1\‘-./:-. ‘3'? if! .E‘t' 3" F5: 1/: f2.-'\'. i r;::.\/
" ‘ ‘Ek ."J ' 3 i\ ”\5 Vanni“

f
v

A Sm: g “durum Qmo‘“

{:30 {he Dfiqt‘aa sf M: a:

L.LULLA STRTE QTY ?b'“”
1 We Vern '-" Num’erge:

$63

ALL _‘

Mnmnmrmmnmm $333; .

31293 01096 99 7 - 94

University

 

 

, 63¢

4W. 3 3 Ma

 

 

 

 

 

 

 

 

.1-

L1

'T'fi“
it.)

on

.- L «23;;

“'\

1" TTI'fi '1'
.Laidv-.-.':.'u

1*;
L4
J4

-'..r
L1.

2".

pl

.La

A _.
y-..
v—A—

oer

“/17.“ 3
A‘

.L

1L

n

1

‘JCIQEO

..
.9- T}

r'a
b.

‘ - A ~r-
11:1:

‘ ‘
[1:1

a1 fix

1

..(\.1r‘\‘r‘
L2'.J. \l

TL“

13:1

6
O

“"I C"
0v 4.4;. .21:

. .
7"

V,
‘-

‘1 3
4.4.2. A.

H -1.
L-JLDL.

_)

I
'-..cr‘~— *“w,
(Lb—AI. .LlL-i

1

“_il

1.5L 1.;-

l--. -4 '1
l ir‘
v LLL C4.J.

ul

9110

A r:

ABSfifiACT

REKOVAL CF LIQUID FRCL POULTRY WASTE BY ELECTED-OSEOSIS

by FRED VERNON NURNBTQG"R

The removal of moisture from chicken excrement by the
process of electro-osmosis was studied using four voltages
5, 10, 15 and 20 volts D.C. Electrodes of two different shapes.
and three different materials were used. Current flow vs.

elapsed time, temperature vs. elapsed time. liquid expelled vs.

elapsed time. and moisture content reduction vs. applied voltage

The plug type electrodes gave higher current flows, tem-
peratures and greater moisture content reduction than the rod
type. Stainless steel electrodes were slightly better than
steel and corroded much less than copper electrodes.

Initial moisture content of the drOppiigs was 78 to 805

on a wet basis. The maximum reduction was $.3fl on a wet basis

which is not sufficient to permit pelleting.

 

ggppiwrvec.

 

Departaent Chiirman

CONTENTS

Abstract
List of Figures
Introduction
Apparatus
Procedure
Data Evaluation
Results
Current
Planinetering
Temperature
Electrode Erosion
Liquid Xpelled
Koisture Content Reduction
Problems for Future Investigation

Summary

ii

Ci)

A) N
CO N N

N N
O)

Fig.

10

11

17

13

1‘4 0 0

LIST OF FIGURES

Excrenent Container

Experimental Apparatus for Plug Type
Electrode

Experimental Apparatus for Bod Type
Electrode

Average
for

Average
for

Average
for

Average
for

Average
for

Average
for

Average

Current Flow Vs. Elapsed Time
Stainless Steel Bod Type Electrodes

Current Flow Vs. Elapsed Time
Steel Rod Type Electrodes

Current Flow Vs. Elapsed Time
Copper Rod Type Electrodes

Current Flow Vs. Elapsed Time
Stainless Steel Plug Type Electrodes

Current Flow Vs. Elapsed Time
Steel Plug Type Electrodes

Current Flow Vs.
Copper Plug Type

Elapsed Time
Electrodes

Current Flow Vs. Applied Voltag

haximum Temperature Difference Vs.
Applied Voltage

Erosion

[Ti

Average

rosion

of Plug Type Electrodes
of Hod Type Electrodes

Electrode Weight Loss Vs. Applied

Voltage

Average

~ -‘1 r.
Average

for

Average

Liquid Expelled Vs. Applied Voltage

Liquid Expelled Vs.
Bed Type Electrodes

Elapsed Time

Liquid Expelled Vs. Elapsed Time

for Plug Type Electrodes

hoisture Content Reduction Vs. Applied

Volt

r) fi A
605; C

iii

10

11

12

13

15

17
19

(\J
K»)

l\)

(‘3
\1'1

5)
\3

NTEODUCTION

The present trend of poultrymen and other livestock pro-
duc€"‘<" i t “:7 w w 1 f c ' L". , ‘1’" a ‘ - _' ." C1“ . , _ ‘ fie-p (‘3) run . ”I

lo s OLde con incment housing and feed lots. inc pitclem
of waste diepcxal from these operations is becoming more critical

with their increasing animal concentration and proximity to

\v

densely populated areas. Field spreading of the waste material
is not generally possible since in most cases the feed is pur-
chased and thus sufficient field area is not available.

The me hods of composting and lagooning are objectionable
for five major reasons. These are: i. the creation of objec-
tionable odors by their bacterial action, 2. the flies that use
these areas for breeding create a public nu sance and health
hazard, 3. drainage from these areas can contaminate underground
water leading to public and private water supply lines, 4. plant
nutrients are lost from the composte pile due to decomposition

and leaching. and S. in lagoons all the plant nutrients are lost

.1

.1

since the material is thoroughly decomposed. Also large areas
are required which makes them prohibitive for large operations.
Thus some new form of waste disposal is needed.

In addition to disposing of the Waste material, it would
be desirable to have a saleable product. This would provide
the livestock owner with additional income from materials he is
presently throwing away. The product Evould be in the form of
a dry or pelleted substance suitable for use on lawns, shrubs,
gardens. etc. From reports on pelletizing feed concentrates

and hay, it appears that the moisture content of the excrement

would have to be reduced to a range of 5-25/ wet basis for
pelleting. This is the range of moisture contents reported
for feed and hay respectively.

Otis E. Cross in his thesis, "The Influence of Variable
Parameters on the Electro-osmotic hoisture Migration in Poultry
Excrement" at Lichigan State University (1963). concluded that
the amount of liquid expelled by electro-osmosis was a function
of: 1. current flow, 2. time, 3. length of sample, and
4. moisture content of the sample.

In his investigation, Cross used a constant voltage that
provided equal initial currents. These values were held to a
sufficiently low level to reduce the affect of Joule heating.
The samples were placed horizontally to eliminate the effect
of gravitational flow. The results of that study showed that
the moisture content of poultry excrement was not reduced to a
pelletable level.

The purpose of this investigation was the utilization of
Joule heating and gravitational flow to assist the electro-
osmotic process in reducing the moisture content to a suitable
level. Preliminary tests indicated that if the voltage was
greater than 30 volts the samples became very hot. The liquid
was thus evaporated, in fact boiled, instead of being removed
by the electro-osmotic process. To keep the Joule heating from
going to this extreme, the voltages selected for this investi—
gation were 5, 10, 15, and 20 volts D.C. The samples were

placed in a vertical position to include the gravitational flow.

C‘Q

In addition to th 5, different electrode materials and
shapes were studied. The metals chosen were: 1. copper,
2. stainless steel, and 3. steel. Each of these three was
obtained in two different forms: 1. plug type 2" dia. X 1"

thick and 2. rod type 3" dia. X 2" long.

APPARATUS

The containers used for the test samples (fig. 1) were
made from plastic tubing with 5" I.D., wall thickness of i",
and a length of approximately 15". A piece of sheet aluminum
was mounted on one end of the tube to form the base and serve
as the cathode. The plastic tube was sealed to the base with
calking compound. Holes, 1/8" dia., were drilled through the
base to allow passage of the liquid. Plywood was used for a
loose cover on the tube. Two holes we“e drilled through the
cover to permit the entrance of the electrode wire and to hold
the thermometer in place. The thermometers used were standard
laboratory -20 to +1100 C thernometers.

The sample tubes were placed on a stand over receptacles
for the collection of the liquid expelled. heasurement of the
liquid was achieved by pouring the contents of the receptacles
into a 100 milliliter graduated cylinder.

The electrical system was composed of a D.C. power supply,

electrodes, voltmeter, ammeters, and the samples (fig. 2 d 3).

E XCREMEN T C 0” TAM/ER

 

 

 

 

 

, . I... -T
I
. -- I.
7'“ I“ \~
I ' ° ' b ‘*
. ' 0 ' .
.. ‘ , \ )— ‘3. y n. IH\II
‘ ') 7 o , .’I
\\~ ”(I 0 ‘j/
f‘\ O ‘ 1/;
;\. ,. 0
\ . u
”I - -- J— D/AM. ’HOLES
6
0 I
,7- . __.I'n_____-,-.- .I
.4 52 --—-- h-J
:..4. 5" .4
' % I; - --_‘_
I | .
I , I _
I I .
I ' I
' l
I
I I t
I I H)
I I | . \
I i I
I I I‘
I . _
| . II ,
I ' ’
l i
I .
TERMINAL , i I g
Sceny " ' “‘ ‘p I 1 : I .i - BASE 3172’
~- ~ -“- "- ~ (a mo.)
L 9 ’l y. -
o I ‘
/.;’ T r .‘ s
a ,, . \ 0‘
, ' \\
I
‘ «h- ‘I '1'}, . I .
‘ J {’1' I
"f a 1 '
0 ' \IN I
\ 7y ’ V I

 

 

N mi

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

lequm
9303 «OK
Wh‘flQx qum .WNUTKQWUIWT
WQQETUIQ
NNQE<WI IIINIJ\‘ a. _
_ - WQS .ué‘
_ . 3
“ ow
_ c . u know .3 o:
. , ...I. _ I L NQQ‘DW q
_ In have ”R 0.3m. Q menu
_ V , "WI "39% .o .Q
_ _ ‘ >U
«NEE >60 _ _
9530‘ u _
_ _
_ .
w ax
FE. NEQQ $3. 5 C

 

 

.onQk umflw Iquik seq
Wok Wok (QVQQT qwb<mx<§muxw

M Swank

QNUNQ kw. t.
n: 30\ V QOQ
WQUEQQUNQ

 

WHY wq thm
WWQEK «0Q-

 

 

Nomi},

 

 

 

 

 

 

 

 

umlau

 

 

 

 

 

 

 

 

 

 

 

 

.. .- f»- ..I:. 1).) Y. .
a. i Fwd _ .
2 a __. xiii _ Arr"_ .
i a. ._ _ a! .333.
a _ I: WWEOQ
— — . m - QVOQ
. n _ A)
«mkékeou _
3&qu " _
_ _
_
m- A?
«mt. SeER-«kk- : a

 

 

mace-kowww was.» 00¢
«6K wax «u 3.6-. v. «wkéwkxwmo-xw

w\<\ V dd.

O:

3

PriOC21. DUKE

Chicken e: {crement for the tests was obtained from the
dropping board of caged laying hens. All tests were made with
excrement collected from the same group of chickens. The
material for a test was placed in a large container and thor-

ougiily Stil red to obtain as uniform a n xture as possible. Two

mples of approy :imately 150 grams each were taken from this
mixture, weighed, and placed in a 2000 F oven to dry. From
these samples the initial moisture content, % wet basis, was
determined.

Each tube was then placed on a scale and four pounds of
the m xed excrement placed in it. This value was chosen for
two reasons: 1. uniformity, and 2. it filled the tubes appr
imately one-half full thus allowing room for any expansion that
might take place. Each sample was tamped in an effort to remove
the air pockets and to obtain constant bulk density of sarnples.
Five sazples were used for each test run: 1. two for each
electrode type eg. two samples with copper plug electrodes and
two sazuples with stainless steel plug electrodes, 2. one sample
for control with no applied voltage. Thus two different electrode
types and/or materials were tested during each run.

Each electrode was weighed before and after each test run
to determine its change in wcight. The elctrodes were placed
in the center of the cylinders. The plug type was subnerged
until the tOp of the electrode was even with he top of the sa1 ple

(fig. 2). The roa type, however, was completely subnerged to the

The data recorded was: 1. time, 2. volta 3e (adjusted
to prede ermined value), 3. anre ter re ading for eacn sanple,
and b. the temperature of eacn ea 11 . These readings were
taken hourly until a d.eiinite cownvait trend becane estailished,
then periodicallv for the duration of the test. The liauid
removed P38 measrred whenever there had been a sufficient amount
collected to cause neeli-ibic ea ror while transferring it from

the receptacle to th graduated cylinder. Thus its frequency

eepeire2 to Le very low. The steps involved lJ the tei 1 tloq
were as follows: 1. she final data readings; 2. shut off
pover supplv, disco1aect, clean, and wei3' h electrole s; 3. taor—

oush nix ng of each sample; and b. weighing of approximately
150 gns. of the contents of each sample and placement of it in
the Ci;i1I oven for final moisture content determination. The

U

remainder of the sample we 3 discarded and the equipment made

a

vere

ready for the next test run. All of the dryin3 samples

left in the 0V en for a air i11um of 36 hours to ensure complete

DATA EVALUATION

Evaluation of the data consisted of plotting the following
raphs: 1. current flow vs. elac psed tiL1e; 2. temperature vs.
elapsed ti e; 3. liguid removed vs. elapsed time; 4. moisture

content reduction, Z wet basis, vs. applied voltage; 5. naxinum

temperature deviation vs. a1plied voltage; and ~. averaae

electrode weight loss vs. applied volta e. The current flow
vs. elapsed time curves were planimetered to detern'ne the
total ampere-hours of current used. These values were then

used for electrode evaluation and electrode weigh less deter-

mination.

C T 1:13 TF7, =71
\4‘-L-L .1- - a.
-—————-

The 3eneral shapes of the curves for current flow vs.
elapsed time were of two forms. The simplest (type 1) was
m notonic decreasing from its initial value until it approached
some asymptotic value. The other form (type 2) started like
the first but after a few hours began to increase. This con-
tinued until an absolute maximum had been reached. Then, it
usually became monotonic decreasing and continued similar to
type one. Occasionally secondary local maximums occurred.

For the 5, 10, and 15 volt D.C. tests all samples with
the rod type electrodes gave curves of type 1, fig. 4, 5, d 6.
At the 20 volt D.C. level the rod electrodes gave results similar
to type 2. however, the secondary local maximums were much less
than the initial current level. The stainless steel plug type
electrodes gave type 2 curves for all four voltages tested, fig.
7. The steel plug type electrodes gave type 2 curves for the
15 and 20 volt tests but type 1 curves for the 5 and 10 volt
tests, fig. 8. The copper plug type gave type 2 curves for the

10, 15, and 20 volt tests and type 1 for the 5 volt tests, fig. 0.

 

 

1»-

\ !\\
WLK u. .If \NW . .
em ON 3 w m e. O
.- I_1 I I» I m I I. IIHTIIIIITI - +I _ i r j. 0
.Ir-II +\Iv1-f-IV1«) _
u). we? Io.\ »/ / xI raw-Es kn.) _
\I / / . ‘1 I: ikpf 1* 4.! N.
/ J.
/» near 2 _
/ X. 2 ”
aflx. .x. x1 J. “
J/f/ Id. .1?
x/ I. J
/J. M 7.
, / . VJ
.- A t w
/ L J.
. . / ._ //
, X r91.
. L
- z
w! 6. 0 IIIIN‘. _
are D. N . VI..." W
M, t.
. . W
i
D
i w
\ . x :4. . .8 s) 14: 6|!) . 1F .
1.1.. 1n. NI. are . .......\.\Nnm, . L New
L
. 41V.\
’
L
a 11.2.
r
v! r ,
W§AK .wus..-\\I1 .2 . . I.-.I)|, - \8... l L1
Lee

 

10

 

Mm m... 1% c .-
TI I + “II+\ I+I . L IILiN v IN\ .1 N.\ + W .1 V W Q
«I! L I IN! 1..I I I
1 IQ]? I” JQIILIQIW. IUA/ I? I“. “If/flag N.
d W/ 0101 Q\ .IIN.’
/ \fl IQ] fl
\ / . Y
(a. a. {N v.
1 Addie: fl
/
r / Ivan 4am
A

«o-
—.
5.:
I
L
'
(U

P
A”
,d/
é
SIC/«4v ‘ M07j Lava-1503

/ x
,

, ,
\I) . . IN.
EQWLUWNN N31... 99% Nah-U) ,9 LV \

/
1.1....
it:

.1.qu ruddy-3G. h 1.11.“... I._..<.......+ NU ...-....<._..w.\

 

11

m .914

.2 2.x ...w\..\.r.. 0.3....- 3....

 

mN VN ON 2 N\ m. V O
.-I..II.II.-I+.I.-IT-. oIiI It I. w v .r .1 a n a ._ a , 0
I I . /
O. I. / 0 I‘QIOIIO I
\\N\VI0IPI9I/ at? 0x . lo 9/9 N.
V we... .2 d. .0. II.- ./
2 , w, J 1. .9 A
/ to la
1... . .. . . 4/
.11“... .. / . I“: Is“.
. was“ ONIIN», 1, 3
x. / n
/ I II 795
(IL-1‘“ 0 I/I ... mm
/ . H
X i 4
- L

”‘11
7j

3)
§a¢q v “‘m 0

”Madman “Nut... MRAR now when. «GU

N

A -!-“""'

ll-

hvfimx. numb... 3N 9V. >1. 0.... .. beck-.90“). mmqw N... .01

L

 

12

 

h .00.
.hi\ «hit 395%

 

 

 

mu wu .3 Q A S m v . 0
« w .1 + -- Jr .1. w + a-.. .5. L ..... i]. .1 w .. IillJ, O
:N.
7/ mv\03. “1%
T 1/» n
3%.
x W
/ 3‘... Lfiw.w
/ . “V. W.
/ $69 9 1
. er:
U
. .... O
/' ibvm
$3 \
P‘Qowkuwum rrwx
NQAAK “aqq ukawl WWWQ§§TKMI ”to? ON
. -
Jr\t\
:3
:W.\

mfiR 93qu §

3 q Q 3.3 $6 m. $33 1

 

1'3

m .wQ

.2 wvéun Q38. q RV

mm VN 9w. . 3 E

D F U P D - P D

. N
F )
1r-

 

, ”in? .3 q

 

”38 ON
.flwQO Wk UNNN WRAP 9o uo\ NNWLM;

NEH». w....;.\u.w m... ....uQ -éwmeU mw‘w‘qu.

 

 

 

m. GE

 

.93 NSC Q qu<qm
NM » NM vN ON 3 S Q
q 4 « " #:ig? 21.... h 0 +5 lie- L w 4.

 

£3 8

WNQQNRva Six 93R QNQQOU

 

 

 

wwwwms <

 

L

 

'
M011 J/Wavawp

<2“
vo\

L

.‘x

 

Q.\

.15

The magnitude of the secondary local maximums increased with

increasing voltage.

RESULTS OF PLAEIEETER KG

The current flow vs. elapsed time curves were planimetered
to determine the total ampere-hours of current used. The fol-
lowing procedure was used: 1. each curve was planimetered
three times and the average value taken; 2. the values for
the two curves for each electrode type and material were then
averaged together; 3. the results from step (2) were then
divided by the total elapsed time of the test to determine the
average current rate. The results of this are given in fig. 10.
The results from this procedure show that the copper elec-
trodes permitted less current flow. The stainless steel elec-
trodes had the largest current flow, but the results from the
steel were very close to it. This was due to the corrosion
rates of the respective metals. The copper electrode corroded
and reacted with the adjacent material much more rapidly than
the stainless steel and steel. The resulting material reduced
the conductivity and thus the current flow.

The larger current flow rates for the plug type electrodes

'were due to the greater surface area of the pl C"s. They gave

'better sustained electrode to sample contact than the rod types.

razcrszmr URE
The temperature differential i.e. (sample temperature) -

Ozontrol sample temperature), increased with increasing voltage.

16

/0 001”

0.9a

AVE/{465 CURREN? Fww, any);
9 .0 9 ,0 9
*3 o \I Ch
9 i? 9 9 .0

o
‘9"

0.201

0. /&

 

APT-"r. ...'. - ._z’,RREN7 FLOW
'~ A x50 VOLTAGE

—-o- COP/‘51}?
+ 57A INAEJSE; .‘i/EEL

  
    
  
  
 
  

—A—— :3 72' .-.'
____._ #ZUA E71”! 2‘.

- -._ Asa 77/12”

 

 

N
OJ“

)5
A PPM-'0 [/04 7/165, tidal: air.

i_F'/c;. IO

17

This was due to Joule heating i.e. W=I23= EZ/fi. The maximum
temperature differentials for the voltages tested are given
in fig. 11.

The plug type had larger temperature differentials than
the rod type. This was due to the higher current rates occurring
with the plug type. The order of copper, steel, and stainless
steel for increasingly larger temperature differences was caused,

also, by the larger current flows occurring in that order.

ELECTRODE EROSION

The erosion of the electrodes is clearly visible in figs.
12 and 13, which show the electrodes as they appeared before
and after the tests were conducted. In both figures the copper,
steel, and stainless steel electrodes are designated A, B, and C
respectively.

The average rate of the electrodes was evaluated by dividing
the total weight loss of the electrode by the ampere hours of
current used. The values calculated are shown graphically in
fig. 14.

All of the erosion rates increased to a maximum and then
decreased with increasing voltage. For the plugs this maximum
was at the 15 volts D.C. level, but for the rods it was 10 volts
D.C. The cause of this was not determined. In general, the
rate of erosion in decreasing order was copper, steel, and stain-
less steel. This is the reason the current flow rates were in

the Opposite order, as mentioned above.

F4
U ‘.

60v

50«

°c
~n.
9

6

7254/05/21: was D/FFfegvcg

l:

 

 

IA

MAX/MW! lie/w 24 MA 70/?! fi/F/Zfl/W’Z
1/‘5 ff”, 'r' _’£‘D VaL7AGE’!

——o— Comm
+ STA/A755“ -,”'1,-¢L
—A- 57511

pl 06) [f/Jé

 

-— ‘— ~— Roa UP;

 

 

4 /
/
/ /
4/ /
/
._ /o/
fa f5

APPUZD Vans; , van, ctc.

FIG. II

19

 

 

. r~_,w

‘v‘w‘

“7 —._..-,_

- _ w
ice.

Erosion of

13

i"

'U

 

Avg/24 61:“ £1. EC7/2006” Wé/GH 7 1.0.55

 

 

T V5. 43/24/th MO\L7AGE
_. // \ \ ‘3‘ ‘ ‘
/o40f / ~ 0
/
/ C
if / '
/.20+ ,, x“ . " o
. 74, l. ._ 73;
~L /'l \‘ \
W‘ ‘ fl/ / \\ \
\z/‘OO‘F / ,..-—-—-)<~ .‘ \A\\\ \
,5 (4/ / \ \ \ \ SA
0) 4 f A
\ I] ‘ 6' ' '
//
g :80‘" //
. / I
k a
3!; /
g 060" /
[I , _o_ COPPER
(k?) _. l / . —+— STA/NLESS Suez.
*1" / —A— STEéL
1‘. 40¢ ’
st. // ---—~ PLUG 7/195
% // - “ " “ 1900 77¢;
0201'
O + L a i
O .5' IO A! 20
APPL/Eo VOLrAéé‘, va/A alt.

FIG. 14

21

[:1

LI QUID 531' Eli .’.1D

The quantity of liquid expelled due to electro-osmosis
increased with both applied voltage and tide as was discovered
by Cr ss. fhe alues were dete‘mined by subtracting the control

sample value from the sa q3le value and then averaginc the re-

sults for like samples. The values are shown in figs. 15, 16,

and 17.
T avoid excessive crowding in the liquid expe elled vs.
elapsed tine curves, only the two extreme cases are shown,

namely those of the stainless steel and cop per electrodes. To

1 '-. 1 4‘ ' --- r- I"
1e sage coorpinace sys em scale was used

)«J

facilitate comps wri.on t
for both the plug and the rod type electrodes. ihe plug type
e::pelled between two anl hree ti es more liquid than did the
corresponding roe type of the same rate; ial.

~tinate of the cost involved to remove this liquid

k)
7
PJ
(D
(’1

:as made. rhe results for the stain es steel plug type elec-

"'\ “ ~ --. ’1'," - J— “ h r" ’1 -v ~“'\ 1 ,‘u
.LOL‘CS at .;U \fOle D.C. {ll/10. 1:432 /r.w1'1 care:

I I
Kl)
\]

CI)

:3

:5

c+

7

From plwaimctei results (18.9 anp-hr)(20 v)
tOtL.l 11:11.11 d I“:3;;LOVCG. = LLILLOS .11. ¢

Cost = (.373 Lw1,( 2¢ /:1h) _ 1/:
(222 05:11) (2 005341 O-T: Fr ,Ll/ [.41)2

l of liquid removed

Cr?
(‘13.

,-. mw1~'-__-:.-\ ““T.'1 ‘1 '3'} "1'7V'TV‘V -r-C-vr
LloyLLJ‘LAjJ CLIL._..J: .-.'.L._1JJKI.I‘V.1..L LA

Tne initial moisture ceite11ts ra iged from 7: to dvp 10b
Exisis. The reduction in moisture content due to electro-osmosis

_,s determine? by subtracting the reduction of the control sam-

’-‘ ~ 4-1. - .~ -. I“ u . ~. .1 -1. 1 - .
“Jr: IrUm the Obnol sanples. lhe average Value of line samples

Ave/91165 0430/0 Exp 5450
V5. 1419/94/50 VOL TAGE

 

 

266T
/75.L -—o— Cop/’54?
-—x— 57A/Nlé‘ss 571:“
\ .4. .57‘EEL
£ /5o~~ .
(f __._. PLUG, 77/91::
-—-- Ron 77705

if MES" \‘e
w ,
‘3.
Lu /
S 00 t
Q
3'
V 7.5T
“1
e
if 5&-
‘h
A
‘1‘

25»-

0

O

APPA/e’o V01. 7A6 E, Ila/7’; 4.6.

FIG. IS

23

 

6L<
WN VN. 0N

P b . p
.

we... T1

1

3 6Q

. ..§$R 0.33. w

3,

’r -.

 

 

£3 u... 11

win? .m.\ R n

 

.10: ON T.

 

1...... ONT

“3th wwweSEm 1T
quqou Io:

3.5.x. 9.33% we

.983.an Skuwum 0k meQ demmxuq 339$ mwflwfixq

 

J

22

t?

I? 0/09/7

§
:1

:1
[D4 ' 6377:?

 

21+

Q .9...

2.x . mes. Sunk 3N

 

NH... m...w.fi.e.~...m.._
3.6%. \

m; 0\o o u \\

$4. .m.
. .13. Q \\ \

n\x\x...A\. XV.
fl... 9.1.1.1.. -\..
.\..\\\

       

E-
MM»: .3 . ,
TNQQOU ndl

WQQQGRUSN mask 03$

1.... 3. 1E 1.81... e
h.w0\<eeooucumu.w Q med. QWVVNQXM 333 V beefing v

 

‘92

Q
no! {7:773:1de (7/069/

31
7

T

.532

luhm\

 

I

12:)

was then takcn. Fir. 18 shows these results. The reductions
increased W'th increased voltages. The maximum reduction
achieved in these tests was #.33 w.b. This was not sufficient
to lower the excrement to a pelletable level of less than 25%
w.b. S’ncc the reductions were much less than anticipated,
the method of determining the final moisture content proved

to be insufficiently accurate. rne error incurred by mixing

th sample and then taking a portion of the mixture for oven

(D

lr'inr was veri sienificant. It was estimated to be not less
Q 9)
than iifl w.b. Thus with the values achieved, errors greater

than 100% could have been caused by poor mixing.

...J

er of efficiency for the electrodes was the sane

s for the previous results reported, i.e. in decreasing order,

m

tainless steel, steel, and copper. The deviations from this

C0

order were within the error caused by poor sample mixing, i.e.
.i',".’.

A point of interest was that all samples with rod elec-
trodes showed less reduction at the 10 volt D.C. level tha
at the 5 or 15 volt levels. Repetition of the 10 volt tests
for rods showed the same tendency. The reason for this was not
known. It could have been a coincidence since the deviations
are well within the mixing error. This point should be studied
further, however, and thus the key may be found to solve the

problem of moisture retention.

m
C'\

Mo ASK/,9; "CW/754i 7 REM/C 7/ch
DUE 70 izécrxzmom/osu
Vs. Apia/go Var/ks:-

 

 

ScO'T
qr- /
~o' 4'0 " _, A
3 -. —
R0 ~~ —o—- COPPER
€ —x~— STA/N465: 57.5.51
3 3-0“ —A— 37254
R
\J’ -_ -——-— PZUG 71%;:
C.
62 -——-— Hoe Thai
L 2.0"
E;
S
6) \ O\
I
u, /A—- - .2
9S / \ \
R) / / \
L": 7‘...
2] wk / ~
g \V \ / /
\_ \ \ f -— o... /
\ \?~_——-/ X/ \ \
. v ‘7:1_* \ %
\\\ \Ido, / / /5 ‘22:
_ APPLIED \ V04 7A6; (flu/+3 dc.
\ /
-/.0 \V/
F IG. /8

 

.27

PROBLELS FOR FUTURE INVESTIGATION

This investigation demonstrated some points that should
be studied further. They were: 1. elimination of all gas
from the sample prior to and during the test run. This will
reduce the expansion and provide better electrode contact and
sample conductivity; 2. eliminate corrosion of the electrodes;
3. removal of dried material near the electrode to improve
conductivity; 4. use of multiple electrodes in a sample; and

5. study the 10 volt D.C. phenomena more fully.

S Ul-LI'IARI

This investigation was undertaken to extend the study
performed by Cross. Its objectives were: 1. reduce the
moisture content of chicken excrement to a pelletable level
of less than 25$ w.b.; 2. study the effects of increased
voltages; 3. evaluate the performance of two electrode types;
and 4. study the performance of electrodes made from three
different materials.

Appropriate data was collected, moisture contents deter-
mined, and graphs plotted. The average current flow rates
were determined by planinetering the current flow vs. elapsed
time curves and then dividing the result by the total elapsed
time.

From the resulting curves it was found that the plug type

electrodes gave larger current flows, higher temperatures, and

greater moisture content reduction than the rod types. Evalua-
tion of the three metals tested showed that stainless steel
gave higher currents, temperatures, and larger moisture content
reductions than steel and cepper. The steel electrodes, how-
ever, gave results very close to those of the stainless steel.
Copper was the poorest because it corroded more readily than
the others, thus reducing its conductivity.

The initial moisture content was 78 to 80% w.b. The maximum
moisture content reduction was 4.8% w.b. which was insufficient
to reach a pelletable level of less than 25% w.b.

Some points for future investigations were observed:

1. elimination of gas from the sample; 2. electrode corrosion;
3. removal of dried material next to electrode; and 4. use of

multiple electrodes.

HICHIGQN STQTE UNIV LIBRQRIE

illllllLlllllllllSLIIIllllllllllllllllllllllllllllillllllHlllHl

 

01096991 7