T he U ltr a v io le t A bsorption o f V ita m in s K i , K 2, and Som e R ela ted C om pounds A C ritical S tu d y o f th e A p p lica b ility o f th e S p ectrop h otom etric D eterm in a tio n o f V itam in A to F ish L iver Oils By JOH N M ELVIN V A N D E N B E L T A THESIS Presented to the Graduate School of Michigan State College of Agriculture and Applied Science in Partial Fulfillment of Requirements for the Degree of Doctor of Philosophy D ep artm en t o f C hem istry E a st L ansing, M ichigan ProQuest Number: 10008491 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest. ProQuest 10008491 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 81 0 6- 1346 346 U ltrav io let A bso rption of V itam in s a No. 450-VA inductance condensed spark betw een tu n g ste n steel electrodes, and th a t in the la tte r p a rt b y a H ilger hydrogen q u a rtz lamp No. H-698. Hexane (b.p. 63-64.5°) was used as a solvent for th e v itam in compounds and others, except in th e case of 2-m ethyl- and 2 -e th y l-l, 4-naphthoquinone-3-acetic acids which were insoluble in hexane. E th y l alcohol was used as th e solvent for these acids. T hroughout this investigation two typ es of 10 m m . q u a rtz absorption cells were used, one supplied b y B ausch an d Lom b, th e other by Zeiss. T he Bausch and Lomb cell was equipped with detachable optical q u artz ends supported w ith m onel m etal fittings. E astm an No. 33 plates, 4 X 10 inches, were used and processed with pyro (No. H -l) developer for 6 m inutes a t 18°. U ltra vio let A b s o r p tio n C u rv e s V ita m in s K x a n d K 2—Fig. 1 shows th e absorption curves of vitam ins Ki and K 2 in hexane. T he m axima, m inim a, and oth er characteristics of the curves are quite similar. Each has a broad band with strong absorption in the region X 240 to 275 m/x w ith a rather fine stru ctu re revealing sharp m axim a a t X 243, 249, 260, and 270 m/x. The right and left portions of th e tw o bro ad bands seem to be rath er separate p arts and are form ed of tw o sm aller b u t definite bands near together, for both vitam ins. T he E\°cia. of one band (249 m/x) in vitam in Ki is 540; the corresponding band in vitam in K 2 has an of 305. T he o th er m axim a of each compound have almost as high an extinction coefficient as the X 249 m/t maximum in fresh solution. Each curve also shows a broad and less intense absorption b an d in the region of X 310 to 340 m/x w ith a m axim um very close to X 325 m/i. The E value of these m axim a for bo th com pounds is about 75 and indicates a fundam ental sim ilarity in chem ical structure. V ita m in i n the S o lv e n ts A lc o h o l a n d H e x a n e —T he ab so rp tio n curve of vitam in K 2in ethyl alcohol (Fig. 2) has a bro ad b an d from X 235 to 280 m/x split definitely into two ra th e r flat, sm aller ban d s with maxima at X 247 and 267 m/x. The coefficient is n o t as large as in hexane. A weak, broad band in the higher u ltr a  violet region has a wave-length value corresponding to th a t of the curve in hexane b u t its E is n o t as great. Ew ing, V and enbelt, an d K am m 347 T he use of hexane as a solvent gives a curve very similar in outline, b u t characterized by the fine stru cture and maxima a t ^ 243, 249, 260, and 270 mju. Because of th is finer structu re F i g . 1 . T h e a b s o r p t io n c u r v e s o f v it a m in s I I ) in h e x a n e . K i (C u r v e I) a n d K 2 (C u r v e / c.r- /< m. F i g . 2 . T h e a b s o r p t io n c u r v e s o f v it a m in h e x a n e (C u r v e I I ) . K 2 in a lc o h o l (C u r v e I) a n d in brought out in hexane, this solvent was used in subsequent ab sorption studies of th e vitam ins and their derivatives wherever possible. V itam in K 2 is unstable when exposed to light (MacCorquodale, U ltrav io let A bsorption of V itam ins 348 Binkley, McKee, Thayer, and Doisy, 1939). An alcoholic solu tion of crystalline vitam in K 2 was received from D r. D oisy in January, 1939, and examined for its u ltrav io let abso rptio n. Significant absorption was exhibited in the region X 240 to 273 m.£i. On the suspicion th a t the sample had changed, since it h ad been dissolved 3 weeks previously, another sample was exam ined w ithin 3 days after its preparation. As expected, the second sam ple showed a finer structure, not only in ethyl alcohol b u t especially in hexane. The hexane curve gave the characteristic fine d istin ct maxima. ffljJ F ig . £40 £60 £00 ~JOO 0£0 J f lO 030 400 3. T h e a b s o r p t io n s p e c t r a o f v it a m in K 2 in e t h y l a lc o h o l Fig. 3 shows ultraviolet absorption spectra of vitam in K 2 in ethyl alcohol (as described above) with m axim a a t X 247 and 267 nip.. Curve II shows the curve of the same solution after 3 days in darkness. The curves are very nearly the same, showing little difference in either tall or short broad bands. The physiological potency of vitam in K 2 is retained for a sim ilar length of time if dissolved in alcohol and kept in darkness. Curve II, Fig. 4, is the absorption spectrum of vitam in K 2 in ethyl alcohol solution after standing 40 hours in a glass bo ttle exposed to diffuse light. The band has undergone little change in absorption intensity, but it is spread out and shifted tow ard the Schumann ultraviolet. Curve I, Fig. 4, shows a comparable spectrum of vitam in Ki Ew ing, V andenbelt, and K am m 349 in hexane after standing 70 hours in daylight. This band has also shifted tow ard a lower wave-length, retaining a shape similar to C urve II. Some p a rt of th e broad band with the maximum a t X 325 mju seems to be remaining. mpi no F i g . 4 . T h e a b s o r p t io n s p e c t r a o f v it a m in s K i (C u r v e I) a n d K 2 (C u r v e I I ) a f t e r e x p o s u r e t o v is ib l e li g h t . F i g . 5 . A b s o r p t io n s p e c t r a o f v it a m in K i in h e x a n e . T h is p r e p a r a tio n w a s o b t a i n e d b y t h e h y d r o ly s is o f th e d ia c e t a t e o f d ih y d r o v it a m in K i. T h e r e a d in g s o f C u r v e I w e r e t a k e n 15 m in u t e s a ft e r s o lu t io n in h e x a n e ; C u r v e I I , 1 h o u r la t e r . I t is evident th a t the vitam in is seriously affected by light. I n s t a b i l it y o f V i t a m i n i n H e x a n e S o lu tio n —Fig. 5 shows the absorption spectra of vitam in K i obtained by hydrolysis of the diacetyl dihydro derivative (regenerated vitam in K i). Curve I was taken as soon as possible after solution in hexane (15 m inutes); Curve I I , 1 hour later. The E ^ n . values of fine structure peaks of the m ajor band have fallen during the hour. U ltrav io let A bsorption of V itam ins 350 To obtain maximum absorption of the vitam in it is necessary to determine the curve im m ediately after solution of th e sam ple. E ffe c t o f R e d u c tio n —Fig. 6 shows absorption spectra of th e com  pounds obtained by the catalytic reduction (Adam s-Shriner) of vitamins Ki and K 2 (iVlcKee, Binkley, IVfacCorquodale, T hayer, and Doisy, 1939). Like the vitam ins, the vitam in K x deriv ativ e has the higher extinction coefficient. There is an absorption b and from X 255 to 275 m/x in each case, with fine s tru c tu re m axim a a t X 260 and 270 m/x. The wave-lengths of these narrow peaks correspond exactly with the wave-lengths of two of th e peaks of the respective vitamins. Reduction of the vitam ins has d estroyed m/i F ig . 6 F ig . 7 6. A b s o r p tio n s p e c t r a o f r e d u c tio n p r o d u c ts o f v i t a m i n s K i ( C u r v e I) a n d K 2 (C u r v e I I ) , T h e s e c o m p o u n d s s t i l l p o s s e s s e d a q u in o n o i d st r u c tu r e . F i g . 7 . A b s o r p tio n s p e c tr a o f th e d ia c e t a t e s o f d ih y d r o v i t a m i n s K i (C u r v e I) a n d K 2 (C u r v e I I ) . F ig . the left half of the major absorption band w ith its two narrow maxima, leaving unchanged the two fine stru c tu re m axim a a t X 260 and 270 m/x. D iacetates o f D ih y d ro V ita m in s K x a n d K 2—To help establish th e structure of the vitamins, the diacetates of the dihydro derivatives of vitamins K x and K 2 were examined. Fig. 7 shows th e ir re spective absorption curves. In place of the broad bands of the vitam ins, w ith fine stru c tu re , there is a narrow band farther in the ultraviolet, w ith a m axim um at X 232 m/x. The E values of the vitam in Kx and K 2 diac etates are 1600 and 1300 respectively. Ew ing, V andenbelt, an d K am m 351 1 , /+-N a.p h th o q u in o n e a n d D ia c e ta te o f N a p h th o h y d r o q u in o n e —■Fig. 8 shows curves of 1 ,4-naphthoquinone and th e diacetate of its dihydro reduction product. As in the case of the corresponding vitam in K i and K 2 derivatives, the E value of the absorption m axim um of the diacetate is greater th a n th a t of the quinone. Also, it is shifted to the farther ultraviolet, so th a t w ith loss in inten sity of the radiation in th a t region, no record was obtained on E astm an No. 33 emulsion. ■/ 5 /< m. my. F i g . 8 . A b s o r p t io n s p e c t r a o f 1 ,4 - n a p h t h o q u in o n e (C u r v e I) a n d t h e d ia c e t a t e o f n a p h t h o h y d r o q u in o n e (C u r v e I I ) . I t is interesting th a t the maximum of the less intense broad band is shifted to a lower wave-length. The band of vitam in Ki is shifted from X 325 to X 283 m/x in the diacetate, from X 328 mp in vitam in K 2 to X 284 m/x in its diacetate, and from X 325 to X 285 m/x in the diacetate of the naphthohydroquinone. V i t a m i n K \ O b ta in ed b y H y d r o ly s is o f D ia c e ty l D ih y d r o D e riv a  tive—T he diacetate of dihydro vitam in Ki was hydrolyzed in order to regenerate the vitam in. The curve is given in Fig. 5. T he m ain band lies in the same position as th a t of the original vitam in Ki, as do the fine stru ctu re m axima a t X 243, 249, 260, and 270 m/x. Also, th e less intense band w ith the maximum in th e longer wave-lengths reverts to its vitam in K i position. The 352 U ltrav io let A bsorption of V itam ins E values of bo th bands of th e regenerated v itam in are sim ilar to those of th e original vitam in. 1 , 4 .-N a p h th o q u in o n e a n d D e riv a tiv e s —Fig. 9 shows ab sorption curves of 1 ,4-naphthoquinone, 2-m ethyl-l ,4-naphthoquinone, 2 , 3-dim ethyl-1 ,4-naphthoquinone, and 2 -e th y l-l, 4 -n ap h th o  quinone, all in hexane. All four compounds have principal absorption bands in th e sam e ultraviolet region with a less intense b and (not shown) w ith th e maximum near X 325 m/u. F i g . 9. A b s o r p tio n s p e c t r a o f 1 ,4 - n a p h t h o q u in o n e (C u r v e I ) , 2 - m e t h y l 1 . 4 -n a p h t h o q u in o n e (C u r v e I I ) , 2 , 3 - d i m e t h y l - l,4 - n a p h t h o q u in o n e (C u r v e I I I ) , a n d 2 - e t h y l - l , 4 -n a p h t h o q u in o n e (C u r v e I V ) , in h e x a n e . The introduction of substituents exerts a notew orthy effect on th e fine structure of the m ajor band. The 1 ,4-naphthoquinone, w ith no substituent, has no fine structure. In tro d u ctio n of a m eth yl group in the 2 position produces three m axim a. Tw o m eth yl groups cause four maxima, as does an ethyl group in th e 2 posi tion. The 2 , 3-dim ethylnaphthoquinone has an absorption spec tru m singularly like th a t of the vitam in (Fig. 1), its fine stru c tu re maxima being a t X 243, 248, 259, and 269 m/*, respectively. U 4 -B e n zo q u in o n e Fig. 10 shows the curve of 1 ,4 benzoquinone, in hexane. I t is very similar in shape to th a t of the 1.4-naphthoquinone. The m ajor absorption is in the region X 230 to 260 m*z, with a maximum a t X 241 mju. T he longer w ave length band is very weak. Ew ing, V andenbelt, a n d K am m 353 2 , 8 D isu b stitu tio n P rodu cts o f 1 ^ -N aph th oqu in on e; A d d s Fig. 11 (Curve I) shows the absorption curve of the quinone acid, C13H10O4, obtained by the oxidation of vitamin X x. The general features of the curve are very similar to those of the naphthoquin— F i g . 10. A b so r p tio n sp ec tr u m o f 1 , 4 -b e n z o q u in o n e in h ex a n e F i g . 1 1 . A b s o r p t io n s p e c t r a o f 2 - m e t h y l- l ,4 - n a p h t h o q u in o n e - 3 - a c e t ic a c id (C u r v e I ) o b t a in e d fr o m o x id a t io n o f v it a m in K i, a n d o f a s y n t h e t ic p r e p a r a t io n o f 2 - e t h y l - l , 4 -n a p h t h o q u in o n e - 3 - a c e t ic a c id (C u r v e I I ) . B o t h c o m p o u n d s w e r e d is s o lv e d in e t h y l a lc o h o l. ones studied, w ith absorption from X 240 to 275 m/x, broken up into a ta ll maximum a t 244 to 248 m/x, and a shorter one a t X 260 to 270 m/x. There is also a less intense band from X 315 to 340 m/x w ith a maxim um a t X325 m/x. The compound, 2 -e th y l-l, 4- 354 U ltraviolet A bsorption of V itam ins naphthoquinone-3-acetic acid (Curve II) exhibits the same m ax ima, general shape of the absorption curve, and sim ilarity m extinction coefficients. This indicates a similar m olecular arrangement in the compounds. D IS C U S S IO N Doisy and his associates (McKee, Binkley, M acCorquodale, Thayer, and Doisy, 1939) have shown vitam ins K i and K 2 to possess quinone structures, whereas our comparison of the u ltra  violet absorption curves with those of known quinones has led to the conclusion th a t the vitamins are derivatives of naphthoquinone rather than of benzoquinone. There is evidence, moreover, th a t one pair of maxima, in the region X 240 to 250 m/x, is due pri marily to the benzene nucleus, whereas the other, in the region X 260 to 270 m/x, is associated with the quinonoid structure. This is shown by our measurements of the reduction products of the vitamins. Both vitamins upon catalytic reduction yield colorless products which upon exposure to air are oxidized to yellow compounds presumably possessing quinone structures (McKee, Binkley, MacCorquodale, Thayer, and Doisy, 1939). Absorption m easure ments showed th a t the maxima a t X243 and 249 m/x had been elimi nated (Fig. 6), whereas the maxima a t X 260 and 270 m/x remained. Dam et al. (1939) found the absorption curve of v itam in K i to have four principal maxima at X 248, 261, 270, and 328 m/x, and an extinction coefficient of a t X 248 m/x of 280. In Fig. 1 a sharp maximum is noted a t X 243 m/x and th e value of the E {cm. at X 249 m/x is 540, practically double th a t reported by Dam . It is obvious th a t our much higher value of the extinction coeffi cient and very pronounced maximum at X 243 m/x are due to the higher purity of the product. The difference between the absorption curves of the hydro genated vitamin K x observed by D am et al. (1939) and ourselves is difficult to understand. If the conception of the St. Louis group is correct, the only points of reduction are the double bond m the side chain, the aromatic rings, and the quinonoid linkages. Dam and coworkers found, after reduction, maxima a t X 248, 261, and 270 m/x but a disappearance of the band a t 328 m/x! Since the 328 mM band or one analogous to it is present in all Ew ing, V andenbelt, a n d K am m 355 of the 1,4-naphthoquinones and the diacetates of the naphthohydroquinones which we have studied, it seems th a t this absorp tion m ust be due to the ring structure. However, hydrogenation of the non-quinonoid ring of vitam in K i causes loss of the maxima a t X 243 and 248 as well as 328 m/x (see Fig. 6). Obviously the discrepancy cannot be cleared up until D am and his collaborators furnish additional evidence on the n atu re of the compound pro duced by th e ir process of reduction. The EX^m . of 1 ,4-naphthoquinone had a value of 1090 a t a sharp maxim um of X 245 m/x and a value of 150 a t X 328 m/x where the band was very broad. The hexane solution of 2-m ethyl-1 ,4-naph thoquinone gave the following approxim ate values of E : 1150 a t X 244 m/x, 1145 a t X 253 m/x, 975 a t X 264 m/x, and 180 a t X 325 to 328 m/x. An extinction coefficient of E 810 was found for th e m inim um a t X 260 m/x (Fig. 9). M acbeth, Price, and W inzor (1935) found only two maxima for these compounds, one a t X 246 m/x and another a t X 334 m/x as well as a m inimum a t X 285 m/x. Fieser, Bowen, e t a l. (1939) rep ort a maximum a t X 250 m/x. We believe th a t this portion of the ultraviolet absorp tion spectrum of 2 -m e th y l-l, 4-naphthoquinone consists of distinct m axim a a t X 244 and 253 m/x, and a m inim um a t X 248 m/x. In a more recent publication Fieser, Campbell, and F ry (1939) give th e absorption curve for 2 , 3-dim ethyl-l ,4-naphthoquinone. A comparison of this curve with the one (Fig. 9) obtained in this lab oratory shows th a t Fieser failed to observe the fine structure of the m ain absorption bands, probably because of his use of ethyl alcohol rath e r th a n hexane as the solvent. The fine struc tu re of the absorption shows th a t the maxima of 2 ,3-dimethyl1 ,4-naphthoquinone and vitam ins K x and K 2 are almost identical. The absorption spectra of the diacetates of dihydro vitam ins K i and K 2 are very similar in th a t sharp maxima occur a t X 231 m/x for the vitam in K 2 derivative and a t X 230 m/x for the cor responding vitam in K x compound. The EXcm. f°r tbe vitam in K i derivative was 1600 and for the corresponding compound of vitam in K 2 the value was 1300. In both of these cases extinction coefficients of 100 were found a t X 285 m/x. I t is noted th a t the diacetate of dihydro-1,4-naphthoquinone shows substantially the same type of absorption curve. The diacetate band (Fig. 7) is simple, and in consideration of the 356 U ltrav io let A bsorp tio n of V itam in s com parative stability of the diacetates, it would seem th a t these compounds have th e proper qualifications as reference stan d ard s for th e vitam ins. From these observations it m ay be concluded th a t vitam ins K i and K 2 are derivatives of 2 , 3-dim ethylnaphthoquinone, th a t th e y contain the stru ctu re O o and th a t the side chains contain no conjugated double bonds. A d d e n d u m , — A fte r t h i s a r t ic le h a d b e e n s u b m it t e d fo r p u b li c a t i o n , w e h a d a n o p p o r t u n it y o f m e a s u r in g t h e a b s o r p t io n s p e c t r u m o f D r . D o i s y ’s s y n t h e t ic v it a m in K i. M a x im a w e r e o b s e r v e d a t X 243, 249, 260, 2 6 9 , a n d 1 9o 325 m/x, t h e c o r r e s p o n d in g E \ Cm. v a lu e s b e in g 410, 4 2 5 , 395, 3 9 5 , a n d 75. T h e s e v a lu e s a re in g o o d a g r e e m e n t w it h t h o s e o b t a in e d in o u r m e a s u r e  m e n t s o n p u r e v it a m in K i fr o m a lf a lf a a n d s h o w t h a t t h e s y n t h e t i c p r o d u c t is id e n t ic a l w it h t h e n a tu r a l. B IB L IO G R A P H Y D a m , H ., G e ig e r , A ., G la v in d , J ., K a r r e r , P ., K a r r e r , W ., R o t h s c h i ld , E ., a n d S a lo m o n , H ., H e l v . c h im . a c ta , 22 , 31 0 (1 9 3 9 ). F ie s e r , L . F ., B o w e n , D . M ., C a m p b e ll, W . P ., F r y , E . M ., a n d G a t e s , M . D . , J r ., J . A m . C h e m . S o c . , 61 , 1926 (1 9 3 9 ). F ie s e r , L . F ., C a m p b e ll, W . P ., a n d F r y , E . M ., J . A m . C h e m . S o c . , 61 , 2206 (1 9 3 9 ). M a c b e t h , A . K ., P r ic e , J . R ., a n d W in z o r , F . L ., J . C h e m . S o c . , 3 2 5 (1 9 3 5 ). M a c C o r q u o d a le , D . W ., B in k l e y , S . B ., M c K e e , R . W ., T h a y e r , S . A ., a n d D o is y , E . A ., P r o c . S o c . E x p . B i o l , a n d M e d . , 40 , 4 8 2 (1 9 3 9 ). M c K e e , R . W ., B in k le y , S . B ., M a c C o r q u o d a le , D . W ., T h a y e r , S . A ., a n d D o is y , E . A ., J . A m . C h e m . S o c . , 61 , 1295 (1 9 3 9 ). Reprinted from Analytical Edition In d u s t r ia l and E n g in e e r in g C h e m is t r y Vol. 12, Page 639, November 15, 194/0 Spectrophotometric Determination o f Vitamin A Critical Study o f Applicability to Fish Liver Oils D. T. EWING AND J. M. VANDENBELT1, Michigan State College, East Lansing, Mich., A. D. EMMETT AND AND O. D. BIRD, Parke, Davis & Company, Detroit, Mich. T IS th e general opinion th a t th e determ ination of vitam in A in fish fiver oils a n d other p roducts by physical m ethods has shown definite prom ise of giving satisfactory results. The several factors in th e spectrophotom etric m ethod have there fore been studied in a n effort to increase our knowledge of its accuracy, in th e hope th a t it m ay come in to m ore regular application as a q u a n tita tiv e procedure. T he am o u n t of vitam in A in a fish fiver oil m ay be esti m ated q u a n tita tiv e ly in a num ber of w ays. T he m ost common are th e biological, th e colorim etric, an d th e spectrophotom etric m ethods. T h e historical developm ent of th e subject has been covered v ery thoroughly b y M unsell (6). Suffice it to sta te th a t th e first biological assays were carried o u t b y D rum m ond an d C ow ard (#), th e first color reaction using arsenic trichlo ride by Rosenheim and D rum m ond (7), and th e color reaction using antim o n y trichloride by C arr and Price (I). Spectro photom etric m easurem ent in th e visible range a t 698 m/x was m ade first by D rum m ond and M orton (5). T akahashi et al. (8) first reported on th e selective absorption charac teristics of vitam in A in the u ltraviolet, while th e absorptive m axim um in th a t region was established a t 328 m/x by M orton an d H eilbron (5). C om parison of th e d a ta b y these m ethods has shown, in th e h an d s of various workers, b o th small and large discrepancies. I 1 P a r k e , D a v is <fc C o m p a n y R e s e a r c h F e llo w in p h y s ic a l c h e m is t r y . Because it is generally recognized th a t th e spectrophoto m etric m ethod is intrinsically capable of m aking very ac curate determ inations w ith b oth organic and inorganic sub stances, th e authors have m ade a careful stu d y of th e factors involved w ith respect to vitam in A in fish liver oils. T his has included the solvent and its action, th e effect of variation in cells, th e application of L am b ert’s and B eer’s laws, th e im  portance of instrum ent adjustm ent, th e fight source, and in  cidentally th e practical lim its of a few interfering substances. Method and Plan H aving investigated th e above factors, th e next step w as to te s t th e precision of the technique. A series of m any de term inations w as m ade on each of several fish fiver oils to ascertain th e spread or variation in th e m axim um extinction coefficient or E \ v a l u e a t 328 m/x (325 to 328 m/x) on each sample. T his was taken as an indication or index of th e reproducibility of th e technique. T he spectrophotom etric setup consisted of a B ausch & Lom b sector photom eter w ith a quartz optical system and a B ausch & Lom b m edium quartz spectrograph. A condensed spark be tw een tu n gsten steel electrodes supplied the ultraviolet radiation, the energy being generated by a B ausch & Lomb 450 VA induc tance transformer. E astm an N o. 33 photographic plates were VOL. 12, N O . 11 IN D U S T R IA L A N D E N G IN E E R IN G C H E M IS T R Y 640 T I. able E ffect of T im e r ! *7o ^ 1 cm . T im e E la p s e d a fte r S o lu tio n M in. 5 15 30 60 3 1 .0 3 1 .6 3 1 .6 3 1 .2 Hours 2 3 4 .5 7 .2 5 25 47 T able 2 9 .7 2 8 .1 2 6 .6 2 5 .6 2 5 .7 2 3 .2 II. C o m p a r is o n I so pro py l, a n d A of E x t in c t io n bso lu te E thyl C o e f f ic ie n t s A lcohol in .---------- e }1%cm . V a lu e s---------- . T y p e of F ish L iv e r O il H a lib u t (3898) H a lib u t (24 5 8 ) H a lib u t (2 6 6 8 ) H a lib u t (2 6 2 8 ) H a lib u t (2 9 7 8 ) M ix tu r e (4288)' C o d (9 7 5 8 ) I so p r o p y l, 1 A b s o lu te e th y l, 11 R a tio , I I/I 3 1 ,5 2 6 .8 3 1 .0 3 1 .3 3 1 .8 7 8 .7 1 .5 9 3 1 .6 2 7 .2 3 2 .0 3 0 .9 3 1 .6 8 1 .4 1 .6 2 1 .0 0 3 1 .0 1 5 1 .0 3 2 0 .9 8 7 0 .9 9 4 1 .0 3 4 1 .0 1 9 used throughout. T hey were processed by N o. D -l developer at 18° C. Samples of the fish liver oils were weighed ou t and dissolved in redistilled isopropyl alcohol (E astm an). The spectrophoto m etric ^ value was then determ ined. T he first photo graphic exposure was made 10 m inutes after preparation of the solution, and other exposures followed im m ediately. The plate was processed under carefully controlled and reproducible condi tions. When the em ulsion was com pletely dry, the “ isodensity†or “ reversal†points were marked on the glass side. T he log I 0/ I of the absorptive maxim um was then deter mined, where 70 equals incident light (100 per cent), and I equals per cent of light tran sm itted a t th e wave length of the absorptive m aximum. The extinction coefficient of the ab sorptive maxim um was calculated from th e form ula: E}*> 1 om. = d X 1 X l o g / p // % concentration where d equals length of light path through the solution in cm., and th e per cent concentration equals 100 tim es the weight of sample in gram s divided by th e milliliters of solvent. are given in T able I I I . I t is evident th a t th e cells should always be checked against each other, in order to elim inate variations due to errors in th e length of cells and differences in th e absorption of th e end pieces. Extrem e cleanliness of the cell wall is essential, particularly w ith reference to oily residues rem aining from th e previous sample. Ju st before each run, th e q u a rtz surfaces should be wiped w ith fresh lens paper. Perpendicularity of th e ends of th e cells to the lig h t beam is necessary, and th e cells m ust alw ays occupy exactly th e sam e position w ith respect to th e optical p ath . L a m b e r t ’s L a w . A h alibut liver oil sam ple (908,871) was dissolved in isopropyl alcohol and diluted so th a t th e absorp tion of th e solution was suitable for an E \ ^ m^ determ ination in the 10-mm. cells. (The log I a/I was equal to approxi m ately 1.0.) The value of this concentration was ob tained also in each of the cell lengths 1, 2.5, 5, 20, an d 50 mm. T he solvent p a th sh utter readings were changed to conform w ith th e various optical densities of th e different length cells. The E\%m_ values (Table IV) determ ined under these conditions did n o t change w ith cell length, except for small variations ascribable to the m ethod, thereby com plying w ith th e requirem ents of L am bert’s law. B e e r ’s L a w . B y proportionally increasing th e concen tration of th e fish liver oil solution in th e cells shorter th a n 10 mm., and decreasing it for exam ination in th e longer cells, the 7 ? ! ^ . values (Table IV) a t a wide range of concentra tion were obtained. Over a range of 50 tim es difference in concentration, th e variation in E \ ^ mm values for th e oil w as no greater th an th a t obtained when a series of a sim ilar n um  ber of determ inations w as m ade a t one concentration and one cell length. Therefore, both L am bert’s and B eer’s laws are valid for dilutions of vitam in A w ithin th e accuracy of th e m ethod used. L ig h t S o u r c e . Tw o sources of ultraviolet radiation were used. T he first was a condensed spark between w edge-tipped tungsten alloy steel electrodes. U sing a slit w idth of 40 m i crons, the am ount of light em itted by th is spark gave, in 15 T able III. C o m p a r is o n M o l e c u l a r E x t in c t io n C of C ells o e f f ic ie n t s M by of P easurem ent o t a s s iu m S o lv e n t C ell S o lu tio n C e ll B & L. 1 B . & L. 2 B & L. 2 B .& L . 1 7 .0 0 7 .0 0 Fundamental Factors Z eiss 1 Z eiss 2 Z eiss 2 Z eiss 1 6 .9 7 6 .9 7 E f f e c t o f T im e o n V a l u e i n S o l u t io n . Table I gives th e values of h alib u t liver oil (3898) in iso propyl alcohol a t varying tim e intervals after addition of the solvent to th e sample. T he absorption reached a maximum value after 5 m inutes and remained fairly constant for about an hour, after which there was a significant decrease. C o m p a r is o n o f S o l v e n t s . Table I I gives extinction co efficients a t 328 mju in isopropyl and absolute ethyl alcohol. T he ratios, being near unity, show th a t these alcohols can be used interchangeably in th e determ ination of the E \ ^ m value of fish liver oils. Isopropyl alcohol was used as a solvent throughout this work because it is a superior solvent for oils. This is an ad vantage in assaying oils of low potency. Z eiss 3 Z eiss 4 Z eiss 4 Z eiss 3 6 .6 7 7 .0 0 I n flu en c e of T y pe s , C l e a n l in e s s , a n d P o s it io n of Bausch & Lomb 10-mm. cells were used in p a rt of this work and Zeiss 10-mm. cells in the other p art. In order to ascertain the relative absorption of the end pieces of the several cells, the molecular extinction coefficient of potassium n itra te was determ ined in both types. The values obtained C e lls . T able IV. C o n c e n tr a tio n , % 0 .3 1 4 0 .1 2 7 0 .0 6 3 0 .0 3 1 4 0 .0 1 5 7 A p p l ic a t io n o f L S u m m a r y (4 0 t e s t s ) : of it r a t e = 301 mfi a m b e r t 's a n d ( H a lib u t liv e r o il N o . 9 0 8 ,8 7 1 ) -------ju en gtn ot <JeU— 1 2 .5 5 10 mm. mm. mm. mm. 1 1 cm . 3 0 .1 2 8 .1 2 9 .9 2 8 .1 2 9 .1 2 8 .7 2 8 .9 3 0 .1 2 8 .2 2 8 .9 2 8 .3 2 9 .2 2 9 .8 2 9 .6 2 8 .8 2 9 .6 2 9 .8 2 9 .6 2 8 .1 2 8 .9 2 9 .7 2 9 .5 2 9 .1 2 9 .1 2 7 .7 2 9 .6 2 8 .2 2 9 .9 0 .0 0 6 3 N — ._ 2 9 .1 2 9 .1 M ean M axim um range, % M axim um d e v ia tio n , % B e e r ’s 20 mm. L aw s 50 mm. 28 .5 2 9 .8 2 9 .7 2 9 .1 2 8 .3 2 7 .7 2 8 .0 * 2 8 .4 2 9 .9 2 9 .9 2 9 .0 4 8 .3 4 .8 N O V E M B E R 15, 1940 T a b le V. E \ V a lu e s T est N o. S p e ctro  gram N o. C oncen  tr a tio n , % 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 390a 390b 391a 391b 392a 392b 393a 393b 394a 394b 395a 395b 396a 396b 397a 397b 398a 398b 399a 399b 400 a 400b 401a 401b 402a 402b 403a 403b 404a 404b 405a 405b 406a 406b 407a 407b 408a 408b 409a 409b 410 0 .0 3 3 0 op H a lib u t L iv e r O il 38 9 8 a t 3 2 8 m /x S h u tter R e a d in g Log Io/I 328 mp E 1, ' cm ° . 9 -1 0 9 -1 0 8 -9 9 -1 0 1 1 -1 2 1 0 -1 1 1 1 -1 2 1 1 -1 2 1 0 -1 1 9 -1 0 10 9 -1 0 1 0 -1 1 1 0 -1 1 12 1 1 -1 2 1 0 -1 1 1 0 (-) 9 -1 0 9 1 1 -1 2 1 1 -1 2 9 9 -1 0 1 1 -1 2 1 2 -1 3 11 1 0 -1 1 1 1 -1 2 1 1 -1 2 12 1 1 -1 2 1 2 -1 3 1 2 -1 3 1 0 -1 1 1 0 -1 1 1 1 -1 2 1 1 -1 2 1 2 -1 3 1 2 -1 3 1 0 -1 1 1 .0 2 1 .0 2 1 .0 7 1 .0 2 0 .9 4 0 .9 8 0 .9 4 0 .9 4 0 .9 8 1 .0 2 1 .0 0 1 .0 2 0 .9 8 0 .9 8 0 .9 2 0 .9 4 0 .9 8 1 .0 1 1 .0 2 1 .0 5 0 .9 4 0 .9 4 1 .0 5 1 .0 2 0 .9 4 0 .9 0 0 .9 6 0 .9 8 0 .9 4 0 .9 4 0 .9 2 0 .9 4 0 .9 0 0 .9 0 0 .9 8 0 .9 8 0 .9 4 0 .9 4 0 .9 0 0 .9 0 0 .9 8 3 0 .9 30 9 3 1 .8 30 4 3 1 .3 32 6 3 0 .9 3 0 .9 3 1 .2 3 2 .5 3 1 .2 3 1 .9 3 2 .0 3 2 .0 3 1 .5 3 2 .2 3 0 .8 3 1 .8 3 0 .5 3 1 .4 3 1 .5 3 1 .5 3 2 .6 3 1 .6 3 2 .4 3 1 .0 3 0 .8 3 1 .4 3 1 .3 3 1 .3 3 1 .7 3 2 .4 3 1 .2 3 1 .2 3 2 .2 3 2 .2 3 1 .1 3 1 .1 3 0 .5 3 0 .5 3 1 .8 0 .0 3 3 6 0 .0 3 0 0 0 .0 3 0 4 0 .0 3 1 4 0 .0 3 2 0 0 .0 3 0 6 0 .0 2 9 2 0 .0 3 1 8 0 .0 3 3 4 0 .0 2 9 8 0 .0 3 2 2 0 .0 2 9 0 0 .0 3 1 2 0 .0 3 0 0 0 .0 2 9 0 0 .0 2 8 8 0 .0 3 0 4 0 .0 3 0 2 0 .0 2 9 6 0 .0 3 0 8 S u m m a r y (4 1 t e s t s ) : 641 A N A L Y T IC A L E D I T I O N M e a n v a lu e M a x im u m r a n g e , % M a x im u m d e v ia t i o n , % stearic acid, has an value of b u t 0.03, which is of no practical consequence. T he m etallic salts of fa tty acids, for th e m ost p art, are in soluble in isopropyl alcohol, and are elim inated thereby. C opper oleate, although slightly soluble, has a deep green color, and can be easily detected. On th e other hand, ferric oleate is soluble in isopropyl alcohol and has significant ab sorption a t 328 m p a t concentrations as low as 0.001 per cent. I t can be detected by m eans of the thiocyanate te st when present to th e extent of only 0.0005 per cent. I t is obvious th a t if th e chemical te st is negative for iron, th e E\^°m value has n o t been significantly increased. Application of Spectrophotometer to Fish Liver Oil S pec tr o ph o to m eter S e r ie s . T able V gives all values ob tained w ith the B ausch & Lom b spectrophotom eter in a rep resentative series of tests on halib u t fiver oil 3898, th is being typical of th e seven fish fiver oils studied in th e long series on reproducibility of technique. T he d a ta show, respectively, th e te st num ber, spectrogram num ber, solution concentra tion, sh u tter reading of th e “isodensity†point a t th e 328 m/x absorptive m axim um on th e finished spectrogram , log I 0/ I value or solution density corresponding to th e sh u tte r reading a t th e w ave-length m axim um , and th e calculated extinction coefficient a t 328 m/x. T he sum m ary a t th e bottom of T able V shows th a t the m axim um range for these 41 tests was 7.0 per cent—th a t is, the difference between th e highest and th e lowest values was 7.0 per cent of th e m ean. T he m axim um deviation, or largest per cent difference between any single value and th e m ean, was 3.5 per cent. T able V I gives th e spectrophotom etric values ob tained for th e other oils studied in th e long series. On account of th e space required, it is n ot possible to present th e detailed d ata, as in the representative series of h alib u t fiver oil 3898. I t is easy to note, however, th e excellent reproducibility and the to tal absence of eccentric values, although th e oils ex am ined have a wide range of potency. 3 1 .4 6 7 .0 3 ,5 seconds, a su itab le exposure a t 328 m/x for an E astm an 33 plate, if th e o p tical d en sity of th e absorbing solution was equal to approx im ately 1.0. Second, a H ilger hydrogen discharge tu b e w as em ployed. A lthough requiring som ew hat longer ex posure tim e th a n th e spark, it w as found to be superior from I n f l u e n c e o f N u m b e r o f D e t e r m in a t io n s o n R e p r e  th e sta n d p o in t of b o th convenience an d reproducibility. In s e n t a t iv e M e a n 1?}^. V a l u e s . I t is obvious th a t all the eith er case only w hen th e various p a rts of th e spectrophoto E i^m. values for halib u t liver oil 3898, presented in th e last m etric se tu p w ere in exact optical alignm ent was it possible to o b ta in o p tim u m w orking conditions as to m in im u m exposure tim e an d transm ission in to th e low u ltrav io let region. T a b le V I. E \ % m _ V a l u e s o f F i s h L i v e r O i l s a t 328 m /x E x p o su r e L a t it u d e a n d D e n s it y of T h e processing of th e photographic p late w as carried o u t under w ell-defined conditions. Using an E astm an calib rated 21-step ta b le t an exposure b y d irect c o n ta c t w as m ade on each plate. T he den sity vs. log exposure curve gave a gam m a of 0.9 for th e p late as processed u n d er care fully controlled conditions. T h e isodensity points h ad a v alue of approxim ately 0.6 as read on an E a stm a n transm ission den sitom eter. I n t e r f e r in g Su bsta n c es. Some few com pounds com m only found w ith fish liver oils exhibit ab so rp tio n a t 328 m/x. N atu rally occurring are th e u n sa tu ra te d long-chain f a tty acids, oleic a n d palm itic, w hich have extinction coefficients (A 'i^ m.) of 0.88 and 0.63, respectively. In low potency oils, such as cod liver oils, these should be rem oved by saponification. T he sa tu ra te d com pound, P h o t o g r a p h ic P l a t e (A s d e t e r m in e d b y t h e s p e c t r o p h o t o m e t e r ) . H a lib u t (1 6 ,5 1 9 ) 5 0 .7 5 0 .5 51 .0 51 .0 5 1 .0 4 9 .5 5 0 .7 5 0 .7 4 9 .6 5 0 .0 5 0 .0 5 1 .0 5 0 .5 4 9 .0 4 9 .8 5 0 .5 4 9 .0 4 9 .8 4 9 .8 5 0 .8 4 9 .2 4 9 .0 4 9 .0 4 8 .7 5 0 .7 4 9 .6 4 8 .7 4 9 .0 5 1 .0 51 .0 5 1 .0 5 1 .0 5 1 .0 4 9 .0 4 9 .0 5 1 .2 5 1 .2 5 0 .2 5 1 .2 4 9 .0 5 0 .0 M ean 5 0 .1 2 M ix e d (4 2 8 8 ) 7 7 .6 7 4 .5 7 5 .2 7 8 .4 7 9 .8 8 2 .0 8 1 .0 7 4 .8 7 9 .4 7 8 .0 7 9 .6 7 9 .6 7 8 .9 7 8 .9 8 0 .8 7 7 .6 7 8 .3 7 5 .3 7 7 .6 8 0 .9 7 5 .3 7 8 .2 8 1 .6 7 8 .3 7 7 .6 7 9 .4 7 8 .4 7 8 .4 7 9 .0 7 9 .0 7 9 .7 79 .4 7 9 .9 7 5 .4 7 9 .0 7 9 .8 7 8 .3 8 0 .0 8 1 .7 8 0 .0 7 9 .0 7 9 .0 7 8 .3 7 6 .7 7 9 .0 8 0 .7 7 8 6S M ix e d (1 6 ,3 1 9 ) 12 5 130 130 127 127 12 7 127 129 12 9 12 9 126 129 126 129 128 127 131 127 127 131 131 129 129 132 134 129 125 125 132 126 132 132 128 126 128 131 126 127 134 129 127 133 12 8 .6 H a lib u t M ix e d ( 1 8 ,8 6 9 ) (1 6 ,9 4 9 ) 1 5 .0 1 4 .8 1 5 .2 1 5 .4 1 5 .2 1 4 .8 1 4 .8 1 5 .0 1 5 .0 1 5 .3 1 5 .4 1 4 .8 1 5 .0 1 4 .8 1 5 .4 1 5 .3 1 5 .4 3 9 .7 4 0 .5 3 9 .8 4 0 .5 4 0 .2 3 9 .2 4 0 .8 4 0 .0 4 0 .0 4 0 .8 3 9 .2 3 9 .2 4 0 .0 3 9 .2 4 0 .0 3 9 .2 3 9 .2 1 5 .0 5 3 9 . So C od “ N o n s a p .†(9 7 5 8 ) —C od0758) 1 .5 6 1 .5 6 1 .5 1 1 .5 7 1 .5 4 1 .5 1 1 .5 9 1 .5 9 1 .5 6 1 .6 3 1 .6 3 1 .5 9 1 .5 3 1 . 56 1 .5 5 1 .6 3 1 .5 6 1 .5 6 1 .6 3 1 .6 3 1 .5 2 1 . 55 1 .5 8 1 .5 1 1 . 56 1 .6 0 1 . 56 1 .5 3 1 .5 6 1 .6 3 1 .6 0 1 .6 0 1 .6 0 1 .5 6 1 .6 0 1 .6 0 1 .6 1 1 . 58 1 .5 7 1 .5 7 1 . 63 1 .6 2 1 .5 7 1 .5 3 1 . 55 1 .6 4 1 .5 7 1 .6 4 1 .6 2 1 .5 S 1 . 57 1 .5 7 1 . 5S9 1 .5 7 1 .5 7 1 .6 7 1 .6 4 1 .5 5 1 .5 5 1 .6 7 1 .6 5 1 .6 5 1 .6 7 1 .6 3 1 . 63 1 .6 3 1 .6 6 1 .5 9 1 .6 0 1 .5 5 1 .5 7 1 .6 3 1 .5 7 1 .5 9 1 .6 3 1 .5 9 1 .6 2 1 .5 9 1 .3 4 1 .3 1 1 .3 2 1 .3 1 1 .2 9 1 .3 1 1 .2 8 1 .3 7 1 .3 7 1 . 35 1 .3 7 1 .3 5 1 .3 7 1 .4 1 1 .4 1 1 .4 1 1 .4 1 1 .4 0 1 .4 1 1 .2 6 1 .2 8 1 .2 8 1 .3 4 6 IN D U S T R IA L A N D E N G IN E E R IN G C H E M IS T R Y 642 — T able V II. V ■■■H a l ib u t L 328 m M alues of at iv e r O il through th e solution. From th e scale reading, calibrated as th e log Io/I, the value of the sam ple can be calculated. 3898 V it a m e t e r S e r i e s . Using this instrum ent, several series of determ inations were m ade on fish liver oil sam ples corre sponding to those run on th e spectrophotom eter (Table V) for oil 3898. T able V II contains th e results obtained w ith the vitam eter on this same oil. These include th e te s t num ber; the per cent concentration; th e scale reading, representing the average of 10 observations on each dilution, m ade by tw o operators; and th e calculated A'l'mn. value for each series of 22 tests. (The authors gratefully acknowl edge th e assistance of Cyrill J . Cam pbell for his cooperation in th e vitam eter determ inations.) The m eans of the E \ 7^ values by th e tw o operators, working independently and on unknow n concentrations, are n ot significantly different. T he m ean E \ ^ values by operators 1 and 2 are 32.00 and 32.40, respectively, w ith a difference of only 0.40 or 1.2 per cent of the mean. (D eterm in ed b y th e v itam eter) T e st N o. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 C oncentratio n , % 0 .0 2 2 2 0 .0 2 1 1 0 .0 1 8 9 0 .0 2 1 1 0 .0 2 0 0 0 .0 2 2 2 0 .0 2 1 1 0 .0 1 8 9 0 .0 1 8 9 0 .0 2 2 2 0 .0 1 8 9 0 .0 2 1 1 0 .0 2 0 0 0 .0 1 8 9 0 .0 2 2 2 0 .0 2 0 0 0 .0 2 2 2 0 .0 2 1 1 0 .0 2 0 0 0 .0 1 8 9 0 .0 2 0 0 0 .0 2 1 1 V itam eter Rcale reading 0 .7 1 3 0 .6 9 0 0 .6 6 3 0 .7 0 1 0 .6 3 4 0 .6 9 7 0 .6 7 0 0 .6 9 1 0 .6 4 8 0 .6 9 9 0 .5 9 0 0 .6 9 3 0 .6 3 9 0 .5 9 1 0 .6 8 1 0 .6 0 0 0 .6 8 9 0 .6 5 9 0 .6 4 1 0 .6 0 0 0 .6 7 0 0 .6 7 0 E 1\%cm . Sum m ary (22): M ean M axim um range, % M axim um d eviation. % S u m m ary (44): M ean M axim um range, % M axim um deviation, % 3 2 .1 3 2 .7 3 5 .1 3 3 .2 3 1 .7 3 1 .4 3 1 .8 3 1 .3 3 4 .4 3 1 .4 3 1 .2 3 2 .8 3 1 .9 3 1 .3 3 0 .7 3 0 .0 3 1 .0 3 1 .2 3 2 .0 3 1 .7 3 3 .5 3 1 .7 V itam eter scale reading 0 .7 5 6 0 .7 0 1 0 .6 7 2 0 .7 1 3 0 .6 6 1 0 .6 7 7 0 .6 3 9 0 .6 4 0 0 .6 5 9 0 .6 9 2 0 .6 4 2 0 .6 7 0 0 .6 3 2 0 .6 4 3 0 .6 7 7 0 .6 4 1 0 .6 9 0 0 .6 1 4 0 .6 4 8 0 .6 1 0 0 .6 3 4 0 .6 8 6 1 cm . 3 4 .1 3 3 .2 3 5 .5 3 3 .8 3 3 .1 3 0 .5 3 0 .3 3 3 .9 3 4 .8 3 1 .2 3 4 .0 3 1 .7 3 1 .6 3 4 .0 3 0 .5 3 2 .0 3 1 .1 2 9 .1 3 2 .4 3 2 .2 3 1 .7 3 2 .4 In 3 2 .4 0 1 9 .8 1 0 .2 3 2 .0 0 1 5 .0 8 .8 3 2 .2 0 1 9 .8 1 0 .3 In addition to the stu d y of vitam in A by m eans of th e ultraviolet spectro photom eter, a similar investigation was m ade w ith th e Hilger vitam eter, an instrum ent developed for th e express purpose of determ ining vitam in A in fish liver oils. A ray of ultraviolet light from a copper arc first passes through a filter transm it ting principally copper w ave lengths 3247.55 A. and 3273.97 A., and then through a solution of fish liver oil in isopropyl alcohol. T he ray im pinges on a fluorescent screen as a fine of light ap proxim ately 1 cm. in length. T he in ten sity of this line is compared visually with th e intensity of a similar line from a corresponding ray which does not pass T able Y III. flu en c e of N u m ber o f D e t e r m in a t io n s o n A ccuracy T able V III shows th e accuracy of vitam eter results obtained for six fish oils when th e values from differing num bers of independent runs were averaged to give an value for an oil. T he last column gives the E\ value obtained by a single determ ination on each oil. In parentheses, the percentage differences betw een these single values and th e means of 18 determ inations are given. The other columns contain similar d a ta for th e m eans of larger num bers of determ inations. The m ean of a num ber of runs becomes, in general, more nearly th a t of the m ean of 18 as the num ber increases. This is best illustrated by the means of th e per cent differences (at the bottom of th e table). These values decrease from 5.37, in the case of the single determ inations, to 0.69 in th e case of nine determ inations. T he maxim um per cent differences decrease in the same fashion—from 11.50 to 1.11. T he maximum per cent difference for six determ inations is 1.79. This value, being w ithin experim ental error, establishes this num ber (6) as th e m inim um num ber of tests for routine laboratory practice. M ethod Ado pted. In th e light of observation th e follow ing procedure has been used in routine analysis. of column of Table V, are in very good agreem ent. This be comes even more evident if th e series is divided into smaller groups. B y comparison of the m eans of these successively smaller groups w ith th e m ean of th e whole series, it is possible to ascertain th e num ber of determ inations necessary for a satisfactory representative m ean E \ ^ mn value of the oil. For example, if th e first 20 values (chronologically ob tained) are averaged, th e m ean is 31.44, which is only 0.06 per cent lower th a n th e m ean of all the values. T he m ean of the second half of th e series, 21 values, is 31.49, b u t 0.10 per cent above th e m ean of all. T he m eans of successively smaller groups of values show slight though increasingly larger deviations from the to tal m ean. If six values be taken in each group, the maxi m um difference of any m ean from th a t of the whole series is 1.46 per cent. If th e m ean of th e two values from a single plate be taken, th e largest difference from th e to ta l mean is 3.05 per cent; th e m ean of these per cent differences being 1.33. This dem onstrates th a t a very satisfactory representative value for an oil can be obtained from th e m ean of b u t two spectrophotom etric determ inations. Application of Hilger Vitameter to Fish Liver Oils VOL. 12, N O . 11 V it a m e t e r M e a n s . D uplicate sam ples (0.20 to 0.25 gram) of a fish oil are weighed on the analytical balance into 50-m l. flasks. T o each is added from a buret a volum e in milliliters of isopropyl alcohol equal to 100 tim es the w eight of sam ple in grams. T his makes a Influen ce of of N um ber of D eterm V it a m e t e r M e a n s in a t io n s on A ccuracy --------------------------------- N u m b er of D e te rm in a tio n s-----------------------18 9 6 3 2 1 T y p e of F ish L iver Oil /------------------------------- — H a lib u t (3898) 3 1 .9 6 M ixture (25,679) 2 3 .5 0 H a lib u t (25,379) 3 0 .7 1 H a lib u t (25,349) 3 3 .1 4 D istilla te (25,709) 1 2 0 .4 T u n a concen trate (25,819) 1 4 6 .4 A verage of per cen t differ ences M axim um per cen t differ ence 3 2 .6 3 ( + 0 .6 7 ) 2 3 .3 4 (- 0 .6 8 ) 3 1 .0 5 ( + 1 .1 1 ) 3 3 .1 5 (+ 0 .0 3 ) 1 2 1 .4 ( + 0 .8 3 ) 1 4 7 .6 ( + 0 .8 2 ) M ean 3 2 .7 0 (+ 0 .7 4 ) 2 3 .3 8 ( - 0 .5 1 ) 3 1 .1 0 (+ 1 .2 7 ) 3 2 .5 8 ( - 1 .7 9 ) 1 1 9 .2 ( - 1 .0 0 ) 1 4 7 .7 (+ 0 .8 9 ) V a lu e s ----------------------------3 3 .3 0 ( + 1 .3 4 ) 2 3 .9 3 (+ 1 .8 3 ) 2 9 .2 3 (- 4 .8 2 ) 3 3 .0 7 (- 0 .2 1 ) 1 2 7 .0 (+ 5 .3 1 ) 1 4 7 .3 (+ 0 .6 2 ) 3 2 .4 0 ( + 0 .4 4 ) 2 5 .4 0 (+ 8 .1 0 ) 2 9 .2 5 (- 4 .7 6 ) 3 2 .1 0 (- 3 .1 4 ) 1 2 3 .5 ( + 2 .5 8 ) 1 4 3 .5 (- 1 .9 8 ) 3 2 .1 ( + 0 .1 4 ) 2 6 .2 ( + 1 1 .5 0 ) 2 8 .8 (- 6 .2 3 ) 3 2 .3 (- 2 .5 4 ) 1 3 1 .0 (+ 8 .8 1 ) 1 4 2 .0 ( - 3 .0 1 ) 0 .6 9 1 .0 3 2 .3 5 3 .5 0 5 .3 7 1 .1 1 1 .7 9 6 .3 1 8 .1 0 1 1 .5 0 N O V E M B E R 15, 1940 A N A L Y T IC A L E D IT IO N w eigh t-volu m e concentration of 1.00 per cent. Further prefl t 18 ar+81 n?^'de w ith calibrated pipets and sm all volu m etric flasks u n til th e concentration necessary to give a v ita m eter scale reading extin ction betw een 0.50 and 0 75 has been established. T h ree su itab le d ilution s are th en prepared for each sam ple. I he m ean of ten indep en d en t scale readings is taken as th e v i tam eter extin ction valu e for each dilution. Six E\%m values are calcu lated b y divid ing th e vitam eter extin ction valu es by the per cen t concen tration s of th e corresponding solutions. T he m ean o f all six v alu es th u s obtained g iv es a satisfactory d eterm in ation of th e vita m in A con ten t of th e oil. T h e nonsaponifiable m atter w as prepared according to the pro cedure g iven by M orton (4). T able X. 643 C o m p a r a t iv e E\ V alues of F is h L iv e r O il s a t 328 uifj. ( D e t e r m in e d b y t h e s p e c t r o p h o t o m e t e r a n d v it a m e te r ) T?\ % 1 c m . wt 3 2 8 R a t io , S p e c tr o  V /S p h o to m e te r V it a m e t e r F is h O il N o . C o d L iv e r O il* 3 0 ,8 3 9 1 0 ,3 5 8 1 0 ,3 5 8 “ n o n s a p .†1 3 ,4 4 8 3 6 ,4 7 0 “ n o n s a p .†1 3 ,2 4 8 0 .8 8 4 1 .3 4 1 .2 8 1 .4 1 1 .4 6 1 .4 0 0 .9 0 9 1 .2 9 1 .2 5 1 .4 1 1 .4 0 1 .4 4 0 .9 7 3 1 .0 3 9 1 .0 2 4 1 .0 0 0 1 .0 4 3 0 .9 7 2 H a lib u t L iv e r O ils Comparison of Methods C o m p a r is o n o f S pectro ph o to m eter V a n d it a m e t e r T h u s fa r th e w ork has d ealt extensively w ith fu n d a m e n ta l factors in th e determ ination of vitam in A b y m eans of th e sp ectrophotom eter an d th e vitam eter. T he degree of precision an d reproducibility of b o th m ethods has been d em o n strated in th e series of 41 an d 44 E values, respectively, w ith spectrophotom eter an d v itam eter, on h a lib u t liver oil 3898 (Tables V and V II). B ecause of th e space required, it is impossible to present th e d etailed d a ta for each oil stu d ied in th e several series. In T ab le IX , how ever, th e su m m ary d a ta for b o th m ethods are p resented. T h is includes th e nu m b er of tests, th e m axim um range, th e m axim um deviation, th e m ean value for each sam ple, a n d in th e la st colum n th e ratio of th e v itam eter to sp ectro p h o to m eter m ean. I t is evid en t, from th e m axim um range an d m axim um de viatio n d a ta , th a t th e v ita m e te r is m uch less precise in its op e ra tio n th a n is th e spectrophotom eter. Including th e values for th e cod liver oil an d its unsaponifiable fraction, th e aver age m axim um range for th e v itam eter is a b o u t tw o and one h alf tim es t h a t of th e spectrophotom eter. D espite these wider ranges, th e m ean values b y th e tw o in strum ents on these long series are in good agreem ent, showing th a t a suffi cient n u m b er of values hav e been obtained to approxim ate a n orm al d istrib u tio n . T h e tw o in stru m en ts m u st be meas uring, therefore, th e sam e significant characteristic of the oils. V a l u e s . IX . S V um m ary alues of F is h L iv e r O il s at 328 mju T y p e o f F i s h L iv e r O il N um ber of T ests M a x i m um R ange, % 7 .0 1 9 .8 M a x i m um D e v ia  t io n , % 3 .5 1 0 .3 E 1% 1 cm . M ean R a t io , v a lu e s V /S a 3 1 .4 6 3 2 .2 0 1 .0 2 3 2 .8 1 2 .0 5 0 -1 2 6 0 .3 9 1 .0 0 5 8 .9 1 7 .8 5 .1 9 .0 7 8 .6 8 7 9 .8 2 1 .0 1 4 42 40 7 .0 2 1 .0 4 .2 1 1 .0 1 2 8 .6 1 3 2 .6 1 .0 3 1 OS) (F ) 17 30 4 .0 1 8 .5 2 .3 1 0 .6 1 5 .0 5 1 5 .1 2 1 .0 0 5 M ix e d (1 6 ,9 4 9 ) OS) (F ) 17 30 4 .0 2 1 .2 2 .0 1 3 .8 3 9 .8 5 4 0 .0 7 1 .0 0 5 C o d (9 7 5 8 ) (5 ) 77 ( F ) 110 1 0 .0 1 8 .0 5 .0 1 0 .4 1 .5 8 9 1 .4 4 4 0 .9 0 9 C o d “ n o n s a p .†(9 7 5 8 ) (5 ) (F ) 22 62 1 1 .2 1 6 .9 6 .7 8 .8 1 .3 4 6 1 .3 6 9 1 .0 1 7 7 .1 1 9 .4 3 .9 1 0 .7 H a l ib u t (3 8 9 8 ) OS)-* 41 44 (F ) H a l ib u t ( 1 6 ,5 1 9 ) OS) (F ) 41 40 5 .0 2 2 .2 M ix e d (4 2 8 8 ) OS) (F ) 46 46 M ix e d ( 1 6 ,3 1 9 ) OS) (F ) H a l ib u t (1 8 ,8 6 9 ) A v . (8 ) 1 S , S p e c t r o p h o t o m e t r ic (5 ) (F ) v a lu e s . V , V it a m e t e r v a lu e s . 1 .6 5 7 .7 0 2 2 .6 2 3 .1 2 9 .3 3 5 .4 5 1 .0 6 9 .0 1 .0 2 2 1 .0 4 1 0 -9 7 6 1 .0 2 4 1 .0 2 4 1 .0 0 0 0 .9 8 4 1 .0 0 0 1 .0 3 1 0 .9 7 9 0 .9 9 5 1 .0 3 0 0 .9 7 1 1 .0 0 0 1 .0 0 4 1 .0 1 3 1 .0 3 5 0 .9 8 8 0 .9 8 5 0 .9 9 6 1 .0 3 1 1 .0 0 7 1 .0 3 4 1 .0 5 1 1 .0 2 7 1 .0 0 3 1 .0 5 0 0 .9 9 4 1 .0 4 6 1 .7 4 7 .7 8 2 3 .1 2 4 .2 2 7 .1 3 3 .9 5 0 .8 6 7 .8 1 .0 5 5 1 .0 1 0 1 .0 2 2 1 .0 4 8 0 .9 2 5 0 .9 5 8 0 .9 9 6 0 .9 8 3 3 .6 0 2 5 .0 2 7 .6 1 3 3 .4 1 .0 0 9 0 .9 6 5 0 .9 8 2 1 .0 1 4 H ic k m a n D is t ill a t e s 3 .5 7 2 5 .9 2 8 .1 1 3 1 .5 8 ,0 0 8 8 ,5 4 8 8 ,5 5 8 8 ,0 4 8 C o m p a r is o n able 1 3 .7 1 5 .2 1 6 .2 1 7 .1 1 7 .2 1 7 .3 1 8 .5 1 8 .8 1 9 .6 1 8 .9 1 9 .4 2 0 .9 2 0 .0 2 1 .4 2 2 .0 2 2 .7 2 3 .9 2 5 .2 2 5 .4 2 5 .8 2 9 .7 2 9 .4 3 0 .3 3 2 .6 3 2 .2 3 1 .7 3 3 .4 3 2 .3 3 8 .5 F is h L iv e r O il M ix tu r e s 1 7 ,7 8 9 1 7 ,7 7 9 1 3 ,1 0 8 1 1 ,5 9 8 3 1 ,5 9 9 1 3 ,4 3 8 1 4 ,7 4 8 1 7 ,8 4 9 o f R o u tin e S p e ctro p h o to m eter an d In T able X are included spec trophotom etric and vitam eter E\% m_ values determ ined in routine laboratory practice. These d a ta dem onstrate th a t reliable E\% m_values can be obtained by m aking a m inim um num ber of runs on each instrum ent. T he corresponding values are in close agreem ent throughout, as is shown by th e ratio values in th e column a t th e right. This again em pha sizes th e fact th a t th e two instrum ents are capable of m easur ing th e sam e characteristic of th e fish liver oils, irrespective of their potencies or type. C o n v e r s io n F a c t o r . All seven oils studied in the long series on b oth spectrophotom eter and v itam eter were sub jected to bioassay according to th e U. S. P . procedure (9). T able X I gives these data as U. S. P. units per gram , and also the E\%m values a t 328 mja for b oth instrum ents. F rom these values one can calculate the so-called conversion factors for th e different oils. In th e m ain these d a ta show a rem arkable agreem ent. One oil (No. 16,519) is 16 per cent lower th an the m ean from the spectrophotom eter d a ta and 15 per cent lower th a n th e m ean from v itam eter d ata. T aking oils from the spectrophotom eter, th e m ean conversion factor is 2152, b u t 3 per cent different from th e factor calculated from the value of the unsaponifiable fraction of the U. S. P . reference V ita m e te r E T 1 3 .4 1 4 .6 1 6 .6 1 6 .7 1 6 .8 1 7 .3 1 8 .8 1 8 .8 1 9 .0 1 9 .3 1 9 .5 2 0 .3 2 0 .6 2 1 .4 2 1 .9 2 2 .4 2 3 .1 2 5 .5 2 5 .8 2 5 .9 2 8 .8 2 9 .2 2 9 .3 3 1 .0 3 1 .3 3 1 .6 3 1 .8 3 2 .5 3 6 .8 3 2 ,5 3 9 3 2 ,5 4 9 3 2 ,4 6 9 3 2 ,5 0 9 3 2 ,4 1 9 3 2 ,4 8 9 3 2 ,4 9 9 3 2 ,5 1 9 3 2 ,4 7 9 3 2 ,3 6 9 3 2 ,4 5 9 3 2 ,3 7 9 3 2 ,5 2 9 3 2 ,3 8 9 3 2 ,4 3 9 3 2 ,4 0 9 3 2 ,3 9 9 3 2 ,5 8 9 3 2 ,4 4 9 3 2 ,4 2 9 3 2 ,5 5 9 3 2 ,5 7 9 3 2 ,5 9 9 2 ,6 6 8 3 ,2 5 8 3 ,6 5 8 2 ,9 7 8 3 2 ,5 6 9 1 5 ,7 6 8 V a lu e s . 644 IN D U S T R IA L A N D E N G IN E E R IN G C H E M IS T R Y T a b le X I. C o r r e l a t i o n o f E \% m a t 328 w ith B io lo g ic a l P o t e n c ie s o f F is h L iv e r O ils T y p e of F ish Liver Oil B io a ssa y va lu e C alcu lated C onversion I cm . V a lu e a t 3 2 8 m /i F actor S p ectro Sp ectro p h o to m eter V itam eter p h otom eter V itam eter U. S. P . u nits/g H a lib u t (3898) H a lib u t (16,519) M ixed (4288) M ixed (16,319) H aU but (18,869) M ixed (16,949) C od “ nonaap." (9758) 6 5,000 9 0,500 185,300 295,100 31,900 8 7,800 3,000 3 1 .4 6 5 0 .1 2 7 8 .6 8 1 2 8 .6 0 1 5 .0 5 3 9 .8 5 1 .3 5 3 2 .2 0 5 0 .3 9 7 9 .8 5 1 3 2 .6 0 1 5 .1 2 4 0 .0 7 1 .3 7 2066 1806 2355 2295 2120 2203 2222 2019 1796 2321 2225 2110 2191 2190 M ean valu e 2152 2122 tube is em ployed, its constancy is definite w ith re spect to the optical axis, and one is able to obtain satisfactory results in regard to reproducibility and applicability to routine assays. For visual comparison, optim um results are ob tained if the photographic emulsion has been proc essed to a d en sity of 0.6. A few substances m a y interfere w ith th e deter m ination of vitam in A in fish fiver oils. Among these th e m ost im portant are: (ct) the unsaturated long-ehain fa tty acids (particularly in low potency oils) and (b) the m etallic salts such as copper oleate and ferric oleate, both of w hich can be detected ana lytically if present in am ounts having significant absorption. B y observing tliese essential factors, a long series of tests was m ade on each, of seven fish liver oils having a wide range in vitam in A potency. Analysis of the spectrophotom etric d a ta shows th at the m ean of as few as two determ inations gave a satisfactory result ( :i=2 per cent). In the study w ith th e vitam eter, reproducibility was de term ined by m aking a series of tests on th e sam e fish liver oils m entioned in connection w ith the spectrophotom eter. Analysis of the d ata established th a t a satisfactory E value can be obtained from a m inim um num ber of six tests. T he vitam  eter mean values on the long series were in close agree m ent with those of the spectrophotom eter. B y correlating the spectrophotom eter and v itam eter data w ith those from the biological procedure, conversion factors have been calculated. The several values are in good agree m ent, the m ean value being, in this instance, 2137. This is, therefore, a close approxim ation betw een th e potency of a fish liver oil in U. S. P. units per gram and th e spectrophoto m etric value. oil, 2222. For th e vitam eter, the m ean conversion factor is 2122, also b u t 3 per cent different from th a t derived from the unsaponifiable fraction of the reference liver oil, 2190. The average of th e m ean conversion factors in Table X I, 2137, is in close agreem ent w ith th a t obtained from routine bioassay and spectrophotom etric results on fish liver oils during the p ast several years. I t is also very close to B arthen’s factor 2064 referred to by Wilkie {10). I t is th u s apparent th a t it is possible, under the conditions of technique as carried o u t (1) to obtain a quan titative evalua tion of vitam in A by measuring th e extinction coefficient a t 328 mju; and (2) b y applying a conversion factor such as 2137, to express the potency in U. S. P. or In ternational units per gram. Discussion and Summary In this critical stu d y of th e applicability of th e spectro photom etric m ethod to the determ ination of vitam in A in fish liver oils, th e characteristics of the operative procedures have been carefully investigated w ith respect to the Bausch & Lomb spectrophotom eter and th e more specialized instru m ent, the Hilger vitam eter. By m eans of an elaborate series of determ inations on several oils, reproducibility has been dem onstrated w ith both, and a comparison of the results shows good agreement. T he investigation has established the following funda m entals w ith respect to the spectrophotom etric technique: Conclusion The E x^m. value can be accurately determ ined spectrophotom etrically under a carefully controlled procedure. It is therefore possible, by em ploying the proper conversion factor, to evaluate satisfactorily the vitam in A potency of fish liver oils in units per gram. Since vitam in A in solution m ay lose considerable of its ab sorptive power after a tim e, the test reading should preferably be m ade w ithin an hour after preparation of the solution. The vitam in is equally absorptive in isopropyl and absolute ethyl alcohols. T he cells should be carefully paired and their surfaces kept scrupulously clean. Especial care should be taken th at no oil residues remain on the cell surfaces. The values of a fish liver oil determ ined through a range of 50 tim es difference in concentration were found to be in agreement, showing that Beer’s law holds. T his is also true when one concentration is examined in cells of different length, showing th a t Lam bert’s law holds within the ex perim ental error of the m ethod. If a condensed spark is used as source of fight, its position with respect to the optical axis of the system m ust be m aintained with the highest degree of precision. T his is ordinarily very difficult to accom plish since erosion of th e electrodes is unavoidable during the course of sparking. On th e other hand, if a Hilger hydrogen P h in t e d VOL. 12, NO. 11 Literature Cited (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Carr, F . H . , and Price, E . A., Biochem. J ., 20, 497 (1926). Drum m ond, J. C., and Coward, K . H ., Ibid., 14, 734 (1020). Drum m ond, J. C., and M orton, R . A ., Ibid., 23, 785 (1929). M orton, R. A . , “Absorption Spectra of V itam ins and H or m ones†, London, Adam H ilger, 1935. M orton, R . A., and H eilbron, I. M ., Biochem. J ., 22, 987 (1928). M unsell, H . E ., “The V itam ins, A Sym posium †, Chap. IV, pp. 87-109, Chicago, Am erican M edical A ssociation, 1939. Rosenheim , O tto, and Drum m ond, J. C-, Biochem . J ., 19, 753 (1925). Takahashi, K ., N akam iya, Z., K awakam i, K ., and K itasato, T., In st. P hys. Chem. Acs. (T okyo), Sci. P avers, 3, 81 (1925). U . S. Pharmacopoeia X I (Second Supplem ent), p. 134, 1939. W ilkie, J. B ., J . Assoc. Official A gr. Chem., 23, 336 (1940). P r e s e n t e d b efo re th e D iv is io n of B io lo g ic a l C h e m is t r y a t t h e 9 8 t h M e e t  in g of t h e A m e r ic a n C h e m ic a l S o c ie t y , B o s t o n , M a s s . in U. S. A.