l

'1!

W“

l
i

Illl

05* ll

N-t
UDNOO

AIR AGETATEON OF MILK

 

I
_{
I

Thesis for “13 Degree cg M. 5.
IfiECEiGAI‘E "513% UNLVERSEYY
Ro‘bert C. NIiEkie

1958

H4. 351:3

This is to certify that the

thesis entitled

AIR AGITATION OF MILK

presented by

Robert Charles Milk ie

has been accepted towards fulfillment
of the requirements for

14.8. Agricultural

degree in? .
11.115 meerlng

JMM W

Major professor

Carl W . Hall

 

Date March LL; 1958

0-169

AIR AGITATION OF MILK

by

ROBERT C. MILKIE

A THES 15

Submitted to the School of Agriculture of Michigan State
University of Agriculture and Applied Science
in partial fulfillment of the requirements
for the degree of

MAhTBR OF SCIENCE
Department of Agricultural Engineering

1958

grU-Jf

«a». «z x 4/ 7

ACKNOWLEDGMENTS

With great pleasure the author expresses his sincere gratitude
to Dr. Carl W. Hall whose guidance and inspiration made the investi-
gation very rewarding. His patience, encouragement, and understanding
throughout the program made the work particularly enjoyable.

The Dairy Industry Supply Association is due great thanks
for the fellowship which made the undertaking of study for the
Master of Science degree possible.

Thankful acknowledgment is also due Dr. G. Malcolm Trout
of the Michigan State University Dairy Science Department, whose
personal guidance and technicalassistance was always a source of
encouragement.

The author feels deeply indebted to Dr. Arthur W. Farrall,
Head of the Department of Agricultural Engineering, for the
graduate research assistantship which allowed continuation of this
investigation.

Dr. W. James Harper of the Dairy Technology Department,

Ohio State University, is thanked for technical assistance and

cooperation on certain phases of the investigation.

ii

TABLE

INTRODUCTION . . . . . . . . .
REVIEW OF LITERATURE . . . . . .
Agitation Theories . . . .
Power Requirements . . . .
Physical Changes . . . .
EQUIPMENT 0 o o o o o o o a

Part A. Pilot Installation
Part B. Plant Installation

OF CONTENTS

Part C. Foam Formation Study . . . . .

EXPERIMENTAL PROCEDURE . . . .

Part A. Pilot Installation
Flavor Tests . . . .
Creaming Ability Tests .
Fat Redistribution Time 1
Vitamin C (Ascorbic Acid)
Rancidity Tests . . . .

Part B. Plant Installation

Test . . .

Part C. Foam Formation Study . . . . .

RESULTS 0 O O O O O O O 0

Part A. Pilot Installation
Flavor . . . . . . . . .
Creaming Ability . . . .
Fat Redistribution Time .
Vitamin C (Ascorbic Acid}
Rancidity Test . . . .

Part B. Plant Installation

Content . .

Part C. Foam Formation Study . . . .

DISCUSSION . . . . . . . . .
SUMMARY . . . . . . . . .
CONCLUSIONS .'. . . . . . . .
RECOMMENDATIONS FOR FUTURE STUDY
LITERATURE CITED . . . . . . .

APPENDIX

iii

PAGE

03-th

10

IO
15
23

25
25
25
26
27
28
29
3O

31
31
51
31
34
34
34
39
42
45

51

55

56

58

5.

6.

7.

10.

11.

12.

13.

LIST OF FIGURES

Pilot apparatus for air agitation

PAGE

of 5 gallons of milk . .

Lead line and distributing ring for 5-gallon pilot

agitator O O O O O O O O O O 0

Equipment for

agitating 5*gallon samples of milk . . . . .

Cut-away view of 2000—gallon storage tank with air flow and
distribution lines and air measuring device . . . . . .

ZOOOrgallon storage tank with air
lines (side view) . . . . . . .

2000-gallon storage tank with air
lines (front view) . . . . . .

Air measuring device installed on

flow lead and distribution

flow lead and distribution

EUOO+gallon storage tank.

Air flow lead line viewed from tank interior . . . . . . .

Air flow distribution line viewed
gallon storage tank . . . . . .

through manhole of 2000'

O O C O O O O O O O O O 0

Equipment used in studying the foaming effect 01 milk . . .

Bar graph (number of samples to reach constant fat test at
various times and various air flow rates) . . . . . . . .

Relationship of airflow rate with time required for
butterfat content of milk to reach equilibrium . . . . .

Action of air in milk at two levels . . . . . . . . . . . .

iv

12

13

14

17

18

19
20

21

22

24

35

3‘6

46

TABLE

5.

6.

LIST OF TABLES

EffeCt of airflow on development of flavors in S-gallon
samples of milk . . . . . . . . . . . . . . . . .

Percentage of cream on surface of milk after air
agitation o o o o 0‘. o o c o o o o o o o o o o o

Ascorbic acid content of milk after air agitation . .
Free fatty acid content of milk after air agitation . .

Fat redistribution tests for milk air agitated in
2000-88]. Storage tank 0 o o o o o o o o o o o o 0

Time required to form 9 in. of foam on a 48-in. column
of milk by air . . . . . . . . . . . . . . . . . .

Fat redistribution tests for milk air agitated in
5-gal pilot apparatus at .01 cfm/gal . . . . . . .

Fat redistribution tests for milk agitated in 5-gal

pilOt apparatus at ’02 Cfm/gal o o o o o o e o o 0

‘Fat redistribution tests for milk agitated in 5-gal
pilot apparatus at .03 cfm/gal . . . . . . . . . .

Statistical analysis of results appearing in
Tables A01, A02 and A03 9 o o o o o o o o o o o 0

Fat redistribution tests for milk agitated at 36°F . .

Fat redistribution tests for milk agitated at .04

cfm/gal air flow . . . . . . . . . . . . . . . . .

Pounds of air per feet of head for 1 cubic foot of air

PAGE

32

37

38

41

44

62

63

64

65

71

72

73

INTRODUCTION

The agitation of milk by air is not completely new to the
dairy industry, but the effect on the finished product by the
method has had little serious consideration. The nature of milk is
such that upon standing undisturbed for a period of time the butter-
fat rises to the surface, producing a non-homogeneous mixture. Once
this occurs, some means is required to redistribute the fat. The
early method was to take a ladle or paddle and simply stir the milk.
With high production and large plants this means was entirely in-
adequate. A mechanically-driven agitator was developed which would
produce the desired results of remixing the butterfat.

Mechanical agitators have been the most pOpular and most
widely utilized method for years.‘ However, the time element of
mixing and the power requirements are factors which are unfavorable
for an ideal means of agitating milk. Processors, in an attempt
to gain efficiency, have attempted to use air as the means of agi-
tation. One of the original problems with the use of air was that
ordinary air compressors caused lubricating oil from the compressor
to be carried into the milk during agitation. No satisfactory means
was deveIOped to overcome this problem until a carbon ring, self-
lubricating compressor was developed which eliminated one of the
major disadvantages of air agitation.

The determination of the effects of air agitation on the

physical and chemical properties of milk is essential. Also some

indication of the power requirements of this method is desired.

The aim of the research is to determine scientifically the
effects of air agitation on milk, and to ascertain if air agitation

is a feasible means of agitation on an industry-wide basis.

REVIEW OF LITERATURE
Agitation Theories

Neilson (1949) defined the word "mix", "to cause a promis-
cuous interpenetration of the parts of, as of two or more substances
with each other, or of one substance with others; to unite or blend
into one mass Or compound, as by stirring together; hence to
mingle; blend, as to mix flour and salt." Agitate is defined, "to
move with a violent irregular action; to set or keep in motion; as
to agitate water in a cup." Serner (1948) summarized these quite
thoroughly, but briefly, by relating that mixing is the desired
result, while agitation is the means for bringing it about. He
noted that motion alone does not accomplish much mixing, but from
the agitation created by motion, mixing does occur.

Peck (1955) related that the general direction of flow
alone does not achieve satisfactory mixing. An intense zone of
mixing must be created, and satisfactory direction and rate of
flow of the material must be maintained. Serner (1948) adds that
the first requirement of mixing is to set the liquid mass in
motion.

Serner (1948) noted that "to accomplish mixing it is neces-
sary to create shear planes between liquid layers." This may be
brought about by:

(a) application of a device that produces multidirectional flow,

(b) use of irregular shaped vessels, or
(c) use of baffles, coils, etc., to obstruct normal flow.

Peck (1955) related that the intense mixing zone is the area
where intense shearing is set up by differential velocity of the
material due to motion of the containers or some other means which
cause motion. He adds that to accomplish complete mixing the shape
of the container must be considered in order for all of the material
to pass through the shear zone.

Kaufman (1930) wrote that mixing by air agitation is largely
‘caused by the expansion of.air as it rises in the liquid and is de-
pendent upon the speed with which the air rises.

Kaufman (1930) and Quillen (1954) both said that better
agitation by air will be obtained from deep tanks because of some

of the previous reasons.
Power Requirements

Several workers have presented ideas regarding the factors
affecting power requirements for mixing and the quantity of air
flow required to cause adequate mixing.

Serner (1948) noted that the power needed to produce and
maintain continuous agitation of a confined liquid body is dependent
mainly on three types of factors:

(a) The amount of initial acceleration imparted to the liquid,
(b) The amount of energy loss caused by surface friction, and

(c) The physical characteristics of the material.

Cl

Kaufman (1930) said the degree of agitation by air depends
upon the quantity of air flowing, and also is dependent on the
velocity with which the air leaves the holes.

Quillen (1954) adds that the degree of agitation depends
upon the process and the liquid under consideration, as well as on
the quantity of air flowing and the velocity of air leaving the
orifice.

Storck (1951) showed a l/2-hp compressor agitated a 4,000
gal tank of milk with air in 2 1/2 min where ordinary mechanical
agitators ranging from 2 to 3 hp required 20 min for the desired
results on the same tank.

Quillen (1954) noted air may be used as an aid for mechanical
mixing to reduce power requirements. He adds that basically how-
ever, air agitation is inefficient; it does not compare favorably
with electrically driven mixers. Further, air agitation has a very
definite role in industry in the mixing of materials in shallow tanks.

Dunkley and Perry (1957) observed that the depth of milk
to width of milk ratio was very important as associated with time
required for complete fat redistribution. The greater the depth
as compared to the width, the more rapid the mixing time.

Implications can be drawn from several articles that air
agitation is very desirable in shallow vessels. The nature of
the process makes it possible to agitate to within a few inches of

the bottom of the tank.

- Physical Changes

Several investigators have approached the problem of air
incorporation or air contact, and its effect on some of the physical
properties of milk.

Sharp gt _1. (1940) noted that the oxygen content of milk
is associated with the ascorbic acid content. The oxygen content
of milk is determined by adding an excess of ascorbic acid to milk
and measuring the amount of ascorbic acid reduced by the oxygen.

The amount of ascorbic acid utilized in the reduction is prOportional
to the original oxygen content of the milk.

Opposing views have been presented on the effect of air on
the probability of oxidation of milk. Greenbank (1948) observed
that the oxidized flavor could be inhibited by aeration. Greenbank
(19ob) lound further that unless excessive'copper was present,
aeration and pasteurization would prevent the development of oxi-
dized flavor. Thurston gt 2;. (1936) showed that a form of aeration,
prolonged agitation, inhibited the development of the flavor. They
found that when milk was agitated for 2 1/2 hr the susceptibility
of oxidation-susceptible milk to oxidized flavor was greatly re-
duced. However, when the milk was agitated for less than 45 min
little effect on its susceptibility was shown.

Sharp gt El. (1940) demonstrated that oxidized flavors
could be largely prevented and the ascorbic acid content preserved
by removing the oxygen from the milk.

Guthrie (1946) showed the flavor of deaerated milk was ex-

cellent when fresh and good at the end of 7 days whereas the flavor

of un-deaerated milk was poor as a result of oxidized flavors after
7 days storage.

Brown 33 2;. (1936) found that milk passed over a surface
cooler and exposed to air showed no greater development of oxidized
flavor than did milk passed through an internal cooler.

Storck (1951) related that milk agitated by air for 50
min, then held in a tank for 24 hr, and reagitatedzfiw 20 min, showed
no evidence of oxidized flavor after holding the milk for 48 hr
after the latter agitation. .

Opposed to the above, Guthrie (1946) showed that one of the
factors causing oxidation of milk is the presence of air in the
milk. He stated that "One logical way to prevent the deve10pment
of the oxidized flavor in dairy products is to eliminate the oxygen."

Dunkley and Perry (1957) observed that incidence of oxidized
flavor was unrelated to the method of agitation whether it be air
or mechanical. These workers also noted that ascorbic acid destruc- _
tion was slightly greater at 61’F for samples agitated by air than
for those mechanically agitated. At lower temperatures there was
no difference in the ascorbic acid reduction even when the samples
were held at 39°F for 48 hr.

Jokay (1956) determined the intensity of air intake affected
the free fatty acid degree of milk in a pipeline installation and
high free fatty acid degrees were effectively currected by reducing
the air intake.

Thurston £3 31. (1936) noted that milk agitated for 150

min developed rancidity in all samples and some samples agitated

for 120 min also developed the flavor.

Dunkley and Perry (1957) determined that samples treated
at 399 to 50°F.by air or mechanical agitation did not produce ran-
cidity, but those samples exposed to air agitation when the samples
were at 61'F produced rancidity frequently.

Little information pertaining to the effect of air agitation
on the vitamin C content of milk could be found.

Herrington (1956) presented information which shows that
foaming, not violent agitation, was associated with rancidity.

'Tarassuk and Frankel(1955) observed that foaming appears
to be the necessary condition for activation of lipolysis by "air
agitation," with continuous mixing of milk and foam at temperatures
that keep the butterfat in a liquid state.

-The effect of air on the develoPment of rancidity in milk
is a problem that has received considerable attention in pipeline
milkers. Pasteurization at accepted minimum temperatures inactivate
the enzyme lipase which is the cause of rancid flavors.

Guthrie and Herrington (1956) noted that one of the factors
concerning rancidity was activation of lipase caused by air. They
further noted that air bubbling through raw milk, especially in a
pipeline, would possibly tend to break the fat globules into smaller
ones such as in homogenization, thus increasing the probability of
rancidity development. It is known that the mixing of raw milk
with homogenized milk greatly increases the action causing rancidity.

"Storck (1951) presented work by Dr. R. F. Holland of Cornell
University which showed, on limited tests, that the vitamin C content

remained the same at 17 mg/liter when the milk was agitated for 50

min. The milk was held for 24 hr, re-agitated for 20 min, tested,
and showed a dr0p of 4 mg/liter in the vitamin C content. According
to the author this was a normal dr0p for milk held for this period
of time.

Storck (1951) noted that this milk showed a drop of tempera-
ture from 41.5°F to 41°F for the original 50-min period of agitation.
After the 24-hr holding period the temperature rose 1°F after 20
min of agitation.

Dunkley and Perry (1957) showed that the overall heat trans—
fer coefficient for mechanically agitated milk was roughly 10 per-
cent greater than for air at a rate of 4.6 cfm per 1500 gal of milk
I in a 4.7-ft x 8.3-ft oval milk tank.

Dunkley and Perry (1957) showed that the size of the air
bubble formed at an orifice is a function of the velocity of the
air passing through the orifice and not a function of the diameter

of the orifice Opening.

10

EQUIPMENT
Part A. Pilot Installation

The primary problem was to determine the cubic feet of air
per minute required to give complete redistribution of the butter-
fat in a tank of raw whole milk which had been undisturbed for a
time sufficient to allow the butterfat to separate, forming a cream
layer. The effect of the volume of air flow on the flavor, vitamin”
C (ascorbic acid) content, and free fatty acid content were also of
major importance.

A small laboratory model with an agitator suitable for use
in a lO-gal milk can was developed to agitate small quantities of
milk. A flexible 0.5-in. inside diameter, plastic tube was used
as the air distributing section called the distributing ring. “The
length was such that when formed into a 100p, the loop was in the
center of the cross-sectional area of the can with half the volume
of milk on the inside and half on the outside.

A molded 0.5-in. plastic pipe Tee was used as a means of
joining the distributing ring in the form of a loop. The Tee was
connected to a rigid, .5-in. clear plastic tube, called the lead
line, which served as the means of getting the air to the distribut-
ing ring.

A rubber tube, with a U-tube manometer for measuring vari-
ation in static air pressure, connected the inlet line to a rotameter.

Figure 1 shows a diagram of the laboratory apparatus.

1}

A Brooks Rotameter model 1110 was used to measure the air
flow from .05 to .75 cfm at 70°F and 14.7 psia.

The Michigan State University Dairy's compressed air supply
was utilized in the research. To insure the air being delivered to
the milk was free from foreign material, a moisture and oil separa-
tor, manufactured by the Beach Precision Parts Company of Boonton,
New Jersey, was installed in the compressed air line.

To prevent complications due to materials which might
cause oxidation of milk, all milk contact parts were made from
inert plastics. The agitating equipment arrangement was simple to

assemble and disassemble for cleaning and sanitation.

0..

.sz so 2260 m .. cozozoo :o .2 38530 SE

._ 059....

 

 

240 v3.2 440-0.

 

 

 

 

 

mmkwfidkom mommwmaioo Ed

 

 

 

 

 

 

 

D 20
F

D:

 

 

 

 

 

 

Figure 2. Lead line and (in
pilot agitator.

. - .
h . .4 0'1 ,

{.‘2‘ ‘. J ‘
* "'. -_
; ‘x. --)’..- ’0. A” J
, J _ fluvks
,- -' -

Q‘“
\ .

.-

liguro 2. Load line and distributing ring for s-gallon
pilot agitator.

 

I. ... .3... .

. . ”3‘3

‘9 a

«4‘ Av JP.

Ids/64.1.? II.» . ti (7

 

Equipment for agit.‘

rim. 3.

 

Figure 3. Equipment for agitating Srgallon samples
of milk.

14

15

Part B. Plant Installation

To substantiate the results obtained on.a pilot installation,
a 2,000-ga1 rectangular milk storage tank was equipped with air
agitation devices. The equipment was installed in the receiving
room of the Michigan State University Dairy.

The assumption was made that the results obtained in the
pilot installation could be directly projected to the larger instal-
lation. The volume of air per gallon of milk was found to be a
constant quantity. Results of the pilot installation showed that
an air flow rate of .03 cfm/gal of milk gave satisfactory butterfat
redistribution in raw whole milk held for 24 hr before agitation,
.with no apparent harmful effects on the physical or chemical proper-
ties of the milk. As a result, 60 cfm of air for a 2000-ga1 tank
of milk was assumed to be adequate.

However, the initial tests conducted in the 2000—gal storage
tank proved this assumption to be invalid. An air flow rate above
20 cfm per 2000-gal of milk agitated the milk so violently and the
resulting foaming was so great that the assumption of .03 cfm of

air per gal of milk was discarded.

Eguipment

Stainless steel tubing, 1 l/2-in. diameter with l/S-in.
.holes on l2-in. centers, was used as-the air distributing device.
The tube was installed 4 in. from the storage tank floor with
the holes on the underside. The lead line, also of l l/2-in.

stainless steel tubing, connected the horizontal distributing line

16

through an observation port, with a rotameter placed on the obser-
vation platform of the storage tank.

Two Brooks rotameters with ranges of .05 to .75 cfm and 1
to 15 cfm, were used as the air measuring device and a dial type
pressure gage was installed in the air line between the rotameter
and the lead line.-

‘\The air source was the same for this phase of the experi-
ment as that described in the pilot apparatus section.

The equipment was installed in a 2000-ga1 rectangular
storage tank.' The tank was equipped with a mechanical agitator
which was not operated when the air agitatiOn equipment was in
use.. Figures 4through 9* show the air agitation set-up for a

2000'ga1 tank.

l7

 

 

 

 

 

 

Figure 4.

Cut-away view of 2000-gollon storage tank
with air flow lead and distribution lines and
air measuring device.

 

18

 

 

 

 

 

 

 

 

 

 

w m . ,
w l a t
L liltitn lllllllllllllllllllllllll
i n .5 mm.
. m
_ _ TI: —
_ _ :

_ __
. _ __
_.
H _ : .
_ _ : _
_ _ Z a
_ H ”u =
. _ Z =
_ . : =
_ _ air:
._ _ a: ._
_ _ : :
_ . __ __
_ . x __
.
u _ x _"
_ _ ._ __
_ n "n __
. . : __
. _ ._ __
. _ : __
. _ prams :
" ..fi..i.Hu.I.HHMHHHHHMMHHHHHHHHHHHHHHHHHILWK __
is. ............................. -f.
... m, . 4|. J
:L» p C »

 

 

OUTLET

 

 

 

 

lead

tank with air flow
(Side view)

and distribution lines.

2000 - gallon storage

Figure 5.

19

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.

2000 -gollon storage tank with air flow lead
and' distribution lines. (Front view)

 

 

 

Figaro 7. Air measuring device thr’ .;a
tank.

20

    
  

 

  
      

 

i

  

\ K IL-t
"Q! \r \' \
""-

i
,“|

#‘yd

\\ \- ‘

     

Figure 7. Air measuring device installed on 2000-gallon storage
tank.

Figure 8. Air {10' lead line i ...

 

 

Figure 8. Air flow lead line viewed from tank interior.

 

 

Figure 9. Air flow distix' '-
manhole of zht' .

22

 

Figure 9. Air {10' distribution line viewed through
manhole of 2000-gallon storage tank.

Part C. Foam Formation Study

A foaming problem was encountered when milk was agitated
with air so tests were conducted to determine the effects of various
air flow rates on the type and volume of foam.

. A 9-ft length of l l/2-in. glass pipe was placed in a
vertical position with an adapter on the bottom to allow air to be
put into the tube through various sized orifices. The orifice was
connected to a compressed air supply through a rotameter for air
measuring purposes.i A tape measure was fastened to the side of the
glass pipe to give a simple means for determining the amount of
foam on a specific height of fluid for a given air flow. Figure 10

shows this apparatus.

 

 

 

Figure 10. Equipment u».
foaming eitr.'

Figure Ice

(32((- 3:1 T, '

:g.’

. -: ~ .
" \‘\ .
-.

53“..) ‘ ..
.‘b‘~é‘,}~ ‘7‘; 4
II'. \

-_;AҤ~;-.f -.
l‘ :I on I'.‘ | II I'- II- I‘ U. ‘3

.—
.4

-.-r-

I
I
l
e
t ‘ D
I
.
l
.
r
u
I
e ’ .
I 4 t
' I
i
I! ‘ ‘
in t
. .‘ ,
I
n\-’
r ‘ a
1 i ,
u
"q I
. x ‘
‘,f . \
K: , 'V I
.' ‘ ’
4 .
.' :9.

\
\ ‘
,I\V‘
' ‘1’: a
V
I

 

Equipment used in studying the
foaming effect of silk.

f .1
(J1

EXPERIMENTAL PROCEDURE

The milk used in the research was raw whole milk received
at the Michigan State University Dairy. The milk was from bulk
tank pick-up and the milk received may have been from a feW‘hOUFS

to over 48 hr old before being subjected to research study.
Part A. Pilot Installation

Flavor Tests

The first tests were conducted to determine the effect of
air flow on the flavor of milk. Five gallon samples of raw milk
at approximately 40°F were exposed to air flow rates ranging from
.01 to .30 cfm/gal of milk. Samples were taken before exposure to
air agitation-and after 20 min, 1 hr, and 2 hr of air agitation.
Samples also were taken for creaming ability test. The temperature
of the milk was taken at the time of sampling. The flavor of the

milk was determined by organoleptic examination.

Creaming Ability Test

\ Tests were run to determine if air agitation affected the
creaming ability of raw milk. The test consisted of placing 100

ml of milk, agitated for the various times of 20 min, 1 hr. and 2 hr,

and a control sample, in IUU'ml graduated cylinders and holding

undisturbed, under refrigeration for 24 hr. After 24 hr the cream

volume '88 read directly in percentage of cream per volume of milk.

26

The milk used had been exposed to mechanical agitation in
farm bulk tanks and in the receiving room tank before samples were

taken.

Fat Redistribution Time

The next phase was to determine the volume of air flowing
through the milk required to give complete mixing. An agitation‘
period to give complete redistribution of the butterfat was the
objective. The equipment used in the previous portion of the ex-
perimental procedure was utilised. Five-gallon samples of raw un-
clarified milk from the storage tanks at the Michigan State University
Dairy were obtained and held undisturbed under refrigeration at ap-
pr0ximately 42°F for 24 hr and then exposed to various air flows
from .01 to .10 cfm/gal of milk. Duplicate control samples were
taken at 2-min intervals. The samples taken at approximately the
same location in the can each time, were obtained_by dipping with a
beaker a portion of milk from the surface. These samples were then
itested for butterfat content using the Babcock test. From these data
the time required to obtain a consistent butterfat test indicating
redistribution of the fat was obtained. After a minimum number of
tests, airflow rates of .01, .02, and .03 cfm/gal were found to give
results which approached an agitation time of 3 min, so the bulk of

the tests were run at these values.

27

Vitamin C (Ascorbic Acid) Test

After an airflow rate which would give complete redistribu-
tion of the butterfat in approximately 3 min was obtained, the re—
lation of the airflOw volume and the vitamin C content was determined.
The vitamin C content of dairy products is determined as the total
of the ascorbic acid content.

The test used to determine the ascorbic acid content is a
method developed by Shapr (1938) and is outlined in the Appendix.

Five gallon samples of raw milk from the Michigan State
University Dairy bulk pick-up were used far these tests. The
milk was agitated using the air agitator described earlier at an
air flow rate of .03 cfm/gal of milk which was determined to be the
flow rate adequate for proper mixing of 5 gal of milk. Control
samples of the milk were taken before agitation and the milk was
agitated for periods of 20 min and 1 hr. After each period of agi-
tation, samples were taken and tested immediately for the ascorbic
acid content, as were the control samples. The control samples,
20-min samples, and l-hr samples also were held 24 hr under refriger-
ation in dark brown bottles and again tested for the ascorbic acid
content to determine if there was a time lag involved with the ef-
fects of air agitation on the ascorbic-aCid destruction.

The milk which had been agitated for 1 hr was held under
refrigeration for 24 hr and again air agitated for 20 min at a
flow rate of .03 cfm/gal of milk. Samples were taken and tested

immediately for ascorbic acid content.

Rancidity Tests

After an airtlow rate which would give complete redistribu-
tion of the butterfat in approximately 3 min was obtained, the re—
lation of the airflow volume and the free fatty acid was determined.

Five gallon samples of raw milk from the Michigan State
University Dairy bulk pick-up were used for the tests. The milk
was agitated using the air agitator at an airflow rate of .03 cfm/
gal of milk for periods of 20 min, 1 hr and 2 hr. Samples of milk
from the same SUpply, held under the same condition with no air
agitation, were taken at the same time intervals and used as
controls.

The samples were shipped to Dr. W. S. Harper of the Ohio
State University Dairy Technology Department where the free fatty
acid degree was determined by the chemical procedure of Harper

t 1. (1954) and Thomas gt El. (1954).

Part B. Plant Installation

The procedure followed in conducting this phase of the re-
search was to agitate volumes of raw, whole milk which had been
held undisturbed overnight in the 2000-gal storage tank. Samples
were taken before air agitation and every minute thereafter through
11 min. The butterfat contents of these samples were then deter-
mined with the Babcock test. Tests were conducted at various air
flow rates until flow rates were obtained which gave complete fat
redistribution in approximately 3 min. The bulk of the tests were
then conducted at these air flow rates.

The effect of air agitation on the foaming and churning

of the milk was also observed and recorded.

30

Part C. Foam Formation Study

Brief tests were conducted on this set-up using orifices
of 1/16-in. and l/B-in. diameter holes. There was no apparent
difference in the bubble sizes from either method, substantiating
the work cited by Dunkley and Perry (1957). The opening sizes
studied were assumed to be the sizes most likely used in a plant
installation.

Raw whole milk at a temperature of 50°F was placed in the
tube to a height of 48 in. above the orifice, requiring roughly one
quart. The milk was exposed to an aixflow rate of .05 cfm and
the amount of foam and the time required to develOp the maximum
amount of foam was measured. The actual effects of the foam such
as bubble size, how foam is formed, and life of the foam were also
observed.

The procedure was repeated using .10 cfm, .15 cfm, .20 cfm,

and .25 cfm of air.

31

RESULTS
Part A. Plant Installation

Flavor

The flavor tests which were conducted as described in the
procedure section indicate that at a flow rate of .03 cfm/gal of
milk or lower, which is adequate to give satisfactory redistribution
- of butterfat, there is no effect on the organoleptic quality of raw
whole milk. At low air flow rates, of .01 to .10 cfm/gal of milk,
there was no change in the flavor of samples agitated for 20 min,
1 hr or 2 hr, when compared to samples of the same milk which had
not been exposed to air agitation. At high flow rates, above .10
cfm/gal of milk, the samples agitated for 2 hr had rancidity

development.

Creaming Ability

The results of the creaming ability tests indicate that air
agitation over a wide range of air flows has no effect on the cream-
ing ability of raw whole milk, even when the samples have been
exposed to .3 cfm of air/gal of milk for 2 hr. The reSults are below
the quantity of cream normally expected on the raw milk, but the
previous treatment of the milk is such that the creaming ability
has probably been affected. The milk has been agitated both in the
farm bulk tanks and the milk storage tank in the dairy before samples
were drawnforstudy. This previous agitation would have a tendency

to reduce subsequent creaming. (Table 2)

32

TABLE 1 '

EFFECT OF AIRFLOW RATE 0N DEVELOPMENT OF OFF—FLAVOR IN 5-GALLON
SAMPLES 0F RAW MILK AGITATED FOR PERIODS INDICATED.
SAMPLES HELD UNDER REFRIGERATION 24 HOURS
BEFORE BEING CHECKED ORGANOLEPTICALLY

Test Number Airflow Rate 'Agitation Time. Off-flavors

 

‘cfm/gal hr

2 A .3‘ 4 I rancid
3 . .3' a a 4 rancid
4 , .3‘ 2 rancid
5 .3" 2 I none

6 . 0:5 f ' 2 none

7 ‘ .03 I 2 none

8 .03 2 none

9 i .03 2 none
10 .03 I 2 none
11 .03 2 none
12 .03 2 none
13 .03 2 none
144 .03 2 none
15 .03 2 none
16 .03 2 none

 

°The air flows are considerably above the rate required for
fat redistribution.

33

TABLE 2

PERCENTAGE OF CREAM ON SURFACE OF MILK AFTER AIR AGITATION,
FOR PERIODS INDICATED, AT .03 CFM/GAL. SAMPLES
HELD UNDISTURBED UNDER REFRIGERATION FOR
24 HOURS AFTER AGITATION

——

 

 

 

 

Agitation Time

 

 

Test Number . Temp.
' Control 20 min 1 hr 2 hr
°F Percent Percent PerCent Percent

6 52 9.0 9.0 9.0 9.0
7 50 9.0 9.0 9.0 9.0
8 42' 10.0 10.0 10.0 10.0
9' 44 10.0 10.0 10.0 10.0
10 44 10.0 10.0 10.0 10.0
11 44 5.5 6.0 6.0 6.0
12 44 5.0 5.0 5.0 5.0
13 43 10.5 .10.5 10.5 10.5
14 43 10.5 10.5 10.5 10.5
15 43 10.5 10.5 10.5 10.5
16 44 10.0 10.0 10.0 10.0
°2 11.0 11.0 11.0 11.0
‘3 11.0 11.0 11.0 11.0
‘4 41.5 10.0 10.0 10.0 10.0
.5 43 9.0 9.0 9.0 9.0

 

°Airflow rate .30 cfm/gal

Law values of cream percentages are probably due to previous

agitation of milk in farm bulk tanks and dairy plant storage tank.

34

Fat Redistribution Time

Preliminary tests indicated that air flow rates of .01,
.02, and .03 cfm/gal of raw whole milk would give complete fat re-
distribution in approximately 3 min, so the bulk of the laboratory
tests were run at these flow levels. After approximately ten sets
of data for each of the air flow rates were obtained, a statistical
analysis was applied to the results. The calculations and data are
shown in Tables A-l through A44. These results indicate
a flow rate of .03 cfm/gal would produce complete fat redistribution
in raw whole milk which had been held undisturbed for 24 hr before

agitation. Figures 11 and 12 show these results.

Vitamin C (Ascorbic Acid) Content

The results of the tests conducted in reference to vitamin
‘C content appear in Table 3.. These tests show that air agitation
has no significant effect on the vitamin C content of raw whole
milk. The milligrams_of ascorbic acid per liter of milk is below
the value normally present in milk, but the previous history of the

milk must be the cause of the low values.

Rancidity Test

The results obtained from a private laboratory showed that
raw whole milk exposed to .03 cfm/gal of milk had no greater free
fatty acid content than did milk kept under similar conditions with
the exception that thecontrol milk was not exposed to air agita-

tion. These results appear in Table 4.

35

 

 

 

 

 

8
a?"
E
a 2. .
s . - a
. l l 1
z , , , ,
4 6 8 l0 l2 l4 l6
minutes
.0! cfm/gal.
3-——1 —-1

 

J

No. of samples
N
L

 

 

 

 

ill)

2 4 6 8 l0 l2 I4
minutes

 

.02 cfm / gal.

i

b
_1

"i

No. of samples
N 01
l L

l

 

 

 

2 4 6'
minutes

 

.03 cfm /gal.

 

Figure II. Bar graph (Number of samples to reach constant tat
test at various times and various air flow rates.)

 

36

€222.33 goo? 2 5.6 so 22:8 .2353
.2 .3253. 0E: 53 22 32:6 .0 aEmcozoEm

 

 

._oo\..£u .30.... :4
No.

 

 

223 23305 as... /
.23 2385 2.0

 

N.

w.

ON

sainugw - aunl

 

 

TABLE 3

AVERAGE ASCORBIC ACID CCNTENT OF THREE SETS
OF SAMPLES OF RAW WHCLE MILK EXPOSED TO
.03 CFM OF AIR PER GALLON OF MILK

 

 

 

 

Immediately after
agitation

24 hr after
air agitation

Milk agitated 1 hr,
held 24 hr,
reagitated for 20 min
and then tested
immediately

 

 

 

 

37

 

 

 

Control Agitation

0 min 20 min 1 hr

mg ascorbic acid per liter milk

11.4 11.5 11.3
7.3 7.3 7.3
7.4 7.3 -—

 

Low value of ascorbic
history of the milk.

acid content is a result of previous

TABLE 4

FREE FATTY ACID CONTENT OF RAW WHOLE MILK
AGITATED AT AN AIRFLOW RATE
OF .03 CFM/GAL 0F MILK

 

 

 

No. Agéizzion Fat Temperature Tisgfzion Ao
percent °F m1

control 4.1 36 0.65 1.585

2‘ control 9.1 35 1.05 1.155
3 20 min 3.7 37 0.55 1.489
4‘ 30 min 3.6 37 0.55 1.530
5 1 hr 3.65 40 0.45 .234
6‘ 1 hr 3.65 _40 0.50 1.369
7 2 hr 3.65 45 0.45 1.234
8‘ 2 hr 4.0 46 0.55 1.375

 

TSamples were from same milk supply, held under same conditions
as others with the exception they were not exposed to air agitatidn.

m1 n/1oo base - m1 n/100 base for blank test

Note: A;
12 A0 = Titration value

 

g fat in sample

39

Part B. Plant Installation

The results of this part of the work, which are shown in
Table 5, show that .50 cfm of air will redistribute butterfat in a
2000‘ga1 tank of milk in 2 min. When this flow rate was used, about
20 percent of the surface of the milk in the tank was covered with
foam to a depth of 2 in. The foam was very light and caused no
apparent effects on subsequent operations performed on the milk
such as standardization with a standardizing clarifier.

A higher flow rate of l cfm of air per tank gave butterfat
redistribution in 2 min, but the foaming was considerably more than
the amount produced by .5 cfm of air. When the milk agitated with
l cfm of air was passed through a standardizing clarifier, con-
siderable difficulty was encountered in adjusting themachine. The
reason for the trouble was not fully understood, but the history of
the milk used may have had some effect on the foaming characteris-
tics of this milk. ’

When an air flow rate of 1 cfm was used, there was a slight
amount of churning as determined by small butter flakes clinging to
the walls of the storage tank when the tank was emptied. No evi-
dence of churning was observed at .5 cfm per tank of milk. No
evidence of air incorporation was found in samples of milk agitated
at 1 cfm of air. ‘ C

When .25 cfm of air per 2000 gal of milk was used the air
volume was insufficient to cause air to leave all of the openings

in the distributing line. “The result was determined by observing

40

the surface of the milk.in the tank. Above each hole through which
air was passing there was a slight fountaining effect and air -
bubbles were formed at the milk surface. At a flow rate of .25
cfm, only the holes in approximately the front half of the dis-
tributing line had air flowing through them. When the flow rate
was increased to .5 cfm all Openings had air passing through them.

Computation of the horsepower reqhirements is shown and is
a purely theoretical approach. The head loss in the pipe, friction
loss due to air passing through holes and other factors are omitted.
These values are extremely small and would have little significance
on the final horsepower. A .25-hp motor is used on the mechanical
agitator on a tank identical to the 2000-gal tank uSed in the air
agitation study. A

hp = lb/min of air x feet head
33,000

05 x 7 ,
h . : .
P .33’000 00001

 

41

 

 

 

 

 

 

 

 

 

 

 

 

 

m.n a.» o.e mo.v m.n = as
m.» s.n o.e H.v s.n = om
as.» s.n s.n as.» mm.» o.¢ mo.e m.n = a
a.» o.n s.» as.» m.n 0.4 H.v na.n = m
s.» mm.» s.» w.n mm.» o.a H.e a.» = s
s.n o.n s.» m.» m.n mm.» no.e m.n = o
a.» ms.n s.n a.» m.n m.n mm.» mm.n no.e so.» = m
s.» m.» s.n m.n w.» mm.n m.n mm.» “.4 a.n = a
e.» o.n. s.n a.» m.n m.» mm.n a.» ~.v a.» = n
ne.n w.» , m.n m.n w.n m.» a.» ma.n a.e a.» = m
a.» a.» mo.n n.» a.» no.» mm.» m.» c.e e.n = a
o.m ¢.m mo.m m.~ w.m mm.m v.m m¢.m ma.m s.m use o
vowhdvm
O O O O I 0 .0 0 O 0 I O O O O O O O O O mm: 30”“ “fl“
a m a a m a s m a a m a _ a m a s a a a m a a m a a m a a m a umaca some»
. - moamfisn ease
ado m. ado mm. Eco m. sac m. sea m. ado a sea a see a ado H see a mans soacuaa
as as ow as. ea as ea as as as a. asos
coma coma ones coma coma cams news news coma omaa sad: so due
cm a m s o m a n a a conesz «was
mamas aaaqussa so so<ms>< se< maunmam .ons<sHe<

mH¢ OB UZHWOAXQ muchmm :mom OB mmw<d 2<mmo BOAA< 09 :GDOZH GmanBwHGZD GAB: mz<8 mummoem
A<wlooom 20mm wA<>GQBZH ZHSIH Ed 2:“ mflqzz<w be .mee xuoum<m wm DQZHXGHBQQ .BZHBZOU B<hmmafiam

m H4m<s

42,

‘Part C. Foam Formation Study

The results show that at airflow rates from .05 to .15
cfm per 48 in. of milk (1 qt) in a 1.5-in. glass tube, milk at a
temperature of 50°F developed a 9-in. column of foam as shown in
Table 6.

At flow rates above .15 cfm in the same apparatus the foam
rate was greatly accelerated and the height of foam rose to 42 in.
before foaming out the end of the column.

When a foam column of 42 in. was formed by an airflow rate
above .20 cfm it tack several minutes far the foam column to return
.to the 92in. length when the airflow rate was cut back to .05, .10
or .15 cfm.

When air bubbles left the orifice they tended to join with
other bubbles, until near the top of the milk column the bubbles
were about 4 in.-in length and nearly filled the cross-section of
the column. As the bubbles neared the surface of the milk they
collapsed allowing the milk carried above them to run down the
tube causing turbulent motion at or near the surface of the main
body of milk. The foam formed at this point was of a rather
permanent type and constituted the majority of foam in the column.
These bubbles were about l/32rin. in diameter at the milk surface
and increased in size to about 1/8-in. diameter at the top of the.
foam column.

I The air bubbles leaving the orifice either joined with

other bubbles in the movement through the liquid, eventually

collapsing as it neared the surface or remained as single bubbles
which rose to the milk surface and became a part of the foam column.
At flow rates slow enough so little or no bubble joining
took place in the movement through the liquid there was very little
or no turbulent motion at the milk surface and the foam was original
bubbles leaving the orifice. The foam column, in this case, was

about 1 1/2 in. in height and was very light.

44

TABLE 6

TIME REQUIRED TO FORM 9 IN. OF FOAM ON A 48-IN. COLUMN OF MILK
CONTAINED IN A VERTICAL 1.5‘IN. DIAMETER GLASS PIPE AT
VARIOUS AIRFLOW RATES. AIR INTRODUCED THROUGH
..125—IN. ORIFICE AT BOTTOM OF PIPE

 

 

t

 

 

Airflow Rate A . Time to Form 9 In. Foam
.05 cfm 30 sec
.10 cfm. 30 sec
.15 cfm 30 sec
.20 cfm I ‘ ‘
.25 cfm ‘

 

’Impossible to measure because column including milk and foam
mixture flowed out open end of pipe.

45

DISCUSSION

Air agitation in a liquid causes bubbles, issuing from holes
in a distributing line, to rise in a stream to the surface. These
bubbles form an air lift which lightens the surrounding milk and
causes it to rise. As the bubbles move upward, they move aside.
the milk above and impart velocity to the milk. ‘The milk rises
to the surface where the air bubbles leave and the retained velocity
carries the milk to a lower region in the tank.

Repetition of this motion sets up a pattern of movement
that insures complete mixing.

.Figure 13 shows the motion of liquid and air in a tank. The
velocity of the air and milk in a deep tank will be greater at the
surface as a result of increased buoyancy of the air causing more
rapid mixing than in a more shallow tank. The greater the width
of tank, the slower will be the mixing time because more milk in
a lateral direction must be moved by the air stream.

At a flow rate of .10 cfm/gal of milk, a rancid flavor is
produced. However, two things must be noted. First, the tempera-
ture increase in the milk was such that a natural inducement or‘
speed up of rancidity deveIOpment was possible, and second, the
flow rate which produced rancidity in the milk was extremely high.
The agitation of the milk was so violent and the foaming so exces-
sive that such a high airflow rate is impractical.’ If the air
agitating equipment is operated properly no rancidity or other

oft-flavors will be developed.

46

.226.

oz s .25 s

so Co c0304

.m. 239...

 

 

 

 

‘\
. \
\
\
N
O
O
O
’5
J
_/
/
/
/
_.——'

 

 

 

 

 

\Twhq mogumH/
\\\\\\\*oo wt/l/
sxx a.... .l//////'
.xx .I/o 0%Fm,_/..
.__ z/lo,o\\\ I...
+.. ///oo-\_\\ f.
_. #J40\\\ _._
_._ _:._.: .
I. .ZOJ: T:
I. _:o;i __
I. Z. i. _.
:. :11". _____
:. __: .
:. .__n.f: :.
Z. : :2 I.
:_ _ :_. __
.__ ____O“F, "V“.
., _«o _.

'p/ /(l\\\\fidprp///In\\\\\‘
///H(HH\ MJ//.l..l.\u.\
is

 

 

 

 

 

The values for the volume of cream on raw milk reported
herein are much lower than the expected value of 1/6 to 1/5 of
the total volume of milk represented as cream. The low values
probably are due to previous agitation of the milk. The milk was
from farm bulk tanks which had possibly been agitated a minimum of
two times before delivery to the dairy. This milk again may have
been subjected to agitation in the dairy storage tank. Frequent
mixing of milk when the butterfat is in a solid state will cause
the subsequent creaming ability to be reduced.

The vitamin C content of the milk was not affected by air
agitation at air flows suitable for fat redistribution which agreed
with work cited by Storck (1951).

The assumption that the ratio of air volume to gallons of
milk required for fat redistribution is a constant, was proven to
be invalid wnen projected from a pilot installation to a plant in-
stallation.

The relatively high value required for the pilot installa-
tion is most likely the rCSult of the low static pressure due to
milk above the Openings in the distribution line. The low head,
causing a low denSity bubble, and the short path of the bubble are
factors which discourage high velocity of bubble rise. Since air
agitation is a function of bubble size and bubble velocity, the con-
ditions in the pilot apparatus were unfavorable for air agitation
and thus for projection to larger and deeper quantities of milk.
The results which showed faster agitation and more mixing occur

when deep volumes of milk are used agree with work noted by haufman

(lUJO) and Quillen (1954).

48

The effect of air agitation on the physical and chemical
'properties of milk should be more closely associated with the air
to milk ratio. The redistribution of butterfat is a function of
the movement created by air and not by the quantity of air as such.
However the physical and chemical prOperties will be affected more
by the quantity of air. As a result theassumption may be made -
that the larger airflow per gallon of milk required in a pilot
installation would have a greater influence on the preperties of
the milk.' Because the flow rate suitable for air agitation in the
pilot installation had no harmful effects on the milk it is logical
to presume that the lower air volume per gallon of milk in the
plant installation would produce no unfavorable results.

When milk has remained undisturbed in a storage tank for a
long period of time the cream layer has a tendency to become
quite rubbery. A satisfactory method for breaking the cream layer
would be to impress a high airflow rate into the air agitating
equipment for a few seconds merely to break the layer and then re-
duce the airflow rate to a smaller value to complete the agitation.
When less‘than .5 cfm per 2000-gal tank was used, the air flow was
insufficient to cause air to leave all holes in the distributing
line. However, when a higher flow rate (.5 cfm or above) was used
to break the cream layer and start motion, all holes had air leaving
them when the flow rate was reduced to less than .5 cfm.

The airflow rate required for fat redistribution was a
function of the length of the tank being agitated for a particular.

tank cross-section. The milk path in a tank being agitated was

49

mainly perpendicular to the axis of the distributing line and each
air stream caused motion half-way to the adjacent air streams.

With air agitation the work required is only the work re-
quired to free the air in the milk. The air moving through the
milk due to its buoyancy does the work in putting the milk in
motion.

Work on the foam formation study apparatus shows that the
size of bubbles leaving an orifice is not a function of the orifice
diameter. This is in agreement with work presented by Dunkley and
Perry (1957).

From observations made in the foam formation study, higher
airflow rates cause considerable bubble clustering, thus creating
greater fluid movement and faster mixing.

Results of this work indicate that air agitation is a
highly satisfactory means of mixing raw milk. The general skepticism
of dairy men to purposely put air in contact with milk appears to
be unfounded. The air used in air agitation is only passed through
the milk to create motion and does not become a part of the milk
volume.

The installation of air agitation equipment should be ap-
proached with several points kept in mind. The distribution line,
3 to 4 in. from the floor, should run the length of the tank with
orifices along the entire length. The holes should be l/B'in.
diameter on lZ-in. centers. Another recommendation is that each
tank should have individual controls at the tank for regulating the

airflow rate. Run tests to determine relationship of airflow rate

which will redistribute the butterfat in the desired period of
time. The air flow should not be allowed to continue for excessive

periods of time.

SUMMARY

The flavor of raw whole milk was not affected by air agita-
tion at an airflow rate of .03 cfm per gallon of milk for 2‘hr in the
the 5-gal pilot apparatus. At an air flow of .3 cfm/gal for 2 hr,
rancid flavor was develOped. The creaming ability of raw whole
milk was not affected by air;low rates up to .3 cfm/gal for 4 hr
in the laboratory installation. The laboratory results showed that
.03 cfm of air per gallon of milk gave fat redistribution in a
mean time of 3.1 min. In the laboratory installation it was found
that the ascorbic acid content and free fatty acid content which
represent vitamin C content and rancidity development, respectively,
are not affected by .03 cfm/gal of milk when exposed to this air
flow for 2 hr. . I

In the plant installation, .5 cfm per 2000-gal tank was
sufficient to redistribute the butterfat in 2 min. At higher flow
rates (1 cfm and above) excessive foaming of the milk and slight
churning of the butterfat occurred.

An airflow rate of .25 cfm Was not sufficient to cause
air to leave all of the holes in the distributing line and as a
result very poor agitation resulted. .

An airflow rate of l cfm per 2000-gal tank caused no evi-
dence of air incorporation in milk.

The results of the foam formation study indicate_that .05

to .15 cfm of air through a 4-ft depth of milk (1 qt) will form a

specific volume of foam, but above these flow rates, the foam
formation is greatly accelerated. The foam is composed, only in a
small part, of the original bubbles leaving an orifice, but is com-
posed mainly of foam resulting from splashing and turbulent action
at the upper milk surface.

No apparent relationship exists between milk to cubic feet
of air per minute ratio for pilot apparatus and plant installation
when determining time required for complete butterfat redistribution.

The depth of milk above the air distributing appears to be

the important factor in air agitation.

53

CONCLUSIONS

1. The ascorbic acid content of milk is not affected by
air agitation for 1 hr inha pilot installation at an airflow rate
of .03 cfm/gal of milk.

2. Development of rancidity in milk as measured by the
free fatty acid content is not affected by air agitation for 1 hr
in a pilot installation at an airflow rate of .03 cfm/gal of milk.

3. The butterfat redistribution time for 5 gal of milk in
a pilot installation was approximately 3 min when the milk was ex-
posed to .03 cfm of air per gallon of milk.

4. No correlations exist between the airflow rate to milk
_ volume ratio between a 5-gal pilot installation and a 2000'gal
storage tank.

5. The creaming ability of milk is not affected by air
flow rates up to .03 cfm/gal of milk for a total agitation time
of 2 hr. I

6. The flavor of milk is not affected by an airflow rate
of .03 cfm/gal of milk for 2 hr in a pilot installation. At an air
flow rate of .3 cfm/gal, rancid flavor was produced on samples
agitated for 2 hrs.

7. The cream layer in a 2000-gal storage tank, unagitated
for 20 hr, will be redistributed in 2 min by .5 cfm of air.

8. 'The pressure in the air lines after the rotameter in the.

pilot installation is not great enough to register on a dial type

pressure gage o

54

9. The results of air agitation on the physical and
chemical properties of milk in a pilot installation can be pro-

jected to a large scale plant installation.

1.

55

RECOMMENDATIONS FOR FUTURE STUDY

Deve10p a milk foam theory using a transparent liquid

with milk characteristics.

2.
agitation.
3.

4.

Compare efficiency of operation of mechanical and air

Determine foaming relationship to churning.

Evaluate heat transfer rates of air and mechanically

agitated liquids.

5.
and Shapes
6.
agitation.
7.
commercial

8.

Determine air agitation requirements for various sizes
of milk storage tanks.

Conduct economical comparison of air and mechanical

Develop air measuring and timing device suitable for
use on milk storage tanks.

Determine the effect of air agitation on dairy products

after processing.

9.

at orifice
lO.
agitation.
ll.

separation

Study the possibility of using one-way air valves
point in air distributing line.

Evaluate CIP cleaning of tanks equipped with air

Determine minimum airflow rates for preventing

of butterfat.

56

LITERATURE CITED

Brown, W. C., L. M. Thurston, and R. B. Dustman. 1936.‘ Oxidized
flavor in milk. J. Dairy Sci., 19:753-760.

Dunkley, W. L. and R. L. Perry. 1957.. Mixing by air agitation in
horizontal milk tanks. J. Dairy Sci., 3Q: 1165-1173.

Greenbank, G. R. 1936. Control of the Oxidized flavor in milk.
International Association of Milk Dealers' Proceedings,
'39: 101-116.

Greenbank, G. R. 1948. The oxidized flavor in milk and dairy
products: a review. J. Dairy Sci., 31: 913-933.

Guthrie, E. S. 1946. The results of deaeration on the oxygen,
vitamin C, and the oxidized flavors of milk. J. Dairy
Sci., 23: 359-369. A

Guthrie, E. s. and B. L. Herrington. 1956.. Rancidity, problem of
farm tanks, pipe lines. Jersey Jour., g: 21.

Harper, W. J., E. W. Basset and I. A. Gould. 1954. Solvent
extraction procedure for determining the free fatty acid
content of homogenized milk. J. Dairy Sci., 23: 622-623.

Herrington, B. L. 1956. The control of rancidity in milk. J.
J. Dairy Sci., 32: 1613-1616. .

Jokay, L. 1956. Effect of variations in pipeline installations
on the acid degree of fat in milk. Master's thesis,
Michigan State University, East Lansing. 68 pages.

.Kaufman, B. L.- 1930. Determining air flow in agitation problems.
Chem. and Metallurgical Engr., 37: 179-180.

Neilson, W. A., ed. 1949. Webster's New International Dictionary,
2nd ed., 50, 1574. Springfield, Mass: Merriam Company.

Peck, W. C. 1955. The design of mixer. The Industrial Chem.,
12(55): 593-597.

(:1
\l

Quillen, C. S. 1954. Mixing - the universal operation. Chem.
Engr., 61(6): 178-218.

Serner, H. E. 1948. Simplified approach to mixing problems.
Chem. Engr., §§(1): 127-129, 136.

Sharp, P. F. 1938. Rapid method for the quantitative determination
of reduced ascorbic acid in milk. J. Dairy Sci., 21: 85.

Sharp, P. F., D. B. Hand and E. S. Guthrie. 1940. Preliminary
report on the deaeration of market milk. J. Milk Tech.,
3: 137-143.

Storck, R. E. 1951. Agitation of milk by air. The Milk Dealer,
13(1): 52, 127, 130.

Tarassuk, N. P. and E. N. Frankel. 1955. The mechanism of acti-
vation of lipolysis and the stability of lipase systems of
normal milk. J. Dairy Sci., 38: 438-439.

Thomas, W. R., W. J. Harper and I. A. Gould. 1954. Free fatty
acid content of fresh milk as related to portions of milk
drawn. J. Dairy Sci., 32: 717-723.

Thurston, L. M., W. C. Brown and R. B. Dustman. 1936. Oxidized
flavor in milk. 11. The effects of homogenization, agi-
tation, and freezing of milk on its subsequent susceptibility
to oxidized flavor develOpment. J. Dairy Sci., l2: 671-682.

58

APPENDIX

The procedure outlined below is from work presented by

Sharp (1938).

Vitamin C (Ascorbic Acid) Titration

Reagents:

A. The dye (2-6 dichlorophenol-indophenol)

B. Sulfuric acid (concentrated)

C. Ascorbic acid
Equipment:

A. Mortar and pestle

B. One liter volumetric flask

C. 500-m1 volumetric flask

D. 200-ml beakers

E. Burrette 10 ml calibrated in 0.05-m1 divisions

F. 100—m1 graduated cylinder

G. 50-ml graduated cylinder.

Place 0.2 grams of 2-6 dichlorophenol indophenol in a mortar
and pestle. Grind thoroughly and add 55 m1 of hot distilled water.
Grind further and decant through a filter (paper) into a one liter
volumetric flask. Add more hot water to the mortar, grind, and
decant, continue this process until the blue color has passed through
the filter. Adjust to room temperature and make up to one liter

with distilled water.

The Sulfuric Acid

Prepare a stock solution of 10 normal C. P. H SO4 by

2
diluting 285 ml of concentrated acid to one liter. For use,
dilute 10 ml of this stock solution to one liter. This is
approximately N/lO, and is used for acidifying and diluting the
milk for titration. Approximately 25 ml of this is added to

exactly 10 m1 of milk.

Stardardizing the Dye Solution

A. Weigh accurately approximately 50 ml of ascorbic acid;
transfer to a 500-ml volumetric flask and dilute to volume with
distilled water. It should be used within one to 4 min after
preparation. V '

B. Transfer 5 ml of the standard ascorbic acid solution
into a 200-ml beaker, add approximately 15 ml of the dilute sul-
furic acid and 15 ml of distilled water. Titrate with the dye
solution to a light pink color permanent for at least 30 sec;
record value.

C. Determine titration of an acid-water blank brought to
the same end-point. A blank of approximately 0.3 ml is usually
‘obtained. Subtract this from the value of (B).

D. Calculations. The amount of ascorbic acid in the 5-m1
sample divided by the number of milliliter of dye solution needed,
gives the milligrams of ascorbic acid corresponding to 1 m1 of the
dye solution. This value times 100 gives, in terms of milligrams

per liter, the amount of ascorbic acid corresponding to each

60

milliliter of dye solution required for the titration. The strength
of the dye solution should be such that this factors in the_1imits
of 5 to 8. I J

E. 1. Restandardize dye at 2 to 4-day intervals.

2. Discard dye solutions over two weeks old.

Titration of the Milk

A. Pipette 10 ml of milk into a 200-ml beaker. Add 25 m1
of the dilute acid and titrate at once with the dye solution.

B. On titration, the milk will assume a pink color upon
the addition of a small amount of the dye. However, upon standing
ya few seconds this color-will.fade. More dye may then be added.
The procedure is continued until the pink color persists for at
least 30 see. This is the end-point of the titration.

C. A blank determination should be conducted and this
blank value represents the residual titration of the milk and is
obtained by allowing milk to stand until the ascorbic acid disap-
pears and until constant value is secured. The blank may be deter-
mined on milk which has been stored in the cold for several days,
or on milk which has been treated with about 1 mg of capper sulfate
or exposed to sunlight.

D. Calculate and express results on the basis of milli-
grams of ascorbic acid per liter 0f milk.

1E. Precautions. 1._ Addition of acid to milk accelerates

the destruction of the ascorbic acid, so that the milk should be

‘titrated as soon as the acid is added.

2. The exposure of the milk to light should be reduced to
the minimum required for the titration procedure. Light markedly

accelerates ascorbic acid destruction.

62

:05 mm.o~ I memo» 005 no 0500 ommso><

mthufi
:wE 0H 0 w :05 m s08 ca :08 00. :05 OH :05 OH :05 0 :«E c ouodaaoo sou
. coswsaom 0508

 

on.» nu.m no.0 mb.n mN.n Om.n Ob.n on.» O@.n Om.» : ON

 

 

 

00.0 00.0 00.0 00.0 , 00.0 , 00.0 00.0. 00.0 00.0 00.0 e 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 = 00
00.0 00.0 . 00.0 00.0 . 00.0 .00.0 00.0 00.0 . 00.0 00.0 e 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 e 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0_ , 00.0 e .00
00.0 00.0 00.0 00.0. 00.0 00.0 00.0 00.0 00.0 00.0 e 0
00.0 00.0 00.0 00.0 00.0 ..0s.0 00.0 00.0 00.0 00.0 e 0
00.0 00.0 00.0 00.0 . 00.0 00.00 00.00 00.0 00.0 00.0 e 0
00.00 00.0 00.00 00.0 00.00 00.00 00.00 0.0 00.0 00.0 e 0
00.00 00.00 00.00 ., 00.00 00000 00.00 00.00 . 00.00 .00.0 00.00 . . . :05 0
.0.0 a .0.0 a .0.0m .0.0 a .0.0 a .0.0 a 00.0 a .0.0 ml, .0.0 a .0.0 a 00009

 

nnk «woe On% amok mm§ owes omfl «woe mmk amoe mufi umos bdfi amok mfi amoa wk ewes nfi 9x08 ouafidm mafia

5"“
In} i I

 

 

 

I l
{I

MAHZ ho A<U\Xho Ho. ho H9<m BOJEEH< 08 dszome mmommm m: «N mOh GHQGDBmHQZD GAE: 2AM: 34%
he A<U n Soak m4<>mmfizH ZHZIN B4 zmm<9 mmam2<m ho Bmfle ¥UOOQ<Q km GHZHZMMBQQ HZEBZOU 9<hmmeaam

H4 mags

 

 

63

:05 mm.o I 00000 000 son 0800 owmso><

 

 

 

 

 

 

 

 

 

 

 

 

 

:05 v :05 0 :0E OH :05 N :05 w :05 o :05 N :08 m :02 N :05 N :05 ea ouodmawu0mww
pohwsvos 0808

00.0 00.0 00.0 00.0 00.0 00.0. 00.0 00.0 00.0 00.0 00.0 = 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 0040 00.0 00.0 00.0 = 00
00.0 00.0 00.0 .00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 = 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 = 00
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0. 00.0 = 00
00.0 00.0. 00.0 00.0 00.0 .00.0 00.0 00.0 00.0 00.0 00.0 = 00
00.0 00.0 00.0 , 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 = 0
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 e 0
00.0 00.0 00.0 00.0 00.0 00.0 .00.0 00.0 00.0 00.0 00.0 : 0
00.0 00.0 00.0 00.00 00.0 00.00 00.0 00.0 00.0 00.0 00.0 = 0
00.00 00.00 00.00 00.00 00.00 00.0 00.00 00.00 00.00‘ 00.00 00.0 :02 0
.0.0 x .0.0 a .0.0 a .0.0 a .0.0 s .0.0 a .0.0 a .0.0 a .0.0 x .0.0 x 0.0 a

00 0000 00 0000 00 0000 00 0000 00 0000 00 0050 00 0000 00 0000 00 0050 00 0000 0.0000 o0aemmxum00

iIIIII'III

. mmqmz<w QP<UHAmDn ho mu<mm>< mm¢ meaawmm
24H: ho A<U\xhu No. he me<m Boqhmn< OB OZHmOQXQ mmoamn a: ma m0& @000 H< Gammabmmnza mam: KAH:
3<m mo 4<U n 20:& m4<>mmBzH ZHEIN B4 zmx<9 wmAAZ<m ho Hmma mooum<m wm szHzmMBmQ Bzmazoo thmmfiaam.

.0.< 00000

64

:05 O0.n l momou 000 000 0500 owmso><

 

 

 

 

 

 

 

 

 

 

 

 

w:0X05

:05 N :05 N :05 v :05 o :05 0 :05 0 :05 N :05 N :05 N 0000a50o mom
6000:0o: 0508

00.0 oo.n om.n n0.n oo.n oo.n nb.n om.n no.0 om.n : ON
on.n oo.n Om.n on.n Oo.n no.n ow.n om.n nw.n om.n : NH
on.n oo.n om.n ou.n nn.n no.n nu.n Cm.n Oa.n Om.n : o0
05.0 oo.n 0n.n nu.n nn.n no.n om.n Om.n Om.n Om.n : v0
on.n oo.n Om.n nb.n oo.n no.n n0.n om.n nw.n om.n : N0
on.n oo.n om.n n0.n oo.n no.n on.n Om.n om.n no.» : 00
00.0 oo.n no.0 on.» Oo.n oo.n oo.n Om.n nw.n Om.n : m
00.0 oo.n nm.n n0.n no.0 00.n om.n 00.0 00.» nn.n : o
oh.n oo.n nw.n no.0 n0.n n0.n on.n nn.n Om.n no.» : 0
00.0 oo.n oo.n nm.n 00.00 om.n om.n ow.n Om.n nn.n : N
no.n 00.00 00.00 .00.0N 00.n0. oc.n0 00.00 om.o Oo.o nn.0 :05 o
.m.m R .m.m 0 .k.m x .m.m X .h.m x .m.m fi .h.m R .&.m R ..m.m R. .h.m m

somda

00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000 00 0000. 0 0000 0 0000 000000 0500

 

 

.mmaa24m QB<UHALDQ ho mw5<mw>< mm< mBADmum .MJHZ ho
A<U\ZLU no. ho me<m 300k¢0< OB UZHmcmxm mmommm m: «N moa bovfi B4 cmmmDBwHDZS can: KAHZ @4013
B<m EC 440 n 20mm wq<>mm820 ZHZIN H< 2mm<8 mm432<m ho BmLH xuoom<m rm omszmmBma BZQBZOU B<hzm999m

n.< HAQ<B

65

TABLE A.4

STATISTICAL ANALYSIS OF RESULTS APPEARING
IN TABLES A01, [‘02 AND A03

 

 

 

 

.01 cfm/gal
_X. .3 .2 £2 _f62
4 1 -6 -6 36
6 1 -4 -4 16
8 l -2 -2 4
10 2 0 0
14 2 +4 +8 32
16 1 +6 , +6 36
r)
+2 =‘Ifd 124 = it‘d“
Assumed average = 10
z fd = 2;; = +025
f 8
arithmetic mean = 10.00 + .25 = 10.25
f ‘ f
0 PT”? 2 ‘ “
8 64
2","[1505 - .06 = \[15044
5" = _+_3.93 one probable error

(3

[‘0‘

ll
|+
x)
9
(p
05

two probable errors

I):

a: a: a to

10

14

.02 cfm/gal

I. 2

3 -4.00
1 ~2.00
3 0

1 +2.00
1 +4.00
1 +6.00
1 +8.00

 

Assumed average = 6.00

 

 

 

 

fd : -+6.00 = +.55
2..—
f 11.0 '
arithmetic mean = 6.00 + .55
\le f f
= 162 - 36
, 11 121
= 1407 -' 030 =
\ \
f1 = 13080
'6'} = £7.60

66

3:9 _rd‘°'
-12 48
-2 4
0 .
+2
+4 16
+6 36
+8 64
IE =5_'fd 162 =zrd2
6.55

one probable error

two probable errors

67

.03 cfm/gal

 

 

 

 

x f d fd fd2
+1.00 -5.00 5
+1.00 +3.00 3
+3.00 +3.00 9
2
+1.00 =zfd. 18 =Eifd
Assumed arithmetic mean = 3.00
z fd = +1.00 + +.11
f 9
arithmetic mean = 3.00 + .11 = 3.11
6“ =/Zfda - Efd 2
. f f .
= /18 -l_._:._-_-..1...
V 9_ 81
6"“1 = 11.42 one probable error

two probable errors

I:1
u
|+
N
(D
m

ANALYSIS OF VARIANCE

 

 

0
.01 cfm (x.) (x.)“ .02 cfm (y.)
1 1 1
6 min 36 14 min 196 2 min
4 " 16 2 4 2
10 100 2 4 2
10 100 8 64 4
14 196 2 4 4
16 256 36 6
8 64 36 4
14 196 12 144 2
82 :2x. 964 =23x.)‘ 10 100 _E
1 1 6 36 28 =§2i
nx :: 8 4 16
_ __ nz
= ‘ =§ 2
. 72 Zyi 64o ~1y1)
05’ = 11
Total sum of squares
2 2 2 2
23:1 +Zyi +221 - (21:1 +‘:.:yi 4» 2:21:
nx + ny + nz
964 + 640 + 104 - (82 + 72 + 23)2
8 + 11 + 9
964 + 640 + 10.1 - 33,124
28
1704 - 1182 426 = I

.03 cfm (2.)
1

4
4
4
16
16
36

Between sum of squares.

 

 

2 2

(2x1) + (gyi)

nx ny
(82)2 + (72)2 + (2

8 11

6724 + 5184 + 784
8 11 9
840 + 171 + 87 -

Degrees of Freedom

Total nx + my + nz -1

8 + 11 + 9 - 1 = 27

Between 2

Within 25

13.2 F

F = .99

exp

‘i . F
b nce exp > 1 ‘99

difference of population

 

69

 

 

 

 

 

( )2 ( r"
+ £21 - Exi + Eyi +2.21
nz nx + ny + nz
)2 - (82 + 72 + 28)2
8 + 11 + 9
- 331124
28
1182 = 216 II
Sum Square Mean Square F
420
216 108 108 = 13.2
8.2
204 8.2
= '7.77

we thvn can nonclude that there is a

with a risk of 1/100 of being wrong.

"t" test
x. y. z. t .93 (25) = 2.06
1 1 1
t .99 (25) = 2.79
r—‘ 'f 1 1"" 1é
C“ (x - z) = mean sum of squares for within 5' — + - '
.nx ny;'
w:.-.._..... 2’.“ ....... "1'“ ' '
: £862 )6 (P. +- )
V
= .j136/98 = ¢r1.78 = 1.34
r- M
X y
1.41m 95 = t .95 X fix-y : 2.06 X 1.34 : 2.76
2.76 <1\3.7" then x and y are different with a risk
of 1/20 of being the same.
- r‘ .' ,' '“ x x I \ '- I ~ .
' y " Z = I‘lIBOJ XLQ + 1711- : lb‘i/gi) '9 \- lat-)6 :3 1.4.19
Lfign : 6.50 - 3.10 = 3.40
exp
1:, "1.95 2 2006 X 1029 = 2006
2.66 <1 3.40. therefore y and z are different with a risk of

1/20 of being the same.

71

TABLE A.5

BUTTERFAT CONTENT DETERMINED BY BABCOCK TEST 0N SAMPLE TAKEN
AT 2'MIN INTERVALS FROM 5 GAL OF RAW MILK HELD UNDISTURBED
AT 36°F FOR 24 HR BEFORE EXPOSING TO AIRFLOW RATE OF
.03 CFM/GAL OF MILK. RESULTS ARE AVERAGE
OF DUPLICATE TESTS

j

 

 

 

Agitation Time Test 50 Test 51 Test 52 Test 53
% B.F. % B.F.' % B.F.' % B.F.
Control 13.0 13.40 13.65 13.40
2 min 4.0 3.95 4.0 4.0
4 " 4.0 3.95 4.0 4.0
6 " 4.0 3.95 4.0 4.0
8 " 4.0 3.95 4.0 4.0

10 " 4.0 3.95 4.0 4.0

 

72

ATABLE A.6

BUTTERFAT CONTENT DETERMINED BY BABCOCK TEST ON SAMPLE

TAKEN AT 2'MIN INTERVALS FROM 5-GAL RAW WHOLE MILK-
HELD UNDISTURBED AT 44°F FOR 24 HR BEFORE
EXPOSING T0 AIRFLOW RATES OF .04 CFM/GAL
RESULTS ARE AVERAGE OF DUPLICATE TESTS

 

 

T199 sample Test 8 Test 23 Test 27 Test 41 Test 43

 

 

 

 

Take" % B.F. % B.F. % B.F. % B.F. % B.F.
0 min '10.00 13.00 '20.60 7.10 13.20
2 " 4.15' 8.50 6.15 3.75 3.60
4 n 3.95 ' 3.50 3.80 3.70 3.60
6 " 3.85 3.50 3.65 3.70 3.60
8 H. 3.90 3.55 3.70 3.70 3.60
10 " 3.90 3.50 3.65 3.70 ‘ 3.60
12 " 3.90. 3.55 3.65 3.70 ‘3.60
14 " 3.85 3.50 3.70 3.70 3.60
16 " 3.85 3.35 3.70 3.70 3.60
18 " 3.85 3.50 3.70 3.70 3.60
20 " 'h3.9o 3.55 3.70 3.70 3.60
'Time required for
complete mixing 4 min 4 min 4 min 2 min 2 min

 

TABLE A.7

POUNDS OF AIR PER FEET OF HEAD FOR 1 CUBIC FOOT
OF AIR AT 30 PERCENT RELATIVE HUMIDITY

 

 

 

 

 

 

 

Ft. of Head 3 Lb. of air/cfm

 

.0858
.0884
.0910
.0934
.0959
.0985
.1009
.1035
.1060
.1086
.1110

@CJQOSMJACANHO

p.-
O

 

Date Due
. 7". 7:1

 

338%

D8111C0'293

 

 

"'TITI'ITH'IJHIIIEH‘I‘MTHfliflhflffllfit'lflfljfiflfl'flfifi'ES