irradiation of milk - ACS Publications

It is therefore possible to unite the vari-. 2 2 u) ... sible by proper equipment design. .... The area included above and to the right of these curve...
0 downloads 0 Views 1MB Size
IRRADIATION OF MILK Factors Affecting Antirachitic Response H. H.BECK, H. C. JACKSON, AND K. G. WECKEL University of Wisconsin, Madison, Wis.

The high-pressure air-cooled mercury arc in quartz produces a maximum vitamin D synthesis a t a 6-inch distance on the normal from the center of flowing milk films. All effective radiation is utilized by a film whose width is no more and length is no less than twice the perpendicular distance to the arc. The vitamin D potency of irradiated milk has a parabolic relation to the amount of active radiation applied. This relation is slightly different for different distances from arc to film, but holds for any other method used to vary the amount of applied energy, such as changing the radiation intensity, the film capacity, or the number of successive exposures. Vitamin D photosynthesis in milk, irradiated in thin vertical rectangu-

lar surface films, flowing by gravity, displays characteristics resulting in the relation where variations in the concentration of vitamin D are only 50 t o 60 per cent as large as the responsible combined variations in intensity and film capacity when the variations are expressed as per cent deviations from a mean. Since the intensity-potency relations are parabolic and the film capacity-potency relations are hyperbolic, it is desirable t o employ sufficiently high intensities and film capacities t o avoid excessive variations in potency when deviations of fixed amounts in intensity or film capacity are a characteristic of equipment or control devices. Experiments indicate that the data obtained using milk are equally applicable t o evaporated milk.

I

RRADIATION of milk for the purpose of increasing its antirachitic potency may be briefly described as a process whereby milk is exposed in flowing films to radiation of suitable quality and intensity. The result of the process, expressed as antirachitic potency, is dependent upon the effect and interrelation of a number of factors. Supplee and co-workers (6, 9, 10) indicated that quality and intensity of the radiation as weP as the amount of energy applied are important. Their studies were made using milk in controlled vertical flowing films exposed to the radiation as many as six times (5, 9). Three rates of emission were employed with the arc a t a fixed distance from the milk film. The radiation was further modified with filters, and, as a result, potency reductions were obtained commensurate with the radiation intensity reductions caused by the filters. Briefly, the conditions of these experiments permitted the irradiation of a relatively uniformly flowing film of milk as many as six times, with the same intensity of radiation each time. Altogether the effect of twelve different intensities of radiation emanating from an arc a t a fixed distance were tested. Other factors studied were the transmission and reflecting properties of milk films towards ultraviolet radiation (6,7 , 8). Transmission was reported (7) as being very low-20 to 40 per cent through films 0.02 mm., and less than 5 per cent through films 0.11 mm. thick. Reflection reached about 6 per cent a t an angle of incidence of 60" as the latter was increased from zero. Fluid and evaporated milk were compared as to vitamin D potency induced by irradiation (6). Higher potencies were reported for fluid than for reconstituted

evaporated milk. These reports by Supplee and co-workers have been valuable guides for improving the technic of milk irradiation. The data reported here show that still other factors are operative. These tend to increase the efficiency of the photochemical process and permit more economical design and reduction in size of commercial irradiating equipment. The factors were revealed as the results of studies to determine the effect produced by varying the distance between radiation source and milk film over a wide range of radiation intensities and milk flow rates. Their applicability to evaporated milk was investigated by measurement and comparison with the responses of fluid milk.

Apparatus The apparatus used in the investigation (Figure 1) was arranged to give flexibility, so that the factors, film area and capacity, radiation intensity, and distance of the radiation source from the film might be controlled and varied: The surface over which the milk was passed and irradiated consisted of a vertical stainless steel surface, 36 X 36 inches. The milk flowed onto the surface from a spillway-typedistributor. The rate of flow of milk (weighed discharge) onto the surface was controlled and maintained by joint use of a sanitary pipe valve and float tank control system. The horizontal width of the film was controlled by means of a superimposed paraffin dam built onto the spillway edge. The vertical exposed length of the fdms was adjusted by interposing metal sheets as shades. A Hanovia quartz mercury-vapor arc provided the radiation, the intensity of which was measured, recorded, and maintained by a circuit employing a Weston photronic cell and Englehard meter. The mercury-vapor arc was mounted on a portable stand which per632

JUNE, 1938

INDUSTRIAL AND ENGINEERING CHEMISTRY

mitted adjLqtment of its elevation and distance from the milk film. In all evperiment,s the center of t,he arc was on the norms1 to the center of the milk film. The antirachit,icpoteno.” of the milk was determined according to a modification of U. 8. P. procedure ( 2 ) .

Procedure The distribution of radiant energy impinging on a vertical rectangular surface from a source such as the mercury arc is not uniform. The variations are the result of qualities (such as size and shape) inherent to the arc, and to the die tame of the different parts of its surface from the arc. The magnitude of these variations depends upon the dimensions or area of the rectangular surface in relation to its distance from the arc or, concisely, iipon the solid angle of radiation intercepted by the surface. The solid angle of radiation may be eoncei\.eri as a rectangular pyramid with apex n t the center of the arc and base delineated by the limits of the rectangiilar snrfaee. The angle between the planes defined by the center of the arc with the upper and loner edgcs of the rectangular surface will be termed the ”vertical angle.” Similarly the angle between the planes defined by the center of the arc with the sides of the rectangular surface will be termed the “liorizontal angle.” A milk film w a superimposed iipoii the rectangular surface. The definitions and staternents pertaining to the rectangular surfaces are applicable to the superimposed milk films. To determine the most suitable solid angle, the film travel distance, length or vertical angle, and the film width or horizontal angle were indepemlently varied. In addition, vertical sections of an evaporated milk film were segregated by means of a partitioned collecting trough. The milk collected from sections of the film symmetrically spaced with respect to the vertical centcr line was mixed prior to bioassay. Studies on the effect of changes in the distauce between arc and milk film were made iising a fixed solid angle of radiation

FIQURE1.

APPARATUSAESEMELED FOR OPERATION

633

chosen on the basis of the data obtained from the experiments just described. In order to maintain fixed horizontal and vertical angles of radiation although changing the distance from arc (apex) to film (base), the dimensions of the film were accordindv varied. Figure 2 illustrates the relation of milk film dim&sions to arc- distance when t h e l a t t e r i s v a r i e d . From preliminary investigations, distances of the arc of 4, 6, 8, and 10 inches from the center of the film were selected as representative within the range of interest. Control equipment and characteristics of the merciiry-vapor arc perinitted a rangc of radiat.ioii emission rates prodiicing intensitie.; in wave lengths loss than 3000 A. of 1000, 2000, 4000, arid 6000 miar~rwat~tsper sq. em., measurcd at a clistanoe of 8 inches horiaoiitally oppositc~ the center of the arc. These 0 4 8 1 2 cmission rate figures may be conDISTANCE BETWEEN verted to average intensities on ARC AND CENTER O f FILM IN INCHES the film by multiplying by the ratio of the spherical surface of %inch radius subtended by the pyramid of radiation to the area POSJTION A N D DIMENof the milk film intercepting the SlOVs OF FILM SURsame radiation. Measurements PACES IN KELATION of the intensity along vertical, TO TX-EIR DIST.ASCE horizontal, and both diagonal FBOM THB ARC great circles on such a spheriThis plan gmrmits $he vadetion of radintion mteiisity cal s u r f a c e p e r m i t t e d t h e while applying a wnqtant estimation t h a t t h e distribunmount Of e n e m y wltliin ths fixed adid angle of tion of i n t e n s i t y averaged rrdiation. 100 ner cent of that at the intersection of these great circles which is the point a t which radiation emission rates were measured. Corrections for intensity distribution in calculating the average intensities on the milk film were, therefore, unnecessary. Within the range of radiation intensities used, total milk flow or delivery rates of 150, 300, 600, and 900 pounds per hour are of interest. Vhen the total flow of milk (pounds per hour) is constant, variation in the area of the film aocompanying arbitrary adjustment of the distance between the arc and film produces changes in 6lm capacity which may be expressed in linear dimensions. For example, a total volume of flow of 600 pounds per hour in a film 8 inches wide represents a film capacity of 900 pounds per foot per hour, whereas the same total volume of flow in a film 20 inches wide represents a film capacity of only 360 pounds per foot per hour. Since the thickness, mean velocity, and average travel time of tho film are fixed by the film capacity, the latter may have an important influence on the irradiation process (B). From the standpoint 01 efficiency,however, the total flow of milk is important. Since the film intercepts the same solid angle of radiation at all distances, the energy impinging from a11 such distances per unit of time is constant if the rate of radiation emission remains constant. If the total flow of milk also remains const.ant, the amount of energy applied to a unit quantity of milk will remain unchanged. Any difference in vitamin D potency produced in the milk as the distance of the arc is varied may then be the result of corresponding changes in intensity or film capacity or both. Irradiation of milk in films having the four rates of flow (150, 300, 600,900 pounds per hour) with each of the four rates of energy emission (1000, 2000, 4000, 6000 mw. per sq. em.) when a t

VOL. 30, NO. 6,

INDUSTRIAL AND ENGINEERING CHEMISTRY

634

especially where the angle is greater than 75". I n the sections between angles of 90" and 105" the potency becomes about half of the probable maximum. On these latter sections the intensity is less than one-fourth of the maximum. Consequently, a horizontal angle of 90" includes the film surface effectively irradiated by the arc. The curve conforms very closely to, and may be estimated by, the cosine of onehalf the -horizontal angle as indicated in Figure 4 by the series of crosses. The data and TABLEI. EFFECTOF VARYINGMILK FILM DIMENSIONS TO INTERCEPT DIFFERENT SOLID curves were interpreted as inANGLESOF RADIATION ON THE ANTIRACHITIC POTENCY OF MILK dicating that both vertical and Radiation Less than horizontal angles of 90" will Pyramid 3000 R . of Radiation Av. in- Energy define a pyramid of radiation Sample -Milk FilmHorizontal Vertical tensity applied Vitamin D effective and applicable under Milk No. Width Length Capacity angle angle on film to milk Potency Lb./ft./ MIL/ u.s. P . the conditions of irradiation In. In. hr. Degrees Degrees sq. cm. Watts/qt. units/&. employed in this investigaFilm-Travel Distance Varied tion.

distances of 4, 6, 8, and 10 inches from the arc permits a comprehensive test of all three variables. The term "fluid milk" refers to fresh whole milk with a fat content of 3.5 to 3.7 per cent. Evaporated milk as irradiated in these experiments was the evaporated product obtained from a processing line following homogenization and cooling but before standardization.

Fluid

E,vaporated

Fluid (aggregate film) Evaporated (segregate film)

R-6 R-7 R-8 R-9 R-42 R-43 R-44 R-45

8 8 8 8 12 12 12 12

4.5 9.0 18.0 36.0 9.25 13.5 19.0 27.75

R-22 R-23 R-24 R-25 R-38 R-39 R-40 R-41

9.25 13.5 19.0 27.75 9.25 12.25 16.00 20.75

20 20 20 20 20.5 20.5 20.5 20.2

307 303 303 303 600 600 600

600

53

53 53 74 74 74 74

Film Width Varied 600 60 600 80 600 100 600 120 600 60 600 60-75 600 75-90 600 90-105

29 59 97 132 60 80 100 120

103 103 103 103 118 118 118 118

Size of Solid Angle of Radiation Both fluid and evaporated milk were used in tests to determine the size of the solid angle of radiation most suitable to milk irradiation with the mercury arc. The conditions and results are shown in Table I and Figures 3 and 4. The curves indicate a fairly uniform increase in vitamin D potency with increase in vertical angle of radiation up to an angle of 100". Conversely, uniform decreases in potency with increases in film width or horizontal angle of radiation are indicated by the curve marked '(aggregate" (Figure 4) which shows these relations as obtained with composite samples. Segregation of the milk flowing in definite vertical sections of a film resulted in the illuminating potency relations indicated by the curve marked "segreFLUID M I L K gate." The potency of the milk from the central portion of the film intercept140 ing radiation within a horizontal 5 120 angle of 60" is relan tively greater. I n the sections of the film intercepting radiation outside 6o0 20 40 60 80 100 120 I of the 60"horizonVERTICAL ANGLE OF RADIATION IN DEGRE tal angle the poFIGURE 3. EFFECTON VITAMIND tency of the milk CONTENTOF IRRADIATED MILK AS decreases rapidly VERTICAL ANGLEOF APPLIEDRADIAas the radiation TION Is VARIED BY CHANGESIN angle i n c r e a s e s , TRAVEL DISTAXCE OF FILM

3400 3160 2440 1480 2800 2400 2160 1680

25.3 54.0 83.4 101.0 13.4 17.45 21.1 23.97

135-150 175 215 215 80 108 120-135 160-175

Conditions of Radiation

The data obtained in the study of various conditions of milk irradiation, when limiting 2220 43.2 160 the applied radiation to that 2000 38.1 120 110-120 1720 passing within a 90" rectan33.1 1360 67.5-80 26.2 gular pyramid, are detailed 2520 26.4 120-135 in Table 11. These data are 1760 110-120 18.6 1445 15.1 80 analyzed with the aid of Fig855 9.0 67.5 ures 5-8 to determine the importance of each of the factors under investigation. Variations in vitamin D potencies with changes in the arcto-film distance are shown in Figure 5. An outstanding feature of all these curves, with one exception probably due to experimental error, is that they have a definite maximum at the abscissa, representing the distance of 6 inches between the arc and film. Since each curve, respectively, represents

K

2 y)

t

-

120 -

= 100 v

l

3

z

EVAPORATED MILK

-

-.

80-

0

J

60

'2

40

{ E ]

FLUID MILK

60

40

20

0

20

40

60

80

100

120

2

HORIZONTAL ANGLE OF RADIATION IN DEGREES

FIGURE4. RELATIVEEFFECTIVENESS OF RADIATION IMPINGING ON ADJACENT SYMMETRICAL VERTICAL SECTIONS OF A MILKFILM, EXPRESSED AS VITAMIN D CONTENT

the results produced by applying a uniform amount of energy per quart of milk, the maxima exhibited by the curves indicate the existence of a condition where the energy is most efficiently utilized. An explanation for the occurrence of maximum potencies when the radiation source is 6 inches from the film in contrast to other distances is not evident. The maximum occurs for each combination of intensity and film capacity studied. Since an intensity range of 600 per cent and a film capacity range of 600 per cent were tested at the 6-inch as well as a t other distances from arc to milk, the greater response a t the

JUNE, 1938

INDUSTRIAL AND ENGINEERING CHEMISTRY

6-inch distance appears to be independent of both intensity and film capacity. Figure 5 shows four curves in one group which have approximately equivalent potencies and which represent the results of application of four different intensities and film capacities. I n each case 44.6 watts of energy were applied per quart. Calculations show that the size of the quartz mercury-arc radiation source is responsible for a reduction of about 3.4 per cent in the amount of radiation incident on the film when a t 4 inches, as compared with 10 inches, from the film. This reduction is within the degree of accuracy by which the radiation was measured and hence is too small to be a determinative factor in accounting for the efficacy of the radiation when the film is a t a distance of 6 inches. If the distance from arc to film and the film dimensions are simultaneously decreased, but maintained a t a ratio of 1 to 2 as in these experiments, these dimensions will approach zero as a limit. As this limit is approached,-the thickness and velocity of the film (total flow remaining constant and ideal conditions being assumed) will become very great. The low penetrating power of the active rays (7, 11) confines their activating action to the vicinity of the surface of the film. As a result, the greater portion of such very thick films would pass uninfluenced by the radiation, regardless of the intensity of the latter. On the other hand, it is conceivable that, if under ideal conditions the distance from arc to film and proportionately the film dimensions were made very great, the resultant concentration of radiation and activatable material would become so low that the frequency of collision of photons and activatable materials would be very low. It is conceivadle t h e n , that between two such limits of effectiveness, there is a condition a t k a 24 0 0 [ which the effectiveness reaches a maximum. L L The maxima o b s e r v e d 0 W in the experiments ocIn t 300 curred when t h e disz 3 tance of the arc to film h u l w a s 6 i n c h e s . If this 3 maximum corresponded z 200 to a definite film cac pacity and i n t e n s i t y , 5 c the explanation for its 0 u o c c u r r e n c e m i g h t be n 100 complete. Since it ocz E curs a t a definite dis< c tance from arc to film of 6 inches, independent of the intensity or film caO2 ' d ' $. ' i ' Ib DISTANCE BETWEEN pacity, it is assumed that ARC AND CENTER OF FILM IN INCHES other little-known factors are involved. MILK ARC APPLIED CURVE FLOW EMISSION2 ENERGY I n Figure 6 the relaNO. LBS./HR. W./Q. MW./CM. I 150 6000 256 tion between vitamin D I50 2 4000 178.3 3 300 6000 134.5 potency and the amount 4 300 4000 89. I 5 I50 2000 89.1 of radiant energy ap6 600 6000 7 0.6 7 I50 IO00 44.6 plied per quart of milk 8 300 2000 44.6 9 600 4000 44.6 is shown. Logarithms IO 900 6000 44.6 I/ 900 4000 were employed to aid in 29.7 12 300 IO00 22.3 1 3 600 interpretation and analy2000 22.3 14 900 2000 14.9 sis of the data, and to 15 600 IO00 11.1 16 900 I 000 7.4 prevent e x p e r i m e n t a1 FIGURE5. EFFECTOF DISerrors from appearing TANCE BETWEEN ARC AND MILK disproportionate. FILMON VITAMIN D CONTENT Differences in potency IN IRRADIATED MILK OVER A levels for the several disW I D E RANGE OF MILK FLOW AND ARC-EMISSION RATES tances from arc to film

\

a-

t

635

TABLE 11. EFFECT OF DISTANCE OF SOURCE, RADIATION INTENSITY, AND FILMCAPACITY ON EFFECTIVENESS OF RADIATION IN INCREASING THE ANTIRACHITICPOTENCY OF MILK Sample NO.

Distance of Arc t o Film Film Dimensions

Film Capacity

Radiation Less than 3000 A. Av. in- Applied tensity energy

Mw./

Watta/ qt.

s.

In.

I%.

R-144 R-140 R-136 R-132

10

20 x 20 16 X 16 12 x 12 8 x 8

R-106 R-102 R-98 R-94

10 8

20 x 20 16 X 16 12 x 12 8 x 8

360 450 600 900

335 525 930 2,090

11.1 11.1 11.1 11.1

67.5-80 67.5-80 80-95 55

R-143 R-139 R-135 R-131

10 8

20 x 20 16 X 16 12 x 12 8 x 8

540 675 900 1350

670 1,060 1,860 4,190

14.9 14.9 14.9 14.9

67.5 87.5 95.0 67.5

R-90 R-86 R-82 R-78

10

8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

180 225 300 450

335 525 930 2,090

22.3 22.3 22.3 22.3

80.0 110-120 110-120 80-1 10

R-105 R-101 R-97 R-93

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

360 450 600 900

670 1,050 1,860 4,190

22.3 22.3 22.3 22.3

80-95 110-120 130 80-95

R-142 R-138 R-134 R-130

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

540 675 900 1350

1,340 2,100 3,730 8,380

29.7 29.7 29.7 29.7

110 115 135 110

R-141 R-137 R-133 R-129

10

20 x 20 16 X 16 12 x 12 8 x 8

540 675 900 1350

1,875 2,940 5,590 1,700

41.6 41.6 44.6 41.6

135 150 175-190 135-150

R-74 R-70 R-66 R-62

10

20 x 20 16 X 16 12 x 12 8 x 8

90 112.5 150 225

335 525 930 2,090

44.6 44.6 44.6 44.6

120 135-150 175 110-120

R-89 R-85 R-81 R-77

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

180 225 300 450

670 1,050 1,860 4,190

44.6 44.6 44.6 44.6

135 136- 150 160 150

R-104 R-100 R-96 R-92

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

360 450 600 900

1,340 2,100 3,730 8,380

44.6 44.6 44.6 44.6

135 175 180 135

R-103 R-99 R-95 R-9 1

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

360 450 600 900

2,180 3,300 5,870 14,670

72.4 70.3 70.6 78.5

160- 175 215 230 200

R-73 R-69 R-65 R-61

10 8

6 4

20 x 20 16 X 16 12 x 12 8 x 8

90 112.5 150 225

670 1,050 1,860 4,190

89.1 89.1 89.1 89.1

200 230 200-215 175-200

R-88 R-84 R-80 R-76

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

180 225 300 450

1,340 2,100 3,730 8,380

89.1 89.1 89.1 89.1

200 230 230-245 190-200

R-87 R-83 R-79 R-75

10 8 6 4

20 x 20 16 X 16 12 x 12

180 225 300 450

1,840 3,100 5,320 13,620

124.5 131.3 128 145

245 300 300-310 230-2 5 5

R-72 R-68 R-64 R-60

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

90 112.5 150 225

1,340 2,100 3,730 8,380

178.3 178.3 178.3 178.3

270 325 340-366 325

R-71 R-67 R-63 R-59

10 8 6 4

20 x 20 16 X 16 12 x 12 8 x 8

90 112.5 150 225

1,910 3,150 5,320 11,950

254 267.4 256 254

365 400 430-460 400

8 6 4

6 4 6 4

8 6 4 8

6 4

8 x 8

L b . / j t . / h r . sq. cm. 540 335 675 525 900 930 1350 2,090

Vitamin D Potency u. P .

7.4 7.4 7.4 7.4

unats/qt.

47.5 55.0 67.5 47.5

shown by the curves in Figure 5 are also evident in Figure 6. To determine definitely whether such distinctions are actually in accord with the data, and to avoid personal errors, the logarithmic curves were fitted to the points by the method of least squares (3). The results are shown by the straight-line curves in the lower portion of Figure 6 and conform to the following equations: Pin. distance: 6-in. distance: &in. distance: 10-in. distance:

log P = 1.129 log P = 1.385 log P = 1.302 log P = 1.154

+ log E + 0.605 0.514 log E + 0.539log E + 0.584 log E

(1)

[E{ (4)

INDUSTRIAL AND ENGINEERING CHEMISTRY

636

VOL. 30, NO. 6

dred times, this may be partly, &tleast, explained by the method used for estimating the vitamin D present. The results are in agreement in that identical variations in applied energy produced '+< & 400 approximately equal percentage changes in 22 W U antirachitic potency. Successive exposures and $a 300 2.6 film capacity were separately shown to be com$? & parable with intensity in their effect on anti2.4 rachitic potency. Since each is directly comparable with intensity, they may be mutually compared. It is therefore possible to unite the vari2 2 u) ous factors in a generalization covering their ' relation to potency. The irradiation of flowing 0 2.0 films of milk with a given radiation source a t a 0 x ARC 6" FROM FILM % given distance from the film produces an antiA 8 1 1 1.8 rachitic potency dependent upon the number of 10" successive exposures, the film capacity, and the radiation intensity only in so far as these affect 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 1.6 the amount of applied energy. LOG WATTS PER QUART (2000-3000A3 The generalization just made indicates that control of the irradiation process is dependent FIGURE6. RELATION BETWEEN VITAMIN D CONTENT AND AMOUNT OF upon adjustment and regulation of the amount ENERGY APPLIEDPER QUARTOF IRRADIATED MILK of energy applied to the milk. However, the Straight lines represent logarithmic relations, and curved lines, numerical relations. amount of available energy is directly proportional to the average intensity of radiation, b u t the amount of radiation applied per quart of milk is inversely These equations may be converted, respectively, into the proport'onal to the number of quarts flowing per unit of time numerical form : or to the film capacity. Control of these two variables is posP 13.5 Eo.605 sible by proper equipment design. Nevertheless, it is desirP = 24.3 Eo.514 able to know the degree of variation permissible to each of P = 20.0 Eo.64 (7) the factors in an acceptable control procedure. P = 14.3 Eo.684 (8) where P = vitamin D potency, u. 8. p. units/quart, E = amount According to empirical formulas 5 to 8 derived from the of applied radiant energy of wave lengths shorter than 3000 A., data, the PotenCY-energY relations are in m~ordancewith the watts/quart. expression, 5oo

APPLIED ENERGY C2000-3000h IN WATTS PER QUART 0 40 80 120 160 200 240 280 320

2 5

$6

4 0

I,

M

II

I(

$8

[:{

The relations expressed by Equations 5 to 8 are illustrated by the parabolic curves in the upper portion of Figure 6. The differentiation permitted by the data is emphasized by the complete and extended separation of the curves, taken in the sequence of the arc-to-film distances. The advantage which may be had by placing the arc a distance of 6 inches from the film is evident. When compared on the basis of equivalent watts of applied energy per quart, the relative advantage of the 6-inch over other distances, expressed as U. S. P. units per quart, is as follows: 4 inches, 30-40; 8 inches, 20; loinches, 55. Analyses of the data for individual curves in Figure 6 reveal that a wide variation of both radiation intensity and film capacity are represented by each curve. The deduction may therefore be made that for any given distance between arc and film, identical variations in the amount of applied energy, whether produced by changes in radiation intensity, film capacity, or both, in any proportion, have identical effects measured in terms of vitamin D potency of the milk. Somewhat analogous data and curves were presented by Supplee, Beck, and Dorcas (6). They compared the results produced by different radiation sources varied over a wide range of intensities and applied in one to six successive exposures. Their data permit a deduction, not stated by them, that for any given radiation source, identical variations in the amount of applied energy, whether produced by changes in radiation intensity or successive exposures of the milk, have identical effects measured in terms of vitamin D potency. A comparison of the data from both investigations reveals the applied energy-potency relations in each to be expressible as power functions with exponents in the range from 0.5 to 0.6, dependent upon the quality of the radiation or the distance from arc to film. Although the proportionality coefficients in some of the expressions differ as much as two-hun-

where K l/n

= =

p = KElln a constant decimal exponent in equivalent fraction

(9)

If dE represents an error in evaluating the energy, dE/E will represent a percentage error. Likewise dP/P will represent a percentage error in potency. The relation between dE and dP can be obtained by differentiating Equation 9; hence, dP =

dE I-n -

nE

It

Dividing through by the equality P = KE"",

n E n

' E n

Since the exponents l / n are all included within the range of values from 0.5 to 0.6, a percentage deviation in energy will result in a percentage deviation in potency only one-half to three-fifths as large. That is, if the applied energy can be adjusted within 10 per cent, the resultant potency can be established within 5 to 6 per cent of the desired amount. But the applied energy is dependent upon two independently controllable factors, arc-emission intensity and milk flow rate. These have the following relations:

1 = time F film capacity w = width of film

INDUSTRIAL AND ENGINEERING CHEMISTRY

JUNE, 1938

637

of intensities and film capacities studied. T h e curves unite to indicate four curved surfaces, each representing the conditions for one of the four arc-to-film distances studied. The curves have steep slopes opposite small values of intensity and film capacity, but become relatively flat as the intensity and film capacity values become greater. Consequently, fixed variations will have the least effect on resultant potency when the operating conditions are in the range where the curves are relatively flat-that is, where the slopes are less than one. I n this range a fixed variation in intensity or film capacity will produce equal or smaller variations in potency. A quantitative study of the slopes of the curves requires conversion to units whose size and importance are equivalent. Figure 6 shows that 55 U. S. P. units per quart is the least potency produced by the application of 10 watts per quart and that 370 U. S. P. units is the maximum potency produced by 200 watts per FIGURE7. RELATIONOF VITAMIND CONTENTOF IRRADIATED MILK TO quart. When converted to Steenbock units, THREE INDEPENDENT VARIABLES OF THE IRRADIATION PROCESS the potency-energy ratios are 2 and 0.685 SteenLinear capacity of milk film, average intensit of radiation on surface of film, and distance between radiation source a d milk film are the variables. bock unit per watt, respectively. There must, then, be an intermediate condition such that the ratio will be 1.0 Steenbock unit per watt. The radiant energy was applied in average intensities from Since A and w can be adjusted with considerable accuracy 335 to 12,570 mw. per sq. cm., which can be expressed as and since an error in w will be reflected to a similar extent in 0.312 to 11.7 watts per square foot. The milk flowed at 0.667 A and cancel in A/w,these can be treated as constants. Also, to 4.0 ounces per second (by weight) over areas, resulting in milk flow is regulated by the quantity discharge per unit 1.0 ounce being spread on 2.54 to 0.93 square foot of surface, of time; that is, Q = Fw. Expression 12 may now be respectively. These figures show that a condition can be written: selected such that 1 ounce per second is exposed on 1 square foot of surface receiving 1 watt per second of active radiation; the result is that each ounce will contain 1.0 Steenbock unit of vitamin D, and that this condition will be well within the This indicates that deviations in I and Q may be counterbalrange included in this study. These units may therefore be anced when in the same direction or additive when in opposite accepted as equivalent for the purpose of studying the curves directions. Allowing for deviations of 10 per cent in ear,h, in Figure 7. the maximum deviations in E are from loo+ lo x 100 - 100 = +22.2per cent 100 - 10

to loo - lo X 100 - 100 = -18.2 per cent 100 10

+

Since deviations in P are only 50 to 60 per cent of those in E , the maximum deviations in P resulting from deviations of 10 per cent in each of I and Q will be 11.1 to 13.3 per cent. The standard procedure in use for biological assays of vitamin D content of' food and drug products for control work has now reached a precision of 10 to 15 per cent deviation (4). The analysis indicates that the accuracy of design and uniformity of operation of milk irradiation equipment can be within the limits of accuracy of biological assay procedure. The latter niay be reserved safely for control purposes, as are determinations of the bacterial content of milk. The sensitivity of control devices is a factor which affects the extent of variations experienced during operation. These frequently require a fixed, as contrasted to percentage, change to institute their operation. It is therefore desirable to evaluate the effect of fixed variations in radiation intensity and film capacity on the vitamin D potency. Equrition 12 indicates that the intensity-potency relation is parabolic and the film capacity-potency relation is hyperbolic. These are shown graphically in Figure 7, which includes the complete range

I

,VITAMIN D LEVELS IN U.S.P. UNITS PER QUART INDICATED BY FIGURES

FIGURE8. CURVESRESULTING FROM ANALYSES MADETO DETERMINE CONDITIONS BY WHICHTHE IRRADIATION PROCESS CANBE CONSIDERED DEPENDABLE AS A MEANSFOR INCREASING THE VITAMIN D CONTENT OF MILK The area included above and to the right of these curve8 represents those conditions considered most reliable.

VOL. 30, NO. 6

INDUSTRIAL AND ENGINEERING CHEMISTRY

638 -

111. FILMCAPilCITIES AXD INTENSITIES AT WHICH SLOPE OF RESPECTIVE CURVES Is UNITYAND ONE-HALFUNITY~

TABLE

--.

Film Capacity Varied Point Point of Range in of, halfAv. unit unit intenfilm sity capacitv slope slope

I

Film capacity Lb./ft./ hr. 150 300 600 900

Intensity Varied Point Point of Range of of av. unit half-unit intensity slope slope -----Mw./sq. cm.930-5590 1100 5060 930-5590 5 3 7 . 5 2246 930-5590 258 1070 930-5590 701 168

Mw./

sq. cm.

930 1860 3730 5590

--Lb./ft./hr~-187 150-900 236 150-900 302 150-900 346 150-900

298 375 478 547

tive energy curves in Figure 6. Since the surfaces representing results for the 6-, 8-, and 10-inch distances between arc and film intersect and are nearly parallel, they virtually combine into a single surface. The data showing where the slopes of the curves for the 6-inch arc-to-film distance are unity and the analysis based on them are, therefore, approximately applicable to irradiating conditions for 8- and 10-inch arc-tofilm distances. Present commercial equipment meets the limitations here developed; percentage deviations plus fluctuations of fixed amounts remain within the accuracy of biological assay error.

a This analysis is limited t o curves representing results obtained when distance from arc to milk film is 6 inches.

Applications to Evaporated Milk Most of the data so far presented were prepared from experiments using fluid milk. Approximately 60 per cent of the evaporated milk produced is irradiated. The application of the data to evaporated milk is therefore of interest. The data in Table I and Figures 1 and 2 indicate some similarity in the response characteristics of fluid and evaporated milk. More directly comparable data are given in Table IV and Figures 9 and 10, including fluid whole milk and reconstituted 1:l evaporated milk. Figure 9 shows the results for variations of film capacity for both fluid and evaporated milk, all other conditions remaining constant. The fact that the two curves cross may be accepted as indicating practical equality in response, and their close parallelism as indicating a high degree of similarity in the vitamin D synthesis in the two milks. Figure 10 shows another comparison-the results obtained by irradiating evaporated milk when varying the intensity by changing the distance of the arc. The total flow of FLUID milk, t h e t r a v e l 180 distance, and the horizontal angle of r a d i a t i o n were - 0 I2 held constant, the 2000 3000 4000 5000 6000 latterby changing AVERAGE INTENSITY I N the film width with MW/C~C2000-3000 FIGURE10. VITAMIN D CONTENT the distance Of the OF IRRADIATED FLUIDAND EVAPOarc. I n this inRATED MILK'IN RELATION TO AVERs t a n c e t h e reAGE INTENSITY ON SURFACE O F FILM sp o se of t h e RESULTING FROM CHANGESIN DISTANCE BETWEEN RADIATION SOURCE evaporated m i l k AND MILKFILX was s o m e w h a t

Using the equivalent units, the points a t which tangents have unit and one-half unit slopes were calculated for all curves representing the results for the 6-inch distance between arc and film. The n u m e r i c a l values 440 1 are given in Table I11 and are graphi$ 400< tally represented in 3 0 360 F i g n r e 8. T h i s a chart contains a 320reproduction of the cn t 3 280-intensity potency Q film capacity curves VL 240 EVAPORATED MILK for the 6-inch arc.FLUID MILK z 200 to-film d i s t a n c e I - shown in Figure 7. 160Indicated on the curves in Figure 8 8 120a are the points where 5 80their tangents have 4 . unit and one-half !5 40u n i t slopes. The ~ 0 " " " " 500 " " ' ' 1000 1500 points are joined by 0

extrapolated permit inclusion of the points a t which their tangents have unit slope. The extrapolated portions are shown by dotted lines. Lines indicating 135, 200, 300, and 400 U. S. P. units per quart levels are included to facilitate interpretation. Figure 7 shows that the surfaces representing the irradiation results for 6-, 8-, and 10inch arc-to-film distances intersect. Equation 12 shows that the relation of vitamin D potency to energy and to the intensity-film capacity ratios is the same for any given set of conditions, such as the distance between arc and film. Consequently, the several surfaces in Figure 7 are as near being parallels as the respec-

to

TABLEIV. INFLUENCE OF EXPOSURE CONDITIONS ON COMPARATIVE EFFECTIVENESS OF ANTIRACHITIC POTEXCY OF FLUID AND EVAPORATED MILK

RADIATION I N INCREASING THE

Mllk

Evaporated

Fluid

Evaporated

Sample No.

-Pyramid of RadiationArc HoriVertiFilm diszontal cal Capacity tance angle angle Lb./ft./hr. In. Degrees Degrees

R-34 R-35 R-36 R-37

300 600 1000

R-26 R-27 R-28 R-29 R-46 R-47 R-32 R-33

600 600 600 600 600 600 600 600

100

8 8 8 8

74 74 74 74

Radiation of Wave Lengths Less than 3000 A. Av.inEnergy tensity applied on film to milk Mw./sq. cm. Watts/&.

100 100 100 100

Arc-to-Film Distance Varied 4 167 90 90 5 152 6 140 90 90 8 118 90 4 152 167 90 5 6 140 90 8 118 90

Vitamin D Potency U.S.P . units/qt.

2140 2140 2140 2140

126.0 41.9 21.0 12.6

325 175-190 120-135 80

5440 4080 3200 2120 5440 4080 3200 2120

106.0 79.5 63.0 41.3 56.6 42.75 33.7 22.2

135 135 160 135 135 160-175 190-200 135-150

JUNE, 1938

639

INDUSTRIAL AND ENGINEERING CHEMISTRY

better than that of fluid milk. Supplee et al. (6) reported results in which the fluid milk potencies were consistently slightly higher than those of evaporated milk. I n commercial operation it is generally the practice to maintain, with a given type of irradiating unit, a slightly greater flow than is used for fluid whole milk. Therefore, it appears that the response characteristics of fluid and evaporated milk are very similar and that the data here presented will apply equally well to evaporated and to fluid milk.

Acknowledgment The authors wish to acknowledge and express their appreciation for the aid given them b y W. S. Slemmons of the Carnation Company, to the Carnation Company for the use of laboratory facilities and equipment, and to the Hanovia Chemical and Manufacturing Company and the Wisconsin Alumni Research Foundation for the funds which made the study possible.

Literature Cited (1) Am. Pub. Health lissoc., Rept. Comm. on Standard Methods for Bioassay of Vitamin D, Food and Nutrition Sect., Oct., 1937. (2) Beck, H. H., and Weckel, K. G., IND.EXQ.CHEM.,28, 1251 (1936). (3) Daniels, F., “Mathematical Preparation for Physical Chemistry,” p. 237, New York, McGraw-Hill Book Co., 1928. (4) Scott, H. T., private communication, Oct. 9 , 1937.

(5) Supplee, G. C . , Beck, H. H., and Dorcas, M. J., J . Bid. Chem., 98, 769 (1932). (6) Supplee, G. C., Bender, R. C., Flanigan, G. E., Dorcas, M. J., and Greider, C. E., J . Dairu Sci., 19, 67 (1936). (7) Supplee, G. C., and Dorcas, M. J., Ibid., 17, 4 3 3 (1934). (8) Ibid., 17, 607 (1934). (9) Supplee, G. C., Dorcas, M. J., and Hess, A. F., .J. B i d . Chem., 94. 7 4 9 (1932). (10) Supplee, G‘. C., Hanford, Z . M . , Dorcas, M. J., and Beck, H. H., Ibid., 95, 687 (1932). (11) Trebler, A. H., Robinson, F. W., and Bird, L. A., unpublished studies. RECEIVED January 17, 1938. Submitted for publication with the approval of the Director, Wisconsin Agricultural Experiment Station.

Autoxidation of Terpenes in Petroleum Solvent J. N. BORGLIN Hercules Powder Company, Wilmington, Del.

T

1JRPENTINE tends to discolor somewhat on long storage, increases in specific gravity, and often forms a precipitate due to oxidation. The use of antioxidants such as hydroquinone, gallic acid, pyrogallic acid, activated carbon, and stannous chloride with turpentine was reported by Dupont and Crouzet (1). Later, Smith and Holman (2) reported that calcium oxide in the form of granulated quick lime preserves turpentine for 2 to 3 years and that sodium sulfite is effective for about 5 years. They also tested a number of other reagents but gave no detailed information as to their effectiveness. While turpentine will oxidize on long storage, a product which contains petroleum and terpene hydrocarbons tends to oxidize and precipitate, owing to the low solvent power of petroleum hydrocarbons for the oxidation products of the terpenes. Such blends of petroleum and terpene hydrocarbons are commercially available and are used as paint and varnish thinners. The present study was made with this type of product, and the results of the above investigators were verified with relation to hydroquinone and calcium oxide. However, it was found that triethanolamine and also aliphatic alcohols are equal to or more effective than the antioxidants previously reported. The action of hydroquinone is not given in the data to be presented ; however, comparable experiments with a similar petroleum-terpene hydrocarbon product indicated that 0.1 per cent of hydroquinone was somewhat less effective than 1.0 per cent of triethanolamine. The petroleum-terpene hydrocarbon blends used in this investigation are normally produced apart from pure turpenbine, if the steam and solvent process is employed for extract-

Solutions of terpenes in petroleum hydrocarbons oxidize on long storage. This oxidation can be largely prevented by the use of relatively small quantities of triethanolamine or a l i p h a t i c a l c o h o l s s u c h a s methanol, ethanol, etc. Less effective antioxidants are sodium hydrosulfite, aqueous ammonia, and also anethole.

ing wood rosin from southern pine stumps. Typical analysis of a product of this type follows: A. S. T. ,M.Boiling Range

90 95

160.5 163.2

This product contains in excess of 50 per cent terpene hydrocarbons of which pinene is the major constituent.

Preservative Tests To a measured quantity of the petroleum-terpene hydrocarbon product in well-stoppered flasks, which were threequarters full, was added 0.5 t o 1.0 per cent by weight of the various materials under test. At intervals of 2 to 3 days the solutions were well contacted with air under comparable conditions without evaporation. At intervals of 1 month throughout a period of 10 months, each solution was tested for specific gravity. The results of these tests are given in Table I and Figure 1. From these results it will be noted that the original product (no preservative) passed through an induction period of about 3 months, after which oxidation was rapid and uniform. The presence of tricresyl phosphate (Lindol) or a-terpineol somewhat accelerated oxidation. However, with metallic