the temperature of pasteurization' for butter making
TRANSCRIPT
U. S. DEPARTMENT OF AGRICULTURE,B U R E A U O F ANIMAL I N D U S T R Y . — C I R C U L A R 189.., -*~^
A. D. MELVIN, CHIEF OF BUREAU. . ; * . ' ' . . - " *
THE TEMPERATURE OF PASTEURIZATION'FOR BUTTER MAKING.
BY
L. A. ROGERS, Bacteriologist; W. N. BERG, Chemist;
and BROOKE J. DAVIS, Assistant, Dairy Division.
{Reprinted from the Twenty-seventh Annual Report of the Bureau of Animal Industry (1910).]
WASHINGTON:GOVERNMENT PRINTING OFFICE.
1912.
THE BUREAU OF ANIMAL INDUSTRY.
Chief: A. D. MELVIN.Assistant Chief: A. M. FARRINGTON.Chief Clerk: CHARLES C. CARROLL.Animal Husbandry Division: GEORGE M. ROMMEL, chief.Biochemic Division: M. DORSET, chief.Dairy Division: B. H. RAWL, chief.Inspection Division: RICE P. STEDDOM, chief; R. A. RAMSAY, MORRIS WOODEN,
and ALBERT E. BEHNKE, associate chiefs.Pathological Division: JOHN R. MOHLER, chief.Quarantine Division: RICHARD W. HICKMAN, chief.Zoological Division: B. H. RANSOM, chief.Experiment Station: E. C. SCHROEDER, superintendentEditor: JAMES M. PICKENS. m
DAIRY DIVISION.
B. H. RAWL, ChiefHELMER RABILD, in charge of Dairy Farming Investigations.S. C. THOMPSON, in charge of Dairy Manufacturing Investigations.L. A. ROGERS, in charge of Research Laboratories.GEORGE M. WHITAKER, in charge of Market Milk Investigations.ROBERT MCADAM, in cliarge of Renovated Butter Inspection.
n
CONTENTS.
Page.Introduction 307Method of handling cream in experimental work 307Bacteriological results of pasteurizing cream , 308Tests for the presence of enzyms 310
Peroxidase 312Catalase - 314Galactase 315
Method of measuring the activity of galactase in buttermilk 315Lipase 318
Method of measuring the activity of lipase in milk 319Discussion of enzym experiments 321
Scoring the butter 322Summary 325
in
ILLUSTRATIONS.
Page.FIG. 24.—Increase in water-soluble nitrogen in 100 c. c. of buttermilk, calcu-
lated to 83 days, digestion period 31825.—Effect of pasteurization temperature of cream on quality of storage
butter 32326.—Average scores of butter, showing effect of pasteurization of cream
at various temperatures . - 325IV
Historic, archived document
Do not assume content reflects current scientific knowledge, policies, or practices.
THE TEMPERATURE OF PASTEURIZATION FOR BUTTERMAKING.
By L. A. ROGEBS, Bacteriologist; W. N. BEBG, Chemist; and BROOKE J. DAVIS,Assistant, Dairy Division.
INTRODUCTION.
The pasteurization of cream for butter making has for its primaryobject the elimination of the normal bacteria of the cream to enablethe butter maker by controlling the ripening of the cream to securea uniform product. Incidentally, it may remove some of the possiblecauses of the deterioration of the butter, as well as destroy the patho-genic bacteria and expel some of the gases and other volatile flavor-giving substances.
There is no fixed standard for the temperature of pasteurizationin this country. In Denmark, where all cream used in butter mak-ing is pasteurized, a temperature of 82° to 85° C. (180° to 185° F.)is used, but in this country the cream is frequently heated to notmore than 63° C. (145° F.), and rarely above 77° C. (170° F.).
It is obvious that it is desirable to determine the most effectivetemperature at which cream should be pasteurized for butter mak-ing. This temperature is the one at which the greatest number ofundesirable factors are eliminated with the minimum effect on thecream itself. Several factors are involved in the determination ofthis temperature, among the most important of which are the uni-form destruction of a large proportion of the bacteria of the cream ;the destruction of the enzyms inherent in the milk; the avoidance*of imparting scorched, metallic, or other undesirable flavors to thecream; and the possible increased loss of fat in the buttermilk.
This paper gives the results of an investigation which had for itsobject the determination of the proper temperature for the pasteuri-zation of cream for butter making as indicated by the destruction ofthe bacteria and the enzyms and by the changes in flavor of the but-ter in storage.
METHOD OF HANDLING CREAM IN EXPERIMENTAL WORK.
The cream used came partly from milk separated at the creamerywith which the field laboratory is connected and partly from handseparators. It was all sweet and of fair quality. The pasteuriza-tion was done in a continuous Jensen machine. The temperature was
307
3 0 8 27TH EEPORT, BUREAU OF ANIMAL INDUSTRY.
controlled by a hand valve and determined by a naked chemicalthermometer inserted in the cream pipe near the machine. Thevariation was not over 1£° F.5 plus or minus. The cream was cooledat once to churning temperature and churned within three hours.This method was followed to avoid the complications due to poorstarters, to contamination during ripening, or other factors whichmight change the flavor of the butter and obscure the influence of thepasteurization temperature. These results, therefore, throw no lighton the effect of the growth of bacteria during the ripening of im-perfectly pasteurized cream. Obviously, also, it does not necessarilyfollow that similar results would be obtained from cream which wassour or otherwise fermented at the time of pasteurization.
BACTERIOLOGICAL RESULTS OF PASTEURIZING CREAM.
The results for the first season's work are given in Table 1 andthose obtained in the second season in Table 2. Bacteria were de-termined in composite samples of the pasteurized cream the firstseason, and in the second season the count of the raw cream^wastaken also. Lactose litmus gelatin was usually employed for plating,but in some cases lactose agar was used. The gelatin plates wereincubated at 20° to 22° C. (68° to 72° F.) for five days and the agarplates at 30° C. (86° F.) for three days.
The lack of any established standard of bacteriological efficiencymakes it difficult to draw conclusions from the bacteriological resultsalone. The number of bacteria remaining in the cream can not betaken as an absolute standard of the efficiency of the pasteurization,since this number varies not only with the temperature and time ofpasteurization but also with the number and kinds of bacteria presentin the cream before pasteurization. The percentage reduction isequally unreliable, as this also is largely dependent on the number ofbacteria in the cream before pasteurization. A large number of bac-teria in pasteurized cream may simply mean that the original creamcontained a very great number of bacteria, or that a large propor-tion were heat resistant. If the number in the cream before pasteur-ization was very high, a large number may be left in the cream afterpasteurization and the result still show a high percentage reduction.This is illustrated by the results obtained for sample 70 in Table 2,in which with nearly 1,000,000 bacteria remaining in the cream afterpasteurization a reduction of 99.5 per cent was obtained. The in-fluence of the kind of bacteria occurring in the cream is seen in a com-parison of samples 54 and 70, both pasteurized at 82° C. (180° F.)*In sample 54, 205,500,000 bacteria were reduced to 14,000, a reductionof 99.9 per cent, while with sample 70, 172,000,000 were reduced to940,000, giving a reduction of 99.5 per cent. A large proportion of
PASTEURIZATION OF CREAM FOR BUTTER MAKING. 309
the bacteria in sample 54 were evidently in the vegetative stage andeasily destroyed, while sample 70 contained many resistant spores.
The cream used in the first season's work was of good quality andabout half of the results were obtained in September when the bac-terial content was low. In the second season the proportion ofhand-separator cream was greater and its inferior quality was indi-cated by the higher number of bacteria in the pasteurized cream.In Table 1 fairly uniform results are shown at 71° C. (160° F.) andabove, and it is evident that 66° C. (150° F.) is too low to secureefficient pasteurization. The results given in Table 2 show efficientpasteurization at 77° C. (170° F.) and higher, and only fair resultsat 71° C. (160° F.).
It is evident from these results that with cream of good qualityefficient pasteurization from the bacteriological standpoint can besecured by momentary heating to 71° C. (160° F.). This is, however,near the lower limit of safety, and if the bacterial content of the rawcream is high a temperature of 74° to 77° C. (165° to 170° F.) mustbe used to secure uniform results.
TABLE 1.—Bacteria in cream after pasteurization—First season.
No. of sample. Pasteurizingtemperature.
Bacteriaper cubic,
centimeter.No. of sample. Pasteurizing
temperature.
747477777777777777797979797982828282828888
165165170170170170170170170175175175175175180180180180180190190
Bacteriaper cubic
centimeter.
1.2.3.4.445..7.9.11.42456.8..101240431315171921
150150150150150155155155155155155160160160160160160165165165165165
Number.1,525,000
960,0001,695,0001,630,000
54,000686,300331,500
1,099,000406,00043,00013,500
395,500686,500162,000189,50016,500
. 26,00050,50018,500
191,500244,00080,500
Number.61,00015,00046,50038,00078,500
145,00034,65023,30017,6006,05012,8007,9008,10024,60029,80014,9004,55033,35017,000124,0004,500
AVERAGES OF TABLE 1.
Number of tests.
556*7
Pasteurizingtemperature.
°C.66687174
°F.150155160165
Bacteriaper cubic
centimeter.
Number.1,172,800
449,560246,00095,800
Number of tests.
755.2
Pasteurizingtemperature.
°C.77798288
•F.170175180190
Bacteriaper cubic
centimeter.
Number.54,78011,91016,88064,250
310 27TH KEPOET, BUREAU OF ANIMAL INDUSTRY.
TABLE 2.—Bacteria in <
No.*of
4660474850616249586364515265
Pasteuriz-ing tem-perature.
°C.
' *60"6060606066666666717171
°F.
"
"iio140140140140150150150150160160160
Per cubiccentimeter
cream.
Number.6,900,000
92,000,0006,900,0004,470,000
230,000,00092,000,00086,500,0004,470,000
136,600,00086,500,00038,000,000
230,000,000180,000,00038,000,000
zream before and after pasteurization—Second
Bacteriaper cubic
centi-meter inpasteur-i zedcream.
Number.
74,500680,000
2,950,0007,250,000
18,400,000740,000775,000
1,155,000795,000
1,000,000100,000304,500
Per centreduc-tion.
Percent.
98.984.898.792.178.783.499.498.797.999.699.999.2
No.of
6656676853546970555771725973
Pasteuriz-ing tem-perature.
°C.7177777780828282888888889393
°F.160170170170176180180180190190190190200200
Per cubiccentimeter
cream.
Number.
45,900,000
122,000,000180,000,000205,500,000122,000,000172,500,000205,500,00045,900,000
172,500,00019,300,000
136,600,00019,300,000
Bacteriaper cubic
centi-meter inpasteur-
izedcream.
Number.1,950,000
160,00056,50011,00064,00014,00012,500
940,00024,500
324,60064,0009,500
23,50018,000
season.
Per centreduc-tion.
Percent.
99.7
99.999.999.999.999.599.999.399.999.999.999.9
AVERAGES OF TABLE 2.
5443
60667177
140150160170
83,774,00066,142,500
149,333,00083,950,000
5,870,900866,750838,60075,830
93.098.799.499.9
3i 4
2
828893
180190200
166,500,000110,825,00077,950,000
322,130105,52520,750
99.899. 899.9
TESTS FOR THE PRESENCE OF ENZYMS.
Certain enzyms occur normally in milk. When cream is churnedthese enzyms pass over into the butter. Their action is as yet un-determined, but it is possible that they take some part in the changeswhich occur in butter, even at the low temperatures of commercialstorage. This possibility makes it desirable that they be destroyedin cream used for making butter.
The following definitions of the milk enzyms are taken for grantedin this paper, no attempt being made to be rigorously exact from thebiochemical point of view:
Peroxidase: An enzym that oxidizes other substances by trans-ferring oxygen to them from some peroxid, such as hydrogen per-oxid.
Catalase: An enzym that decomposes hydrogen peroxid, formingwater and oxygen.
Galactase: The proteolytic enzym of milk.Lipase: An enzym that splits fats (or fatty esters) into free fatty
acids and the corresponding alcohols.Although normal milk (or cream) probably contains no oxidase,
this enzym was always looked for when testing for peroxidase. Upto the present time it has not been found in the materials tested.The naturally occurring oxidases are looked upon as mixtures ofa peroxidase and a peroxid, or as mixtures of a peroxidase and asubstance that can easily form peroxids.
Enzyms are unstable substances or agents and are easily destroyedby high heat. For convenience, the temperature at which an enzym
PASTETJBIZATIOST OF CREAM FOR BUTTER MAKING. 3 1 1
is destroyed is called its thermal death point, although an enzymbeing unorganized can not have a real death point. I t should beremembered, however, that the temperature at which any particularenzym can be destroyed is somewhat indefinite, the enzym beinggradually weakened by heat when the temperature approaches thedeath point or when the time of exposure is lengthened. Further-more, the exact temperature at which it is destroyed varies with theconditions under which it is exposed to the heat. The reaction ofthe medium, the presence or absence of the substrate, the amountof moisture present, etc., may raise or lower the thermal death point.
During the summer of 1908 tests for peroxidase, catalase, andgalactase were made on samples of cream. Formaldehyde solution(40 per cent) was added to the samples—1 or 2 liters of cream—inthe proportion of 1 to 1,000. After the addition of the preservativeportions of the sample were withdrawn for use and the remainderstored at room temperature. This method of preservation wasfound to be ineffective, since it did not prevent the growth of moldin many of the samples. Furthermore, the cream would not re-main uniform in its composition in spite of repeated attempts toredistribute the fat by mixing. On account of the large amount offat present chloroform in ordinary amounts could not be used. Al-though proper sampling at the end of the summer was impossible,the material was used to obtain approximate results for futureguidance. In this paper are incorporated only those results of thetests for peroxidase and catalase which were made in the beginningof the summer. By suitable controls it was easily ascertained thatthe results of the tests were not appreciably affected by the pre-servative used.
During the summer of 1909 the tests for peroxidase and catalasewere made on samples of cream obtained from the creamery soonafter pasteurization.
For the work on galactase the buttermilk obtained after the churn-ing of the different lots of cream was used. The samples of butter-milk were preserved at room temperature in stoppered bottles, 5 c. c.of chloroform being added to 1 liter of buttermilk. This concen-tration of chloroform has been found to have very little effect onthe proteolytic enzym of milk.1 Suitable bacterial counts showed thatit preserved the buttermilk very efficiently. Toward the end of thesummer tests for peroxidase and catalase were made on the samplesof buttermilk to confirm the tests previously made on cream and toascertain whether these enzyms were affected or destroyed duringthe changes that had meanwhile taken place. It was apparent from
1 Harding, H. A., and Van Slyke, L. L. Chloroform as an aid in the study of milkenzyms. New York Agricultural Experiment Station, Bulletin No. 6. Geneva, 1907.
22114°—Cir. 189—12 2
312 27TH REPORT, BUBEATJ OF ANIMAL INDUSTRY.
the qualitative tests for these enzyms that they were not yet verystrongly affected (see pp. 313, 315).
The results for lipase were obtained in the early part of 1910 in thelaboratories at Washington. Portions of the same sample of milkwere pasteurized in a small apparatus specially constructed for thepurpose, and the effect of the pasteurization on the activity of thelipase present was determined.
PEROXIDASE.
The method of testing for peroxidase was as follows: A smallamount—5 to 10 c. c.—of cream (milk or buttermilk or curd solution)was transferred to a test tube. Two or three drops of a freshly pre-pared alcoholic solution (about 10 per cent) of gum guaiac wereallowed to run down the side of the test tube so that the tincture re-mained on the upper surface of the liquid to be tested. The tube wasallowed to stand 5 to 10 minutes. The nonappearance of color indi-cated the absence of oxidase. Two or three drops of a dilute solutionof hydrogen peroxid (or old turpentine) were added, and if per-oxidase was present a blue color developed where the reagents camein contact. Usually a blue ring on or near the upper surface of theliquid was formed.
Suitable controls were made to insure that the results would not bevitiated by variations in the degree of acidity of the cream, by theamount of preservatives used, or other factors. The peroxidase andcatalase reactions were as strong in very sour raw cream or milk asin the original samples.
Results for peroxidase.—In Table 3 are summarized the results ofthe tests on the various samples of cream. Tests made on the cor-responding .samples of buttermilk when three months old gave prac-tically identical results. From the table it is obvious that whencream was pasteurized under the conditions described on page 307 theperoxidase reaction was not obtained in cream pasteurized at 79° C.(175° F.) or above; at 77° C. (170° F.) it was generally absent, andin cream pasteurized at 74° C. (165° F.) or below positive testswere always obtained.TABLE 3.—Showing destruction of peroxidase and catalase in cream by pas-
teurization.
Temperature ofpasteurization.
Raw.60666871747779828893
Raw.140150155160165170175180190200
Test forperoxidase.
+
4-
±
—
Test forcatalase.
+
- j -
Weak.Very weak.
—
PASTEURIZATION OF OBEAM FOB BUTTER MAKING. 3 1 3
In general, these results are in acoord with those obtained by otherinvestigators who pasteurized for comparatively short periods oftime. It must be borne in mind, however, that the death temperatureof an enzym is so strongly influenced by so many conditions that anexact agreement between the above results and those obtained byother investigators can hardly be expected. Thus, Herholz * foundthat milk heated one minute at 75° C. (167° F.) gave an uncertaintest with guaiac for peroxidase. Hippius2 obtained the same resultwith milk heated in exactly the same way. Wilkinson and Peters 8
found that at 78° C. (172° F.) the guaiac test for peroxidase becamenegative. They do not say how long the milk was maintained atthat temperature.
The question of how long the peroxidase can exist in butter underordinary conditions of storage is important, since the peroxidasepresent in butter, in spite of the absence of a peroxid, may be able totransfer slowly some of the inclosed oxygen to the several oxidizablesubstances present. It is possible that such slow oxidation may bematerially assisted by the presence of organic salts of iron or ofother metals and by the fine subdivision of the inclosed oxygen byoverworking the butter.4
The tests for peroxidase in buttermilk were repeated in June, 1910,when the samples were one year old. The reaction was given by onlyabout half the samples that had previously given it, clearly indi-cating the instability of the peroxidase. The disappearance wasprobably brought about by the slow digestion that had taken place.The peroxidase reaction in raw milk can be made to disappear inabout 20 or 25 days, by allowing the milk to sour, adding chloroformto eliminate subsequent bacterial action, and then allowing proteolysisby the lactic acid to go on, either with or without the aid of pepsin.
Peroxidase in raw-cream butter can be detected as follows: Meltthe sample at about 45° C. (113° F.), let the curd solution settle tothe bottom of the beaker, pour off the clear supernatant fat, and use5 to 10 c. c. of the curd solution for the test with guaiac, as usuaLAll the samples thus tested (five, fresh) gave positive reactions,,although the reaction appeared to be weak when compared with thesame reaction in raw cream.
1 Herholz. Beitrftge zu bisher bekannten Reaktionen zur Unterscheidung von roher underhitzter Milch mit besonderer Berticksichtigung der Guajakproben. Dissertation. P. 70,Table C. Braunsberg, 1908-
a Hippius, Alexander. Biologisches mir Milchpasteurisierung. Jahrbuch fur Kinderheil-kunde, Band 61, p. 375^ Berlin, X805.
•Wilkinson, W. Percy, and Peters, Ernst R. C. Eine neue Reaktion zur Unterscheidung:von roher und erhitzter Milch «owie zum Nachweise von Wasserstoffsuperoxyd in der MilcluZeitschrift fur Untersuchungen der Nahrungs- und Genussmittel. Band 16, Heft 3, pp.172-174. Berlin, 1908.
4 Rogers, Lore A. Fishy flavor in butter. U. S. Department of Agriculture, Bureau ofAnimal Industry, Circular 146, p. 12. Washington, 1909.
314 27TH KEPOKT, BUREAU OF ANIMAL INDUSTRY.
Five samples of butter that had been in cold storage for periodsranging from 1\ to 4£ years were tested for peroxidase. They werecontained in ordinary tin cans, and only the inner portions of thesamples were used because of the presence of iron rust on that part ofthe sample in contact with the can. All of these gave such unusuallystrong peroxidase reactions with tincture of guaiac and hydro-gen peroxid as to suggest an unusual cause. In marked contrast toraw cream, these curd solutions obtained from old butter whenboiled and cooled still gave the peroxidase reaction. Even repeatedboiling and cooling did not remove from this material the propertyof giving this reaction. Evidently the peroxidase in raw cream isdifferent from that present in these samples of old butter.
In view of the fact that the catalytic oxidation of guaiac resin andof other substances by metals or their salts has frequently beenobserved,1 it was natural to suppose that the peroxidase reactionin the boiled curd solutions was due not to the enzym occurring nor-mally in raw milk, but to some metallic compound of extraneousorigin. The old-curd solutions were qualitatively tested for iron byadding to the unconcentrated material a few drops of dilute hydro-chloric acid and potassium ferrocyanid or ammonium sulphocya-nate. Positive tests were easily obtained. Control tests showed thatmuch more iron was present in the curd solutions than in milk. Rawmilk or buttermilk containing approximately 0.1 per cent of ironadded as chlorid or sulphate will give the peroxidase reaction evenafter being boiled. It would seem reasonable to attribute the perox-idase reaction in the boiled curd solutions to their iron content.
Observations such as these may be of value in explaining the detri-mental influence of rusty milk cans, etc., on the flavor and keepingquality of butter.2
CATALASE.
The method of testing for catalase was as follows: Fifty cubiccentimeters of cream, milk, or buttermilk was introduced into a 100c. c. Erlenmeyer flask and 25 c. c. of commercial hydrogen peroxidsolution added. The flask was quickly closed with a rubber stopperprovided with a bent glass tube, the other end of which was insertedinto a fermentation tube (capacity 20 c. c ) filled with water. Ifraw cream were used, the oxygen liberated by the catalase rose inthe fermentation tube and completely filled it three or four times ina few minutes. With raw milk or pasteurized cream the liberation
1 Alsberg, Carl A. Beitrage zur Kenntnis der Guajak-Reaktion. Archiv fiir Experi-mented Pathologie und Pharmakologle. Supplementband 1908. Festschrift Prof. OswaldSchmiedeberg, pp. 39-53. Leipzig, 1908.
Colwell, Hector C, The catalytic oxidation of guaiac resin by metallic copper. Jour-nal of Physiology, vol. 39, pp. 358-360. London, 1909-1910.
2 Olson, George A. Rusty cans and their effect upon milk for cheesemaking. WisconsinAgricultural Experiment Station, Bulletin 162. Madison, 1908.
PASTEUKIZATION OF CREAM FOR BUTTER MAKING. 3 1 5
of gas was slower, until with cream pasteurized at high temperatureonly a few bubbles were obtained. even on long standing. By thismethod the distinction between raw cream and that in which cata-lase has been destroyed was easy to make. The above proportions ofcream and hydrogen peroxid were used because they were found tobe convenient; slight deviations from them do not detrimentallyaffect the results.
Results for catalase.—From the results (obtained in June, 1909)summarized in Table 3, it is apparent that the catalase in cream wasdestroyed when pasteurized at 70° to 71° C. (158° to 160° F.). Thisis a few degrees below the temperature at which peroxidase is de-stroyed. Confirmatory tests made in September, 1909, on the corre-sponding samples of buttermilk gave similar results. The tests onbuttermilk were repeated in June, 1910. None of the samples thengave the catalase reaction except No. 59, from cream pasteurized at93° C. (200° F.), which gave it strongly. Accidental bacterial con-tamination was the probable cause. It did not give the peroxidasereaction, showing that in this instance, at least, the two reactionsare due to two different substances or agents. This has been ob-served before.1
Catalase in appreciable amounts probably is not present in buttermade from properly pasteurized cream. As a possible factor influ-encing the quality of storage butter it may obviously be left out ofconsideration when butter is made from pasteurized cream. A dis-cussion of its possible action in raw-cream butter must be deferreduntil more data are obtained.
GALACTASE.
METHOD OF MEASURING THE ACTIVITY OF GALACTASE IN BUTTEBMILK.
To determine whether the galactase in cream was destroyed orpartly inactivated at the different pasteurizing temperatures, water-soluble nitrogen was determined in the buttermilk shortly afterchurning and after preservation (with chloroform) for an average of83 days at room temperature.
Into a 500 c, c. volumetric flask 50 to 200 c. c. of buttermilk wasintroduced. The same volume of buttermilk was used at the be-ginning and at the end of the period. Distilled water was addedup to about 400 to 450 c. c. One-fifth normal acetic acid was slowlyand carefully added until the casein separated completely in largeflocculi, leaving the supernatant liquid practically water clear. Foramounts of acetic acid used see Table 5. Distilled water was added tothe 500 c. c. mark, and the mixture was filtered on a 32 cm. foldedfilter (S. & S. No. 588 or 595) into a clean, dry 500 c. c. volumetric
1 Kastle, J. H. The oxidases and other oxygen-catalysts concerned in biological oxida-tions. U. S. Treasury Department, Public Health and Marine-Hospital Service, Hygieniclaboratory, Bulletin 59. Washington, 1910.
316 27TH REPORT, BUREAU OF ANIMAL INDUSTRY.
flask. During the filtration the funnel was covered with a well-fitting watch glass to minimize evaporation. The first part of thefiltrate was returned to the filter two or three times until the filtratewas freed from all particles of suspended protein. When 200 c. c»of buttermilk were used the filtrate was colored faintly yellow andwas not more opalescent than other solutions of equal protein content.In two 200 c. c. portions of the filtrate total nitrogen was determinedby the usual Kjeldahl method. The remaining part of the filtratewas measured and the total amount recovered noted. The figureswere recorded to make certain that the amount of filtrate recovereddid not vary enough to affect the results appreciably. Variations inthe amount of filtrate evaporated will introduce little or no error intothe calculations, if the same amounts of buttermilk are used at thebeginning and the end of the period. The average amount of filtraterecovered was 465 c. c ; both in June and in September, 1909, itvaried between 445 and 490 c. c. Some of the filtrate was of courseretained by the precipitate on the filter paper.
With a little practice the amount of acetic acid to be used in effect-ing the complete precipitation of the casein can be accurately deter-mined by noting the appearance of the precipitate and of the liquidin which it is suspended. The precipitation is very nearly maximalwhen the casein separates out in large flocculi that are suspended ina practically water-clear fluid. This is easily observable at the lineof contact between the upper surface of the liquid and the side ofthe containing vessel. When the precipitation is properly made fil-tration is rapid, requiring usually from two to four hours. It wasfrequently found convenient to allow filtration to go on overnight.The funnel is then lowered so as to close almost competely the receiv-ing flask. From time to time the filtrates were tested by the addi-tion of more acid or of alkali to make certain that the amounts ofacetic acid used were such as to give the maximal precipitation. Nadifficulty was experienced in correctly judging these amounts.
The number of possible sources of error in this determination islarge, and care must be taken to maintain uniform conditions whenmaking these determinations at the beginning and the end of theperiod.
The average of the two nitrogen determinations was multiplied byj : , as 200 out of 500 c. c. was used, and the result was called (total)water-soluble nitrogen. Obviously this is not absolutely correct,,because the volume of the filtrate is 500 c. c. minus the volume of theprecipitate. The figures for the increase in water-soluble nitrogen(Table 4) were obtained by subtracting the amounts obtained inJune from those obtained in the same way in September.
Results for galactase.—In all of the samples of buttermilk testedwater-soluble nitrogen increased very appreciably. The increase was
PASTEUEIZATION OF CREAM FOB BUTTER MAKING. 317
greatest in the buttermilk from raw cream and least in the buttermilkfrom cream pasteurized at 93° C. (200° F.)- Between 71° and77° C. (160 and 170° F.) this increase was markedly diminished,indicating that pasteurization between these temperatures stronglyinhibited the activity of the galactase. The proteolytic agent, pre-sumably galactase, was not completely destroyed by the pasteuriza-tion at any of the temperatures used.
As the pasteurizing temperature was raised the amount of water-soluble nitrogen in the fresh samples was appreciably diminished.All of the filtrates contained protein coagulable by heat in neutralsolution, so that the diminished content of water-soluble nitrogenmay be due, therefore, to a partial, but not complete, coagulation ofmilk proteins other than casein.
TABLE 4.—Influence of pasteurization on
Cream.
Sam-T»loNo
4 6 . . .4 7 . . .4 8 . . .6 8 . . .6 3 . . .6 2 . . .5 1 . . .6 6 . . .6 3 . . .6 4 . . .6 5 . . .5 7 . . .5 9 . . .
Pasteur-ized at—
°C.. Raw.. 60. 60. 66. 06. 71. 71. 77. 80. 82. 88. 88. 93
°F.Raw.
140140160160160160170176180190190200
the activity of galactase in buttermilk.
Water-soluble nitrogen in 100 c. c. of buttermilk expressed as—
Cubic centimeters ofN/5 nitrogen.
June.
c. c.41.840.886.737.536.933.435.728.633.428.023.627.4-24.0
Sept.
c. c.99.590.381.382.180.287.885.459.266.658.642.051.542.5
In-crease.
c c.57.749.645.644.643.354.449.730.633.230.618.424.118.5
G
June
•ams of nitrogen.
Gram.0417
.114
.100
.105
.103
.094
.100
.080
.094
.078
.066
.077
.067
Sept.
Gram.0.27V
.258
.228
.230
.225
.246
.239
.166
.187
.164
.118
.144
.119
Increase.
Gram.0.162
.139
.128
.125
.122
.152
.139
.086
.093
.086
.052
.067
.052
Per cent of total nitrogen.
June.
Percent.25.324.721.722.722.420.321.717.320.317.014.316.614.6
Sept.
Per cent.60.354.849.349.848.653.351.835.940.435.525.531.225.8
Increase.
Per cent.35.030.127.627.126.233.030.118.620.118.511.214.611.2
Total nitrogen in 100 c. c. buttennilk=0.462 gram.
TABLE 5.—Analytic data obtained' in Tableperiod.
calculated to average digestion
Cream.
SampleNo7
46474858636251566364665769
Pasteurizedat—
*CRaw.
606066667171778082888893
•F.Raw.
140140150150160160170176180190190200
Increase in water-soluble nitrogenin 100 c. c. of buttermilk calcu-lated to 83 days digestion period,expressed as—
C. c. of N/5nitrogen.
c. c.56.348.345.145.144.950.245.331.733.231.418.925.018.7
jGtramsofnitrogen.
Gram.0.158.136.126.127.126.140.127.089.093.088.053.070.052
Per centof total
nitrogen.
Percent.34.229.427.327.427.230.427.519.320.119.011.415.211.4
Digestionperiod.
Days.85858482809091808381818082
Volume> ofbutter-milkused.
c. c.5050
100200200150150200150200200200200
Volume of N/5acetic acid usedfor precipitation.
June.
c. c.15132744414134444044444439
Sept.
c. r.99
2017171413202120172817
318 27TH REPORT, BUREAU OF ANIMAL INDUSTRY.
According to Babcock, Russell, and Vivian * " it is apparent thatheating the enzym solutions for 10 minutes at 76° C. suffices to de-stroy the digestive ferment galactase, and even at 71° C. for the sameexposure its action was materially reduced." Hippius2 found thatthe proteolytic enzym of cow's milk can withstand an exposure to65° C. (149° F.) for one-half hour and is not destroyed till near100° C. (212° F.), at which temperature a short exposure is sufficient.In the light of such results it does not seem remarkable, therefore,that cream pasteurized at 93° C. (200° F.) for about 30 secondsshould still contain active galactase. I t is of course possible that inpasteurized milk or buttermilk that has been preserved for longperiods of time the observed proteolysis is due not alone to unde-stroyed galactase, but to the hydrolytic action of water as well.
€O
58 €3 52 5/MO A5O /SO /CO /6O /70co ee ee 7/ •?/ 77 GO
S7/ao /so /9O 2oo°,re2 ae ea 33 °c
FIG. 24.—Increase in water-soluble nitrogen In 100 c. c. of buttermilk, calculated to 83days digestion period. (Graphic representation of analytical data in Table 5.)
LIPASE.
Portions of the same sample of raw; milk were pasteurized at dif-ferent temperatures and the activity of the lipase present determinedby titrating the amount of acid liberated by it from ethyl butyrate.The apparatus in which the pasteurization was done consisted of a2^-liter acid bottle containing the milk to be pasteurized. This wasinverted over a 5-inch funnel connected by means of rubber andglass tubing with a coil of glass tubing (obtained from two verticalcondensers) immersed in a pan containing hot water. From the coilthe milk was passed through two ordinary Liebig condensers.
About 3 liters of cold raw milk as received from the dealer wastransferred to two large beakers and heated in a hot-air bath to 40°
1 Babcock, S. M., Russell, H. L., and Vivian, Alfred. Properties of galactase : A digestiveferment of milk. Wisconsin Agricultural Experiment Station, Fifteenth Annual Report,1898, p. 82.
2 Hippius, Alexander. Biologisches zur Milchpasteurisierung. Jahrbuch ftir Kinderheil-kunde, Band 61, p. 380. Berlin, 1905.
PASTEURIZATION OF CREAM FOR BUTTER MAKING. 319
to 45° C. (104° to 113° F.), requiring 20 to 40 minutes. The milkwas transferred to a 2f-liter bottle and inverted over the funnel con-nected with the pasteurizer. The water in the pan having been pre-viously heated to 90° to 93° C. (194° to 200° F.), the milk wasallowed to flow through the glass coil in which the pasteurization tookplace. The temperature of the milk as it left the coil was indicatedby a thermometer placed in it for the purpose. When the r&ilkentering the coil was at 45° C. (113° F.) and the water in the panat 93° C. (200° F.) the outflowing milk was at 82° C. (180° F.).The burner under the pan was lowered or turned out, according tocircumstances, and as the temperature fell the receivers (300 c. c.flasks) at the end of the condenser were changed so as to receive ineach one of them the milk that had been pasteurized over a range oftemperature not exceeding 2° C< (3.6° F.). In some of the experi-ments the range was but 1° C. (1.8° F.). The amount of milk re-ceived in each flask varied from 150 to 250 c. c. When the pasteur-izing temperature was highest the milk as it left the condenser wasat 45° C. (113° F.). The receivers were placed in ice and salt for afew minutes to cool the milk still further.
The height of the 2^-liter containing bottle above the coil, etc.(approximately 60 cm.), was so adjusted that the milk passed throughthe apparatus in about the same time as it does in the " flash " pas-teurizer, i. e., 15 to 30 seconds. The pasteurization of 3 liters ofmilk and its collection in 9 or 10 separate receivers, with the rejec-tion of milk pasteurized at#intermediate temperatures, was accom-plished without difficulty in 13 to 14 minutes. In general, the pas-teurization was conducted as closely as possible under the same condi-tions as existed in the creamery at the field laboratory. The severalseries of samples of pasteurized milk were tested for the peroxidasereaction, using tincture of guaiac and dilute hydrogen peroxid solu-tion. The peroxidase reaction disappeared at 77° to 79° C. (170°to 174° F.), indicating that in this respect at least the pasteurizationin the two places had been carried out in substantially the same way.
METHOD OF MEASURING THE ACTIVITY OF LIPASE IN MILK.
Into each of two 100 c. c. Erlenmeyer flasks a 50 c. c. portion ofthe sample was transferred from a measuring cylinder, beginningwith the sample that has been pasteurized at the highest tempera-ture. One-half of a cubic centimeter of chloroform and 5 drops ofphenolphthalein solution were added to all the flasks. To one ofeach pair of flasks 0.5 c. c. of ethyl butyrate was added, the other,containing the milk, chloroform, and phenolphthalein mixture, beingused as a control. The contents of all the flasks were immediately
' titrated to a distinct and uniform pink with tenth normal sodiumhydroxid, after which the flasks were tightly stoppered with rubber
320 27TH REPORT, BUREAU OF ANIMAL INDUSTRY.
stoppers. This titration gave the amounts of alkali required to neu-tralize the acidity of the milk, ethyl butyrate, etc. The flasks wereset aside (a thermostat at 26° C. [79° F.] was convenient for thispurpose) and rotated about once in 12 hours. At 24-hour intervalsthe liberated acid was titrated, both in the ethyl butyrate mixturesand in the controls, to a distinct and uniform pink, the color fromthe~previous day's titration having been discharged by the acid lib-erated. In titrating the controls at the beginning of an experimentapproximately 11 c. c. of tenth normal sodium hydrate was requiredto bring the mixture of milk and chloroform to a reaction faintlyalkaline to phenolphthalein. By the following day the pink colorhad disappeared, and the reaction could again be made alkaline bythe addition of approximately 1 c. c. of tenth normal sodium hy-droxid. In this way it was ascertained that in 24 hours about 1 c. c.of alkali was in some way consumed, probably by slow combinationwith some constituent of the mixture. If more than 1 c. c. of alkaliwas added, the pink color persisted for a longer time; thus, afterthe addition of an excess of 3 c. c. of alkali, the control remainedpink for over 8 days. In mixtures containing milk pasteurizedat 74° C. (165° F.) or over (and presumably no lipase) and ethylbutyrate, chloroform, etc., the pink color disappeared more rapidly,which indicated that, in addition to the alkali consumed as describedabove, a small amount of alkali was used up in hydrolyzing ethylbutyrate.
The figures in the last column of Tabte 6, headed " Corrected totalacid," are the figures in the preceding column minus 6.6 c. c, whichrepresents very closely the amount of alkali consumed in the twoabove-described ways; i. e., the correction used here represents theamount of alkali consumed in those mixtures in which the lipase wasapparently destroyed, and which also contained ethyl butyrate.
Results for lipase.—The figures in Table 6 are typical of threeother series giving similar results. It is evident that the lipase wasdestroyed very near 70° C. (158° F.). It is possible that it was nottotally destroyed until a few degrees above this temperature, but themethod did not permit the distinction between the very slow hy-drolysis of ethyl butyrate by the small amounts of alkali presentor by the weakened lipase possibly present in the milk pasteurizedabove 70° C. (158° F.). An attempt to strengthen the lipase by theaddition of an excess of alkali showed this to be inadvisable, becausethe excess of alkali tended to hydrolyze the ethyl butyrate directly.
The destruction of lipase at somewhat lower temperatures wasobserved by Terroine,1 who found that pancreatic juice lost its lipo-lytic power when heated for 10 minutes at 65° C. (149° F.), and by
1 Terroine, Emile F. Zur Kenntnis der Fettspaltung durch Pankreassaft. I. Bioche-mische Zeitschrift, Band 23, p. 424. Berlin, 1910.
PASTEUBIZATION OF CREAM FOR BUTTER MAKING. 321
Hippius,1 who found that the lipase in human milk was destroyedwhen exposed for a short time to a temperature of 64° C. (147° F.)
TABLE 6.—Influence
Milk pasteurizedat—
•cRaw.
55576063666870
. 747780
•FRaw:
131135140145150155158165170176
N/10 so-dium hy-
droxidrequiredfor neu-
tralizationof milk,
etc.
c. c.12.912.312.512.312.612.412.412.312.412.413.1
of pasteurization on the activity of lipase in milk.
l day .
c. c.10.27.87.77.66.33.72.52.21.71.91.6
N/10 acid liberated at end of—
2 days.
c. c.6.25.15.25.03.92.81.91.61.21.21.0
3 days.
c. c.4.33.84.03.43.02.01.51.21.11.11.3
4 days.
c. c.3.02.82.52.52.11.71.31.21.21.11.0
7 days.
c. c.3.33.53.63.42.82.21.91.91.71.61.7
Totalacid lib-erated.
c. c. N/10.27.023.023.021.918.112.49.18.16.96.9
. 6.6
Correctedtotalacid.
r r N/10.20.416.416.415.311.55.82.51.50.30.30.0
DISCUSSION OF ENZYM EXPERIMENTS.
In general, the results obtained on the thermal death points of themilk enzyms are in accord with those obtained by other investigators*But close comparisons with their work could not always be made,because the pasteurization process was seldom carried out by twoinvestigators in exactly the same way or for the same length of time.Frequently their descriptions were so imperfect in detail as to makea reproduction of their work, were it desired, practically impossible.For example, one investigator omitted to state how long the pas-teurization lasted. In addition, variations in the methods of testingfor these enzyms make comparisons still more difficult.
It has been shown that in the pasteurization process catalase andlipase are destroyed at 70° to 71° C. (158° to 160° F . ) ; peroxidasewas destroyed at 77° to 79° C. (170° to 175° F . ) ; and galactase,though strongly inactivated between 71° and 77° C. (160° and 170°F.), is not totally destroyed even at 93° C. (200° F.).
It is highly probable that under the pasteurizing conditions pre-vailing at most creameries that use pasteurized cream for buttermaking the first two of these enzyms are destroyed. Therefore, aspossible factors influencing the keeping qualities of butter madefrom pasteurized cream, these may be left out of consideration. Buton account of their higher thermal death points it is probable thatperoxidase is not always destroyed and galactase is rarely, if ever,completely destroyed. Consequently both of these enzyms may beconsidered possible factors in the deterioration of storage butter.
1 Hippius, Alexander,kunde, Band 61, p. 377.
Biologisches zur Milchpaeteurisierung.Berlin, 1905.
Jnhrbuch fiir Kinderheil-
322 27TH REPORT, BUREAU OF ANIMAL INDUSTRY.
It has been shown that both of these enzyms are present in butter-milk obtained from cream pasteurized at high temperatures. Re-searches are now in progress to determine whether the peroxidase isengaged in slowly oxidizing one or more of the constituents of thebutter or in assisting in such oxidation. The figures in Table 4 showthat the galactase remaining undestroyed in the buttermilk con-tinued to hydrolyze the proteins present. The possible activity ofgalactase in storage butter is likewise being studied.
It is obvious that a similar study of the action of the other twoenzyms would be of interest because of their possible influence on thekeeping quality of raw-cream butter.
It must not be inferred that the writers assume a relation betweenthe milk enzyms present in butter and its keeping qualities. Theresults thus far obtained, however, indicate that the probability ofthe existence of such a relation is great enough to warrant the studiesalready undertaken along these lines. The present paper, on thechemical side, is therefore preliminary only.
SCORING THE BUTTER.1
The butter made in these experiments was packed in hermeticallysealed cans holding about 2 pounds each and stored at —12° C.(10° F.). For scoring, one can was sent to each of three experiencedbutter judges who worked independently of each other and with noknowledge of the history of the sample. The scores for the firstseason's work were so inconclusive that they are not included in thetables. In the second season's work a wider range of temperaturewas used, exceeding the limits of safety at both ends of the range.The first scoring was made after about 40 days in storage, and thesecond after about 150 days.
The scores are given in Table 7, and presented graphically infigures 25 and 26. The scores of the three judges are not averagedbecause, while in the main they agree closely, it is believed that onaccount of varying standards an average will not always indicatethe true conditions. The average score at the various temperaturesby each scorer is given in Table 8 and shown graphically in figure 26.
The tables show, as would be expected, a marked difference betweenthe unpasteurized and the pasteurized cream butter. They indicatealso that pasteurizing temperatures between 60° and 66° C. (140°and 150° F.) leave in the cream some factors causing a deteriorationof the butter. Good results were secured from the cream pasteurizedat 71° C. (160° F.). At 81° C, (180° F.) some of the samples weremarked scorched or cooked.
1 The authors acknowledge their indebtedness to Mr. P. H. Keiffer, of Gude Bros. KeifferCo.; Mr. Cromer, of the Fox River Butter Co.; and Mr. John B. Newman, assistant foodand dairy commissioner of Illinois, who kindly scored the butter.
"1 PI
,JT
i i
??
6O -
-T7-
43
-S
O-
€/
-6
2 -
49
-S
S-
€3
-
5/
-52-
€5
-
se-
€7
-€
8-
S3
-S
4--
69
-7O
SS
-57
-7/
-7
2-
59
-
\
\
<h
(0
JOOi
0
0) i;
> ?7:
X
\
\
\ N
r77
Z? t
4
%
%
%
\
i i ij—B>
i I
ss #̂ s7 1 h ^r̂ k /5 if ^i N J
r !
>M
A«>
\_ r1 \\
\
\ \
0U
77
ER
SC
OR
E
<b
to
$ <*
8 Q
3
oi
i i i
i. S s k i
/ *̂"
4> \
< y
j
y
V 1 V
324 27TH REPOBT, BUBEATJ OF ANIMAL INDUSTRY.
At 88° and 93° C. (190° and 200° F.) the quality of the butter wasnoticeably affected by the high heat, although some of the scorchedflavor disappeared on long standing. The temperature at whichcream acquires a scorched flavor is no doubt influenced by the amountof fat in the cream, the promptness of the cooling process, the relationof the amount of cream pasteurized to the capacity of the pasteur-izer, and possibly other factors. The upper limit of pasteurizationtherefore can not be determined arbitrarily, but is dependent onvarying circumstances. Certainly nothing is gained by exceeding82° C. (180° F.), and it is probable that there is no advantage ingoing above 77° C. (170° F.) except the increased certainty of thedestruction of the organisms of tuberculosis. The lower limit of effi-cient pasteurization may be set at 71° C. (160° F.), although underordinary conditions it will be much safer to use at least 74° C.(165° F.).
It should be remembered, however, that these statements applyonly to the conditions under which this work was carried out; thatis, the pasteurization, in a continuous machine, of sweet cream forbuttermaking. If cream is pasteurized in a vat or other holdingdevice lower temperatures may, undoubtedly, be used.
TABLE 7.—Effect of pasteurization of cream at various temperatures on qualityof storage outter.
Sample
4660474850616249586364515265665667.6853546970555771725973
Pasteurizing tem-perature.
•c.
Raw.Raw.
6060606060666666667171717177777780828282888888889393
•F.
Raw.Raw.140140140140140150150150150160160160160
' 170170170176180180180190190190190200200
Scored by N.
40 days.
868693*91*93*868993*9393*9393*9493
93938891*938789*91}9389*868986*
150 days.
85868892*899292*90*93*89*879293*93*93493*899387*90929193*909087*87*90
Scored by C.
40 days.
8885899392889094919191*93*93*9393*909293*9391939392J93*93859084
150 days.
85878790918890a9289a939393*91939292929393493*92858986
Scored by K.
40 days.
85889192898790939189919392939290919190919191888890889088
150 days.
82859293938890939191889393939393919293939293939093909390
PASTEURIZATION OF CREAM FOR BUTTER MAKING.
TABLE 8.—Average butter scores.
325
Numberof
tests.
254431342
Pasteurizing tem-perature.
°C.
Raw.6066717780828893
° F .
Raw.140150160170176180190200
Scored by N.
40 days.
8690A93,93J9U9189{9087f •
150 days.
85}90909391.
. 87i9190}8 8 |
Scored by C.
40 days.
86}90491A93^f91ft9392}9187
150 days.
8689*9293}92}9292}9187}
Scored by K.
40 days.
87}89*9192}903909188}89
150 days.
83}91*90|939293923
m91}
92
i-64
//
•A
\> BY
\
AS
}li
It
r
A1 A\
arc
^ C G
\ iii
1\*scc
/K
*Qff£/t/~7&?4O 1X4*.WIT Af7&?&OA4KS
A\/\
\
er A
92
90 \
02
FIG. 26.—Average scores of butter, showing effect of pasteurization of cream at varioustemperatures.
SUMMARY.
The proper temperature for the pasteurization in a continuousmachine of sweet cream for butter making was determined by observ-ing (1) the temperature at which the greater part of the bacteriawere destroyed, (2) the temperature at which the various enzymsof the milk were inactivated, and (3) the relative keeping qualityof butter made from cream pasteurized at different temperatures.
The uniform destruction of the vegetative bacteria is uncertain attemperatures below 74° C. (165° F.).
Peroxidase was destroyed at 77° C. (170° F.), catalase and lipaseat 70° C. (158° F.). Galactase was much weakened by temperaturesbetween 71° C. (160° F.) and 77° C. (170° F.), but was not destroyedat 93° C. (200° F.), the highest temperature employed.
326. 27TH REPOKT, BUREAU OF ANIMAL INDUSTRY.
Examination of the butter after storage indicated that pasteuriza-tion at 66° C. (150° F.) or lower left in the cream some factor caus-ing a deterioration of the butter. This was not evident in the butterfrom cream pasteurized at the next higher temperature, 71° C.(160° F.)5 or higher. At 82° C. (180° F.) the flavor of the butterwas affected by the heat. This action, however, may be controlledto some extent by the skill of the butter maker.
For the continuous pasteurization of sweet cream for butter mak-ing a temperature not lower than 74° C. (165° F.) nor higher than80° C. (175° F.) is recommended.
IS PUBLICATION may be pro-cured from the Superintendent of.
Documents, Government Printing OfficeWashington, D. C, at 5 cents per copy