1
2
2
1
1
1
1
2
1
.311
.799
.097
.592
.437
.860
.219
.282
.584
.168
.303
.598
.765
.821
.700















1















-13-69
0.917
0.774
0.804
0.808
0.760
0.601
0.866
0.389
0.484
0.264
0.204
0.324
0.633
0.316
0.524
2-24-69
0.
2.
1.
1.
1.
1.
1.
1.
0.
0.
0.
0.
0.
0.
0.
899
148
096
259
875
467
558
736
761
575
466
528
946
892
692
Dieestor     2-10-69    2-13-69    2-17-69    2-20-69    2-24-69     2-27-69

   1A         1.274      0.855      1.740      2.311      0.899       0.943
   IB         1.110      0.921      1.915      2.799      2.148       1.270
   1C         1.198      0.688      2.319      2.097      1.096       0.914
   2A         1.106      0.855      2.115      2.592      1.259       0.979
   2B         0.890      1.136      2.646      3.437      1.875       1.117
   2C         1.085      0.895      1.445      1.860      1.467       1.147
   3A         0.688      0.855      1.951      1.219      1.558       1.372
   3B         0.990      0.444      2.341      2.282      1.736       1.147
   3C         0.579      1.147      1.410      2.584      0.761       0.681
   4A         0.717      0.430      0.870      1.168      0.575       0.815
   4B         0.968      0.579      1.023      1.303      0.466       0.797
   4C         0.874      0.612      1.227      1.598      0.528       0.848
   5A         0.535      0.597      1.219      1.765      0.946       0.859
   5B         0.568      0.510      1.390      2.821      0.892       0.830
   5C         0.491      0.622      1.241      1.700      0.692       1.041
                                     50

-------

Digester
1A
IB
1C
2A
2B
2C
3A
3B
3C
4A
4B
4C
5A
5B
5C
Digester
1A
IB
1C
2A
2B
2C
3A
3B
3C
4A
4B
4C
5A
5B
5C

3-3-69
0.655
0.560
0.542
0.560
0.651
0.778
0.793
0.716
0.331
0.371
0.120
0.360
0.462
0.371
0.469
3-27-69
1.916
1.207
1.604
1.873
1.113
1.545
1.789
1.120
1.124
1.291
1.353
0.855
2.043
1.873
1.887
TABLE XI
3-10-69
0.345
0.455
0.578
0.695
0.836
0.575
0.691
0.484
0.444
0.444
0.527
0.469
0.433
0.495
0.447
3















(Concluded)
3-17-69
0.545
0.724
0.713
0.644
0.112
0.935
0.858
0.705
0.331
0.356
0.585
0.465
0.658
0.422
0.367
-31-69
2.105
1.960
1.993
1.491
1.356
1.429
0.825
0.978
0.393
0.342
0.469
0.484
1.156
0.680
0.735
Extraction of acidified samples.
                                                 3-20-69      3-24-69

                                                  1.000        1.310
                                                  0.898        1.385
                                                  1.062        1.156
                                                  1.276        0.884
                                                  0.796        1.094
                                                  1.047        1.178
                                                  1.342        0.935
                                                  1.149        0.898
                                                  0.404        0.225
                                                  0.465        0.422
                                                  0.607        0.509
                                                  0.531        0.458
                                                  0.538        0.589
                                                  0.538        0.502
                                                  1.058        0.716
                                                4-3-69         4-7-69*

                                                 3.618          8.425
                                                 2.912         10.970
                                                 2.865          5.600
                                                 1.923          9.970
                                                 2.920         11.991
                                                 2.876         11.450
                                                 1.734          7.923
                                                 1.628          5.868
                                                 0.284          4.945
                                                 0.756          3.607
                                                 0.363          6.272
                                                 0.458          5.770
                                                 1.289          8.850
                                                 1.607          7.577
                                                 0.636          5.530
                                 51

-------
                                  TABLE XII

                   MIXED LIQUOR VOLATILE SUSPENDED SOLIDS
Digester

   1A        19.5      17.3      12.8     12.8     7.6      --      13.2
   IB         --       21.2      12.9      6.2     6.7      --      16.0
   1C         --       20.4      13.1      4.6     6.0     16.2     14.2
                                                                    15.2
   2B        17.7      18.6      12.9      5.6     --      16.8     13.8
   2C        19.0       5.3       8.4      3,1     4.4     12.3     14.6
   3A        15.9       5.8      12.4      5.9     3.6      8.6     11.2
   3B        18.7       5.0       7.6      3.3     7.1     14.2     18.9
   3C        20.0       5.2       8.4      3.1     1.1     10.4      8.2
   4A         --        4.9       6.6      2.2     1.2      --      16.3
   4B        17.4       6.1       8.9      4.1     3.8     12.0     15.
   4C        16.2       9.7       8.0      3.0     4.0     13.0     17.3
   5A        14.0       5.7       7.3      4.4     2.4     12.7     19.7
   5B        14.8       6.7       8.0      4.3     1.8     12.8     13.3
   5C        13.2      12.4       7.4      4.2     1.9     13.8     14.1
(grams /liter)
12-3-68 12-10-68
19.5
--
--
16.9
17.7
19.0
15.9
18.7
20.0
--
17.4
16.2
14.0
14.8
13.2
3-4-69
13.9
14.2
10.1
14.6
14.4
17.0
16.1
15.9
12.7
14.5
5.0
14.5
14.0
14.1
12.3
17.3
21.2
20.4
19.8
18.6
5.3
5.8
5.0
5.2
4.9
6.1
9.7
5.7
6.7
12.4
















12-30-68
12.8
12.9
13.1
13.0
12.9
8.4
12.4
7.6
8.4
6.6
8.9
8.0
7.3
8.0
7.4
3-11-69
13.5
20.1
18.3
20.1
20.0
17.7
16.9
16.6
16.2
15.8
16.9
13.5
14.8
13.8
14.4
1-7-69 1
12.8
6.2
4.6
4.3
5.6
3»1
5.9
3.3
3.1
2.2
4.1
3.0
4.4
4.3
4.2
3-18-69
22.0
23.8
19.4
20.8
24.6
10.6
18.6
15.2
8.9
17.6
15.1
15.4
17.2
17.0
15.4
-14-69
7.6
6.7
6.0
--
--
4.4
3.6
7.1
1.1
1.2
3.8
4.0
2.4
1.8
1.9
















2-18-e

--
16.2
14.9
16.8
12.3
8.6
14.2
10.4
--
12.0
13.0
12.7
12.8
13.8
3-25-69
24.2
24.9
21.0
20.7
24.8
14.9
16.8
16.4
16.8
16.2
16.2
15.0
17.4
17.0
14.7
 Digester       3-4-69       3-11-69       3-18-69      3-25-69      4-1-69

    1A           13.9         13.5          22.0         24.2         16.5
    IB           14.2         20.1          23.8         24.9         14.5
    1C           10.1         18.3          19.4         21.0         18.4
    2A           14.6         20.1          20.8         20.7         16.5
    2B           14.4         20.0          24.6         24.8         23.0
    2C           17.0         17.7          10.6         14.9         11.8
    3A           16.1         16.9          18.6         16.8         10.5
    3B           15.9         16.6          15.2         16.4         10.2
    3C           12.7         16.2           8.9         16.8         14.9
    4A           14.5         15.8          17.6         16.2         15.3
    4B            5.0         16.9          15.1         16.2         13.4
    4C           14.5         13.5          15.4         15.0         14.5
    5A           14.0         14.8          17.2         17.4         13.4
    5B           14.1         13.8          17.0         17.0         13.0
    5C           12.3         14.4          15.4         14.7         13.3
                                     52

-------
                               TABLE XIII

                              SOLUBI£ COD
Digester

  1A          --       15.6      15.8     10.4     6.1      --      18.9
  IB           6.9     14.1      15.6      9.9     5.5      --      21.8
  1C           8.8     12.8      14.4     11.1     6.0     15.6     20.0
  2A           7.4     15.0      17.7      9.7     6.6     20.3     23.1
  2B           7.9     14.2      14.8      6.4     5.5     22.3     25.9
  2C           8.1     13.7      14.9     12.4     8.3     16.2     19.6
  3A           8.0     14.7      15.3     14.9     7.1     29.8     27.0
  3B           8.8     12.9      15.0     11.7    10.2     19.3     29.0
  3C           8.8     13.5      15.6     11.0     3.8     22.8     31.3
  4A           7.3     12.4      11.5     12.2     6.9      —      24.4
  4B          10.6     15.5      14.6     13.1    10.6     16.9     23.6
  4C           7.9     14.5      19.7     14.8    13.2     16.5     21.9
  5A          12.6     13.9      12.3     12.6    10.7     22.6     29.1
  5B           8.5     14.0      14.2     13.8    12.3     32.7     31.1
  5C           8.7     14.3      17.6     13.7    12.3      3.7     23.1
(grams /liter)
11-26-68
— —
6.9
8.8
7.4
7.9
8.1
8.0
8.8
8.8
7.3
10.6
7.9
12.6
8.5
8.7
12-3-68 12-10-68 1-7-69 1















15
14
12
15
14
13
14
12
13
12
15
14
13
14
14
.6
.1
.8
.0
.2
.7
.7
.9
.5
.4
.5
.5
.9
.0
.3
3-4-69
16.
18.
15.
15.
21.
15.
22.
24.
27.
16.
16.
21.
18.
21.
16.
1
1
5
0
1
5
1
2
4
0
1
0
6
8
0













































15
15
14
17
14
14
15
15
15
11
14
19
12
14
17
.8
.6
.4
.7
.8
.9
.3
.0
.6
.5
.6
.7
.3
.2
.6
3-11-69
15
19
17
22
21
16
26
26
27
22
22
25
19
17
16
.9
.9
.1
.2
.2
.8
.0
.8
.6
.1
.6
.3
.6
.6
.2















10.4
9.9
11.1
9.7
6.4
12.4
14.9
11.7
11.0
12.2
13.1
14.8
12.6
13.8
13.7















-14-69 2-18-69
6.1
5.5
6.0
6.6
5.5
8.3
7.1
10.2
3.8
6.9
10.6
13.2
10.7
12.3
12.3
3-18-69
18.
25.
20.
23.
23.
21.
17.
21.
22.
18.
18.
17.
22.
22.
18.
9
5
7
5
4
3
6
2
2
5
0
5
0
0
4






























..
--
15.6
20.3
22.3
16.2
29.8
19.3
22.8
--
16.9
16.5
22.6
32.7
3.7
3-25-69
26
21
18
19
17
19
17
18
21
17
17
18
21
21
16
.3
.2
.7
.7
.8
.6
.2
.5
.5
.5
.3
.6
.2
.0
.8
Digester       3-4-69        3-11-69        3-18-69       3-25-69      4-1-69

   1A             16.1          15.9           18.9          26.3         31.0
   IB             18.1          19.9           25.5          21.2         22.7
   1C             15.5          17.1           20.7          18.7         23.7
   2A             15.0          22.2           23.5          19.7         21.2
   2B             21.1          21.2           23.4          17.8         22.6
   2C             15.5          16.8           21.3          19.6         19.1
   3A             22.1          26.0           17.6          17.2         15.2
   3B             24.2          26.8           21.2          18.5         16.0
   3C             27.4          27.6           22.2          21.5         20.6
   4A             16.0          22.1           18.5          17.5         14.3
   AB             16.1          22.6           18.0          17.3         17.0
   AC             21.0          25.3           17.5          18.6         13.4
   5A             18.6          19.6           22.0          21.2         17.5
   5B             21.8          17.6           22.0          21.0         14.8
   5C             16.0          16.2           18.4          16.8         16.3
                                    53

-------
                                              TABLE XIV

                                    METHANE CONTENT OF DIGESTER GAS
                                               (percent)

Digester    10-8-68    10-31-68   11-6-68    11-12-68   11-21-68   11-26-68   12-5-68   1-10-69   1-17-69
1A
IB
1C
2A
2B
2C
3A
3B
3C
4A
4B
4C
5A
5B
5C
46
45
57
50
50
62
41
48
44
55
55
44
52
50
51
45
49
46
47
49
49
45
47
47
46
48
47
44
47
48
57
55
55
54
45
57
53
56
53
54
55
55
53
55
55
56
57
53
58
35
57
52
57
56
56
55
55
53
57
58
50
49
49
47
46
48
49
48
49
43
52
52
45
33
50
55
56
53
54
55
55
55
55
54
54
55
55
50
52
53
54
61
56
56
54
54
57
53
55
55
54
54
55
55
54
57
57
57
58
58
41
34
42
49
43
44
39
41
37
38
64
63
62
63
65
52
48
50
52
52
46
38
41
42
34

-------
                                               TABLE XIV  (Concluded)
      Digester
2-7-69
                            2-14-69
                     2-21-69
Ul
1A
IB
1C
2A
2B
2C
3A
3B
3C
4A
4B
4C
5A
5B
5C
50
42
51
50
50
50
41
58
43
53
52
50
50
44
46
53
54
54
50
55
53
53
47
55
55
56
51
51
51
f- f- i U7
49
47
47
50
42
52
51
50
19
50
54
54
51
47
53
i-^O-O^
56
57
56
62
44
58
58
57
34
51
61
9
56
26
59
j-i^-oy
56
50
53
52
54
50
51
51
52
52
52
49
52
52
52
3-21-69
31
50
50
52
54
52
52
52
43
53
53
54
51
51
52
3-28-69
38
52
52
53
51
58
53
53
52
55
54
55
53
53
53
4-4-69
50
48
46
47
46
48
45
48
46
32
49
52
40
41
50
5-16-69
54
52
54
54
53
53
55
53
53
54
54
52
51
52
52

-------
                              BIBLIOGRAPHY
 1.  PHS Publication No. 1065, "1962 Inventory Municipal Waste Facilities,"
       U. S. Government Printing Office.

 2.  Loehr, R. C., and T. J. Kukar, "Removal of Lipids by Conventional
       Waste Treatment Methods," International J.  Air Water Poll.,  9,  479
       (1965).

 3.  Hunter, J. V., and H.  Heukelekian, "The Composition of Domestic Sewage
       Fractions," J. Water Poll. Control Fed., 37, 1142 (1965).

 4.  Davis, J. B., Petroleum Microbiology, Elsevier, New York, p.  260 (1967),

 5.  Raymond, R. L., and J. B. Davis, "n-Alkane Utilization and Lipid
       Formation by a Nocardid," Appl. Microbiol. . 8, 329 (1960).

 6.  Beerstecher, E., Petroleum Microbiology,  Elsevier Press, Houston,
       p.  192  (1954).

 7.  Jeris, J. J., and P. L. McCarty, "The Biochemistry of Methane Fermen-
       tation Using C14 Tracers," J. Water Poll. Control Fed.. 37,  178
       (1965).

 8.  Chouteau, J., E. Azoulay, and J. C. Senez, "Anaerobic Formation of
       n-Hept-1-ene from n-Haptane by Resting  Cells of Pseudomonas
       Aeriginoso." Nature. 194, 576 (1962).

 9.  Sarda, L., and P. Desnuelle, "Action of Pancreatic Lipase on Emulsified
       Esters," Biochem. et Biophys. Acta. 30, 513 (1958).

10.  McKinney, R. E., Microbiology for Sanitary Engineers, McGraw-Hill
       Book Company, New York, 1962.

11.  Dr.  William Spangler,  Midwest Research Institute, Personal Communica-
       tion (1969).

12.  McCarty, P. L., "Anaerobic Waste Treatment Fundamentals," Public
       Works. November 1964, p. 94 and October 1964, p. 126.

13.  Mixing Equipment Company, Inc., "Town of  Tonawanda Eliminates  Scum
       Formation on Sludge  Digestion Tank," News from Lightning. April 1966.

14.  Basu, A. K., "Treatment of Effluents from the Manufacture of  Soap and
       Hydrogenated Vegetable Oil," J. Water Poll.  Control Fed..  39, 1653
       (1967).                                                  ~~

                                    56

-------
15.   Technical Practice Committee,  "Anaerobic Sludge Digestion - MOP16
       Sec.  3 Raw Sludge Feed to Digesters," J.  Water Poll.  Control Fed.,
       38,  1840 (1966).

16.   Technical Practice Committee,  "Operation of Wastewater Treatment
       Plants," MOP11, p. 44.

17.   Lyman, B., G. McDonnell, and M. Krup, "Startup and Operation of Two
       New High-Rate Digestion Systems," J. Water Poll. Control Fed., 39,
       518 (1967).                                                    ~~

18.   Cookson, J. T., and N. C. Burbank, "Isolation and Identification of
       Facultative Bacteria Present in the Digestion Process," J. Water
       Poll. Control Fed.. 37, 822   (1965).

19.   Filbert, J. W., "Procedures and Problems of Digester Startup,"
       J. Water Poll.  Control Fed.. 39, 367  (1967).

20.   Grune, W. N., C.  F. Chueh, and C. H. Kaplan, "Effects of C14  and
       Sr90 on Anaerobic Digestion," J. Water Poll. Control Fed.,  35,
       493  (1963).

21.  Alonzo, W. L., and  P. L. McCarty, "The  Role of Sulfide in  Preventing
       Heavy Metal Toxicity  in Anaerobic Treatment,"  J. Water Poll.  Con-
       trol Fed., 37,  392  (1965).

22.  Standard Methods  for  the Examination  of Water  and Wastewater.  12th
       Edition, American Public Health Association, New York  (1965).

23.  Loehr,  R.  C., and G.  A.  Rohlich, "A Wet Method for Grease  Analysis,"
       17th Ann.  Purdue Industrial  Waste  Gonf., p.  215 (1962).
                                     57

                                            4 U.S. GOVERNMENT PRINTING OFFICE : 1910 O - 408-301

-------