THE EFFECT OF ULTRAVIOLET IRRADIATION ON THE PHOTOCHEMICAL CHANGES OF ERGOSTHROL AND THE CHROMATOGRAPHIC SEPARATION OF THE PRODUCTS By James F . r.irn A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 19?3 THE EFFECT CV ULTRATZGUT IRRADIATIQM OH TH£ rHOrOCHJMZCAL CHARGES Or EROOflTBKM. AMD THE CHROMATOGRAPHIC SEPARATION Or THE PRODUCTS »y J « h w F. Kira AM ABSTRACT 3ufa«ltt«d to tho Sobool of Qraduat* StodlM of Michigan Stott Collogo of Aplevltart and Applied Soloaoo In partial fulfil inwit of tbo ro^iilrwonta for tbt dogma of DOCTOR or rHZLOSOTHI Dapartaont of Chaolotry loar Approrod 1953 Jmmmm F . K ir a TH3SX3 ABaraiCT Tbt coawaraion of «th«r aolatlmM of «rfofWrol late vitaala D» and otbar and pvodnotc « u (tadlod twinf oltraviolat radiation of tho waralongth rtgloaa 230 m , to 261* m . # 28U m . to tho vl«lbl«f and tho ontlro oaorgy dlatrlbotloo of a " V v t m 11 liogi . Othor otnrtlaa wora oarrlod oot In tho proaoaao of aonoehroaatlo radiation of wamloagtha 25U m . t 260 n . f 265 m . i 275 ■».« 281 aa . f 285 m . f and 296 no. Kadh raaotlon vac oarrlod oat In on airtight gnarta e d l and tho progroaa of tho roaotlon dotoralaod by tho ultrarlolot abcorpiion apootra. To aid la tho Intorpr at atlon of tho roaultlag apootra with rogard to oolotlon cnopodilon, no rural norloo of thoorotloal aboorptlon ourvoa warn oo Iom Iatort fron lltoraturo apootra graphic data, for oonblwatlona of tho latoraodlato produoto a n d orgoatorol. Fran thoao ourroa a non aorloa of data uaa oaloalatod banod on tfaa ratio of tho oxtinotioo raluoa at a glran oaralongth to tho axtlnetlon viltn of tho ourra eonaldorad at 281 an. Thla aaluo waa plottod agalnat lnoroaaod oonoontratlona of oaa of tho onqpoaonta at a glran wnwoloagth. Tho alopo of tho roaoltlag owrraa aro oharaotorlatlo of build up troada during Irradiation. Fran thoaa data it la pooalblo to oatlaato tho orayioaitlon of tfaa Irrodlatod aolutlona. Tho oocpoaltlon of tho IrradLlatod aolutlona aro dlaouaaod with roapoot to wavoloagth of tho activating onargy. Tho aonaiiivity of othar and othaaol aolutlona of oalolforol In ultra** rlolot radiation waa atudlod at unvolongtha 25b nu., 265 ana., 281 an.. •ad 196 ant. Tbs d— tnwiion aaa ftUowd through t)a nltrtvlolvt abeorpUoa tddblttd by these solutions at >6$ m . It vm foaad that in atlmmt the oaloiferol aaa mere rapidly destroyed than aaa oaloifaral la ethanol and la ftntral the rata of breakdown aaa greater with tfaa abortar wavelength radiation. Tba separation of pore oalolfarol by obroaatogrephlo prooadnraa aaa studied using two types of adsorbents, a dlstnwaosiwis earth, "Superfilterol"j and alumina. Tba solutions separated consisted of Irradiated ergo sterol abloh bad been Irradiated under varying conditions of sail distensions, wavelength of activating energy, and the atmosphere of the covering gases to determine bow these conditions, through alteration of tba composition of the irradiated ergssterol solutions, w>uld offoot the separation. The call dimensions were varied froa 10 sst. to 0.2? an., the radiation oonsistad of either the oaatplete spoctrvm of a "Uviaro" naraury lanp or tbs radiation from this 1.wup filtered by a p * ^ l a m solution, and tba gas atmospheres over tho solutions during irradiation waa either oarbon dioxide, or air. Kaoh irradiated ergoaterol solution aaa passed over either a "Superfilterol" oolusn of activated alualna and the separation followed by the ultraviolet absorption ourvee of the eloate fractions. In general, waa found that tba "Superfilterol" divided the irradia­ tion product into two major bands, the aajor constituent of one being unchanged ergoaterol and the other calciferol, tachgrsterol, and a compound having absorption maxima at 272 au. and at 260 mu. Variations in relative quantities of tba components were observed depending on the Jaaaa F. Kirn conditions of Irradiation. Ali— lno M h l m d nova oonploto upcrttltn of tho oaldforol and tho ooapound etaaraetorlood toy pooks at 272 m * and 280 no., than «&• oboonrod for "Soporflltorol .• ACKNOWLEDGMENT The author wishes to express his sincere appreciation to Professor D. T. Ewing for his guidance and counsel during the course of this investigation. •WUHW* * TABLE OF CONTENTS PAGE I. II . ThE EFFECT OF ULTRAVIOLET IRRADIATION ON THr- PHOTOCHEKiCAL CHANGE OF ERGOSTEROL........................................ 1 Experimental................................................ 3 Procedure.................................................... 5 Results...................................................... 7 Discussion................................................... 13 General Discussion.......................................... 2h Summary...................................................... 26 Literature Cited............................................ 28 THU CHROMATOGRAPHIC SEPARATION OF THc, IRRADIATION PRODUCTS OF ER GOST'EROL................................................... 29 Procedure.................................................... 33 Discussion................................................... 37 General Discussion.......................................... ii6 Summary................................ „.......... 52 Literature Cited............................................ 53 1 I. THE EFFECT OF ULTRAVIOLET IRRADIATION ON THE PHOTOCHEMICAL CHANGE OF ERGOSTEROL The effect, of ultraviolet irradiations as a function of wave lengths of light on the activation of vitamin D has drawn the attention of many researchers. Prior to the discovery of the provitamins of vitamins D most of the work was carried out Dy the direct irradiation of animals. however, after the conclusive proof of O. Rosenheim (9) showing ergosterol to De a parent compound of vitamin D, much of the work has been carried out using pure ergosterol. The early literature is voluminous with descriptions of experiments of both qualitative and quantitative nature proving the attributes of one wave length of light to that of another. It is generally concluded, how­ ever , that the greatest yield of vitamin D 3 is obtained using radiation of 2750 A. to 3131 A. (10). Windaus and co—workers (11,15) in their classical works have de­ termined the accepted mechanism of activation to De TOXISTEROL ERGOSTEROL LUmj.STEROL PROTACHYSTEROL TACHYoTEROL CALCIFEROL SUPRASTEROL I SUFRASTEROL II and have isolated and proven the structure of the intermediate compounds which result during the activation with ultraviolet radiation. Windaus, K. Ditiunar ano Fernholz (16), have recommended wave lengths from 290 mu. to 300 mu. as the most advantageous range for the formation of lumisterol with short periods of irradiation. Short wave lengths of ultraviolet 2 radiation of 265 mu. or less are more advantageous for the formation of tachysterol (16). The final products of the photochemical reaction, suprasterol I and suprasterol II, do not indicate preference of wave length for their formation (17). The existence of toxisterol, the other end product, has been questioned (5). L. Velluz and G. Amiard (12) and G. Amiara and Petit (13) have isolated and characterized a precalciferol, while j. Green (5) has proposed an absorption spectrum for a suprasterol III. The effect of different solvents on the activation of ergosterol in solution has ueen investigated by Bills, Honeywell, and Cox (2) who ob­ tained absorption curves of the same general shape when the irradiation was carried out in ethanol, cyclo hexane, and diethyl ether. activation in the different media is said to be different. The rate of They have shown that ether solutions have a cod liver oil coefficient of 710,000 while cyclouexane and alcohol solutions exhioit coefficients of 330,000 and 250,000 respectively. The effects of temperature of irradiation has been studied by Webster an.; oourdill on (lh) at 72c°, 30.6°, 1°, -16° and approximately -163° and -195 C. They found, with the exception of the extremely low temperatures, that very little change in the activity' of the resulting proauct could be detected. This investigation was carried out to study- the changes wliich take place in the absorption spectrum of diethyl ether solutions of ergosterol under the influence of various wave lengths of ultraviolet radiation ana to interpret trie resulting absorption curves wi Ui respect to the build-up and break-down of the irradiation products and intermediates under the conoitions described herein. 3 EXPERIMENTAL Apparatus and Materials Ergosterol — A very pure grade of ergosterol (Montrose) was used. The ergosterol was recrystallized from an alcohol mixture containing 902 volumes etlianol, hi volumes methanol, and U5 volumes water. The re­ crystallized ergosterol was dried under vacuum and stored in vacuum at below freezing temperatures . Calciferol — The calciferol was synthetic crystalline vitamin D 2 obtained from the Winthrop Chemical Company exhibiting an E (1$, 1 cm) of U60 at 265 mu. Diethyl Ether — Anlydrous ether, C. P., was stored over sodium chips and distilled from ferrous sulfate prior to each experiment to re­ move both moisture and peroxides. The resulting ether was spectrographic­ ally tested for transparency in the lower ultraviolet wave lengths, down to 2260 A°. Skelly Solvent — Commercial Skelly Solvent MBM was redistilled. The fraction boiling at 8U - 65° C was collected. This fraction was then passed chromatographically over silica gel, activated at 25h° C . The resulting solvent will transmit 97% of the incident light at 228 mu. Ultraviolet Source — The light source used for the ir’-ddiation was a Uviarc lamp of Cooper Hewitt manufacture, operating at 2iiO volts D C ., 12 amperes. Spectrometer — All absorption spectra were determined with a Beckman Quartz Model DU, spectrophotometer. u Light Filters — Para-xylene filters consisting of a 2 cm. thick quartz cell containing a $% solution, (volume/volume) , of redistilled para-xylene in purified Skelly Solvent HBn (8) were used. These filters transmitted h5.h% at 288 mu. The bromine-chlorine filter consisted of a 2 c m . thick quartz cell filled to a depth of 3 cm. with liquid bromine into which chlorine was bubbled at a constant rate (l4.,7), transmitting wave lengths from 210 rau. to 2 80 m u . Monochromatic Light — The monochromatic light for irradiation was obtained using a Bausch and Lomb Quartz Monochromator in connection with the bviarc lamp and a cylindo-plano quartz lens for illumination of the slit. 5 PROCEDURE Diethyl ether solutions of ergosterol (0,0025 g/lOO ml.) or calciferol (0.0015 g/LOO ml.) were exposed to the monochromatic radiation at the exit slit of the monochromator in a sealed quartz cell (0.5 cm thick) for a pre­ determined time. The irradiation cells were reversed at five minute intervals to in­ sure uniform irradiation and were enclosed in aluminum foil to protect the solutions from stray radiation and to reflect the desired radiation tlirough the solution. Representative samples were taken for the determin­ ation of the absorption spectrum. The studies involving filtered and unfiltered radiation were carried out in a manner similar to that described above. The monocliromator was removed and where applicable replaced by the previously described filters. The absorption spectra of the solutions were determined in the irradiation cells. The ergosterol solutions were exposed to various wave lengths of ultraviolet radiation in quartz cells for increased lengtlis of time to determine the effect of exposure time and wave length on the absorption characteristics of the ergosterol solutions. Ultraviolet light of a mono­ chromatic nature was used with the wave lengths selected to give the most intense bands of the mercury spectrum between 230 mu. and 300 mu. Filters were selected roughly to divide the ultraviolet radiation into tws bands: the short wave length band, 210 mu. to 280 mu., by the use of a brominechlorine filter, and the long wave length band by the use of a para-xylene filter, 280 mu. to visible region. The ether solutions of calciferol were exposed to various wave lengths of ultraviolet radiation in quartz cells and for various time intervals to determine the effect of wave length and time on the photo­ chemical change. 7 RESULTS The results will be described in detail as to the conditions of irradiation of the solutions. A description of the change in the ab­ sorption spectra will be given for intervals of the irradiation period. Plate I shows the experimentally determined results of the irradi­ ation of an ether solution of ergosterol by the unfiltered Cooper Hewitt Uviarc lamp. It will be noted that during the first seconds of irradiation, curve ho. 2, there is a general increase in the absorption of the ergosterol solution at all wave lengths except 281 mu. The absorption curve indicates large increases in the absorption in the lower absorption wave lengths; i.e., 230 mu. extending through the maximum at 271 mu. The minimum at 276 mu. increases with respect to the maximum at 281 mu. In the longer absorption wave length region, the minimum at 290 mu. increases with respect to the maximum at 293 mu. however, this wave length region exhibits an over-all increase in absorp­ tion with respect to the maximum at 261 mu. As the time of irradiation increases, the absorption at 230 mu. con­ tinues to increase until a new minimum appears at 2U0; the maximum at 263 mu. disappears with the formation of a small plateau at 263 mu. The maximum at 271 mu. shifts to 270 mu. with the minimum at 276 mu. decreas­ ing in depth with respect to 270 mu., shifting to form a new minimum at 273 mu. The maximum at 261 mu. shifts to 260 mu. with greater absorption at this wave length with respect to the maximum at 270 mu. than that 8 manifested by pure ergosterol. The maximum at 293 mu. shifts to 290 mu. with the elimination of the minimum at 290 mu. of ergosterol. After the over-all maximum absorption of the irradiated solution is reached (curve No. 3), the absorption curves tend to decrease- in the longer wave lengtiis, i.e., from 260 mu. and longer, wliile the relative shape of the curve is maintained, the extinction decreases resulting in the form­ ation of a maximum at 263 mu. Continued irradiation under these condi­ tions results finally in the destruction of all absorption by the solution. Plate II shows the results of irradiation of an ergosterol solution by the Uviarc lamp filtered by para-xylene. It will be noted that during the early minutes of irradiation the over-all extinction of the absorption curves decrease at all wave lengths. however, as the time of irradiation increases, the extinction at 230 mu. increases as does the maxima at 263 mu. and 271 mu. The maximum at 281 mu. after the initial decrease remains nearly fixed as does the extinc­ tion of the longer absorption wave lengtiis, i.e., from 281 mu. to 300 mu. The extinction value at 230 mu. increases with increase time until a minimum at 2U0 mu. results. The maximum at 263 mu. is maintained through­ out the entire irradiation period, however, this maximum increases with respect to the maximum at 271 mu. out shift in wave length. The maximum at 271 mu. increased with­ The minimum at 273 mu. shows an increase of aosorption with respect to the maximum at 281 mu. until the minimum disappears. There is a small change in the maximum at 293 mu. Thus, it can be seen that the absorption is increased in the lower wave length regions to wave length 281 mu. and the absorption at wave lengths greater than 281 mu. remains nearly constant. Plate III shows the experimentally determined results of the irradi­ ation of an ergosterol solution using a Uviarc lamp with a brominechlorine filter. In the early minutes of irradiation, the absorption spectrum of the ergosterol solution shows a definite increase in over-all absorption from 230 mu. to 300 mu. At 230 mu. the absorption increases at a very rapid rate with an increase of irradiation time. At wave length 263 mu., the maximum increases in the absorption in the early minutes, but as the time of irradiation increases, t U s maximum disappears and can no longer be identified. The maximum at 271 mu. increases with a shift towards the shorter wave lengths resulting in a maximum at 270 mu. The minimum at 276 mu. increases with the over-all increase in absorption of the entire spectrum until the minimum is no longer identifiable. The absorption at the minimum 290 mu. increases rapidly during irradiation to form a new maximum coupled with the disappearance of the maximum at 293 mu. With increased time of irradiation, the absorption of the irradiated solutions reaches a maximum value at curve w o . 2. Any further irradiation results in an over-all decrease in absorption in the range 260 mu. to 300 :nu., but with increased absorption in the shorter wave lengths. Plate IV shows the results of the irradiation of an ergosterol solution by a Uviarc lamp using the wave length 2Sh mu. as obtained from the monocliromator. 10 The ergosterol solution on exposure to radiation of this wave length exhibits an over-all increase in the absorption over the entire absorption range with increase time of irradiation. The general shape of the curve is maintained except for shifts of the maxima 2o3 mu., 271 mu., 281 mu., and 293 mu. towards the lower wave length end of the spectrum. at 263 mu. shifts to 260 mu., 271 mu. to 270 mu., and The absorption at 263 mu. increases absorption at 271 mu. The maximum 293 mu. to 290mu. at a greater rate than does the The minimum at 276 mu. increases in extinction value and shifts to wave length 273 mu. Plate V shows the absorption curves resulting from the irradiation of ergosterol solutions by a Uviarc lamp using wave length 260 mu. as obtained from the monochromator. There is an over-all increase in the extinction of the absorption spectrum with the following ciianges in the absorption characteristics of the spectrum with increased time of irradiation. The absorption maximum at 263 mu. of ergosterol does not shift to shorter wave lengths, but remains fixed in absorption magnitude with respect to the maxima 271 mu. and 281 mu. from 271 mu. to 270 mu. The maximum at 271 mu. slowly shifts The minimum at 276 mu. slowly shifts to 278 mu. with increased absorption in this region with respect to both maxima 271 mu. and 281 mu. The maximum at 281 mu. shifts to 280 mu. with increased absorption with respect to peak 271 mu. The maximum at 293 mu. shifts towards shorter wave lengths to form a new maximum at 290 mu. in place of the original minimum. 11 Plate VI shows the results of the irradiation of ergosterol solu­ tions by a Uviarc lamp using the wave length 265 mu. as obtained from the monochromator. In the early minutes of irradiation, the absorption curve of ergos­ terol remains relatively stable with respect to the maximum at 281 mu. with increases in the extinction coefficients at all other wave lengths. The maximum at 263 mu. increases with a shift to 260 mu. and with increased absorption with respect to the maximum at 271 mu. The maximum at 271 mu. shifts towards the shorter wave lengths with increased absorption with respect to the two maxima 271 mu. and 281 mu. The minimum at 276 mu. shifts to 273 mu. with increased absorption with respect to the two maxima 271 mu. and 2 81 mu. The maximum at 293 mu. shifts to 290 mu. with increased absorption with respect to the maximum 261 mu. Plate VII shows the results of irradiation of ergosterol solutions by a Uviarc lajmp using wave length 275 mu. The irradiation of ergosterol at this wave length causes no shift in the absorption peaks at 263 mu., 281 mu., or 293 mu. The only variation is in the relative intensity expressed at each peak. From the early minutes of irradiation the peaks at 271 mu. increases rapidly with respect to the increase of absorption of peak 261 mu. A.s the time of irradiation increases, both peaks have the same extinction coefficients. The minimum between these two maxima experiences an increase in absorption, but main­ tains itself at 276 mu. The maximum at 293 mu. shifts towards the shorter wave lengths to form a new maximum at 290 mu. 12 Plate VIII, Plate IX, and Plate X show the results of irradiation of ergosterol solutions at 261 mu., 285 mu., and 296 mu. respectively, as obtained from the monochromator. The absorption spectrum of the irradiated ergosterol shows no alter­ ation of the four absorption peaks with regards to wave length. The changes in the spectrum are manifested by the relative rate of increase in the absorption at the various wave lengths. Absorption peak 263 mu. increases rapidly with respect to the maxi­ mum at 261 mu. wiiile the absorption at peak 271 mu. increases with re­ spect to peak 281 mu. The minimum at 276 mu. remains fixed but experi­ ences an increase in absorption with respect to maxima 271 mu. and 281 mu. It is also found that little change results in the absorption spectrum at wave lengths above 261 mu. wiiile the maximum at 293 mu. remains in its same relative position with respect to the peak at 281 mu. 13 DISCUSSION In order to interpret the changes which take place in irradiated ergosterol solutions, it was necessary first to determine the changes wiiich would take place if only one intermediate resulted during an irradi­ ation. Literature values for the absorption curves of the isomers (3,6) were obtained and from these data, assuming Beer*s law holds true in each case, the absorption curves for various concentrations of the isomers in ergosterol were calculated. Figure 1 shows the absorption curves wiiich result from the calculated curves of ergosterol with increased concentra­ tions of calciferol. Figures 2 , 3 and h are similarly calculated for increased quantities of luraisterol, tachysterol, and toxisterol respectively. The curves for mixtures of suprasterol I and II were not included since the absorption of these compounds does not extend beyond wave length 210 mu. and would manifest themselves only in this region. In order to study the trends of various absorption wave lengths through­ out the photochemical reaction, it was found advantageous to use a ratio of the extinction values of any given wave length of the absorption spectrum to that of the extinction value at 281 mu. of the irradiation mixture being considered. For example: The ratio of the extinction value at 230 mu. to the extinction value at 281 mu. for the curve being considered > (E 23o/^aex)• This ratio will be hereafter referred to as "the ratio" at a given wave length. This method has the advantage of accenting the changes which take place in the absorption spectra at all wave lengths other than the reference 500 400 300 M 200 lOO 250 270 VKave l e n g t h Fife. 1. mixtures - Absorption calculated compounds. (1) pure curves of £90 In mu. ergosterol and calciferol f r o m t h e e x t i n c t i o n v a l u e s o f the p u r e ergosterol, calciferol, 20;t,(h) ergosterol, 40>4;c a l c i f e r o l , calciferol, tfOJi,,(6) (2) ergosterol, pure 60Jt, ergosterol, 80}t; 60jb; calciferol, 40%, (b) ergosterol, 20;*,; calciferol. (4} 500 i.avc lpngtii In au,. r it . : Abi.i'Tftl .r* c u r v e s s .1 ciiatf-d ouritsr.. I; .■ • 1 £••t < 1*. I , . *♦ ) t J I ’ jt. ' S t , ' .. ^ t. i, ; ■, ] 1‘r o u i v ., of ertostorol U / P r t u S L ere 1, ■ 1 - 1 - ' r lurnis t e r c l t-.ii e).tir.ct»or. v o l t e s srt;cstej^., A ; uno t t .:I'e. 1 r vfJ of ei'^^terui, t..e , -re t)u^; 6 o /b ; iu.u i ub ero 1 , -io/u, 0 V i *— 1 i t e r A I. e 1 .1 F *' < ee t. 1 _> i. , ■ ; 500 400 300 •h 8 100 275 £90 Wave length In au. Fig. 3.- Absorption curres of ergosterol and tachysterol mixtures calculated from the extinction values of the pure compounds. (1) pure ergosterol, tachysterol, £0%, (3) ergosterol, 60%; tachysterol, 40%, (4) ergosterol, 40%; tachysterol, 80%. (£) ergosterol, 80%; tachysterol, 60%, (5 ) ergosterol, £0%; 500 400 300 S 200 100 Fig. 4.- Absorption mixtures calculated compounds, toxlsterol, (4) (l) pure 20%, ergosterol, toxlsterol, BO*, 270 :50 W a v e l e n g t h In mu. curves of ergosterol from ergosterol, toxlsterol (2) e r g o s t e r o l , ergosterol, fi0%; 40%; toxlsterol, 6 ( 5 ) pure and tne e x t i n c t i o n v a l u e s o f t n e p u r e (£) (6) 290 toxlsterol. 80%; toxlsterol, 40%, ergosterol, 20%; lli wave length and reducing any experimental error which might be present due to weighing and evaporation of solvents during the irradiation period. For a basis of comparison, the ratio E i/ E 2ax values were calculated from the curves found in Figures 1, 2 , 3 and U for the wave lengths most affected by the presence of the intermediates. Figure 5 is the calculated ratio curves of mixtures of calciferol in ergosterol at various wave lengths. Figures 6, 7 and 8 are similarly calculated for mixtures of ergosterol and lumisterol, tachysterol and toxlsterol respectively. Figure 9 is the calculated ratio curves for mixtures of calciferol and tachysterol. While this device does not offer a quantitative method for the analy­ sis of the irradiation mixture, since the formation of trace quantities of any one product may be overshadowed by a comparatively large change in absorption of the other forming isomers, it still, however, enables one to observe more accurately the general trends of the reaction. When the above observations are applied to the irradiation data of ergosterol in the least complex condition, i.e., with irradiations carried out in the presence of para-xylene filtered light, Figure 19, it is ob­ served that the ratio values increase in the early minutes of irradiation. However, at the end of 16 minutes of irradiation, there is a change of slope in the ratio lines of wave lengths shorter than 27U mu. When the ratio curves are compared with the plots of ergosterol and calciferol, Figure 5, the general trend is the same for short periods of irradiation with the exception of the ratio curves 290 mu., 29U mu., and 300 mu. Furthermore, it is noted that wave lengths 230 mu. and 2hO mu. l.ao x.oo 0.60 -o -o. 12 .k 0.80 O Fig. f..- The 80 10 lOO 60 00 40 Per cent calciferol ratios for mixtures of ergosterol and calciferol (Fig. 1.) as a function of per cent calciferol. (1) 2fO mu., (2) 240 mu.,(1) 200 mu., (4) 260 mu., (5) 264 mu., (9) 230 mu., (6) 2 7 0 m u . , (10) 294 mu., (7J 2 7 4 mu. , (8) 2 7 6 mu., (11) 296 wu.,(l 2 ) 200 mi., 1.00 ■O’ 0 -0 J 0.60 4^ i- Per Fig. 6.- The luaisterol fc*]/£ 2 8 1 r a t i o s (Fig. 2.) as ii.O m u . , ti) 240 (.5) 2 6 4 m u . , (6) 2 7 0 tnu. , 294 m u . , (l0) mu., 80 60 40 cent lunlsterol for olxtures of a function of per (1) (9) n ■o- (2) 2 0 0 (7) 296 mu.,(ll) uu., 2 7 4 mu. , SOO mu. ergosterol lOO and cent lumlsterol. (4) 260 mu., (6) 290 mu., ^ -^ t a c h y s t e r o l Jo 1^0 P i g . 7 .- T n e r a t i o s for m i x t u r e s o f e r g o s t e r o l and t e e n y s t e r o l (.Pig. S.) a s a f u n c t i o n o f p e r c e n t t a c h y s t e r o l . (i; tie au., (i-J i^40 rr.u., (i:) J-bO nu., (4) £60 mu., (i.) 1 ^ 4 mu., (6) ^ 7 0 m u . , (7) i7t u., (O) U 9 0 au. , O (.o; 1.94 m u . , (lo) Per cent 1:96 m u . , (ll) iUU :u. '-20 lO o 20 Per Pig . a.- The toxisterol 40 cent ratios (Fig. (1J 220 mu. , ( 2) (0) 2 7 0 au. , (6J LdJ 2 9 6 nu. , (10) GO toxisterol for fixtures o f ao lOO ergosterol and 4.) a s a f u n c t i o n o f per c e n t t o x i s t e r o l . 240 n u . , \i-) 2 00 mu., 2 7 4 mu. , (7) 300 a u . , 29G mu., (4 ) 260 m u . , (d) 2 9 4 m u . , 1.00 1 —4 CO 0.80 .1 0.60 0.20 O 20 Per Fig. 9.- The tachysterol ratios (Fig. 5.) as 80 60 40 cent tachysterol for mixtures o f a function of per lOO calciferol and cent tachysterol. (1) 2 2 0 m u . , (2) 1.40 n u . , (?) l&O mu., (4) 2 6 0 m u . , (5) 270 274 mu., (7) 27G mu. , (a) 2 8 8 mu. , (9) 2 9 0 m u . , (lO) m u . , (6) 294 mu., (ll) 1.96 m u . , (12) 3 0 0 mu. 15 have ratio values higher than would be possible if all the conversion products were calciferol. There is also a steady increase in the ratio values at 290 mu. and 29k mu. which are of a greater magnitude than would be possible if this assumption were true. The greatest change is observed at 300 mu. where the extinction should decrease if calciferol were the only conversion product, but actual­ ly increases are realized at all periods of irradiation. The key to the solution of this problem lies in the fact that of all the irradiation products known, there are but two products for which the ratio manifests itself with increases at these wave lengths. (.Figure 7) and toxisterol, (Figure 8 ). They are tachysterol, Fortunately, the absorption characteristics of these molecules vary greatly at other wave lengths. If the formation of toxisterol (Figure 8 ) in the irradiated solution in addition to calciferol is assumed, the ratio curves which result would show a decrease in the ratio values at 29U mu. and a slight increase at 296 mu. with a rapid increase at 300 mu. This condition would then con­ form with the requirements of the experimental results. However, with these changes there must also be an increase of ratio values at 2 UO mu., 2$0 mu., and 260 mu. of greater magnitude than is obtained experimentally. It must therefore be concluded that tlie third constituent present in the irradiation mixture is tachysterol. It can be seen from the graph (Figure 7) ttiat for increasing concentrations of tachysterol in the presence of ergosterol, the ratio values at 300 mu. would increase rapidly as well as the ratio values at 290 mu. and 288 mu. with very slight increases at 230 mu. and 2ij.O mu. While on the other hand, there should be a large decrease 16 in the ratio values at 270 rau. and 2JU mu. with smaller decreases at 260 mu. and 26h mu. It will be noted that the ratio value is constant for ratio line 250 mu. for all concentrations of tachysterol. If it is assumed that any increase in ratio value at this wave length may be attrib­ uted to the formation of calciferol, it then is possible to calculate the ratio curve attributable to calciferol and ergosterol, (Figure 22). These calculated ratio curves conform with the experimental findings for wave lengths 250 mu., 270 mu., and 27U mu. for the first eight minutes of irradiation. However, the calculated values for longer periods of irradi­ ation yield values of greater magnitude than were experimentally realized at shorter wave lengths, 230 mu., and 2 UO mu., the calculated values ex­ ceeded the experimental (Figure 22) for all periods of irradiation. It will be further noted that the ratio value for wave length 230 :nu. deviated further from the calculated values than the same values calcu­ lated for wave length 2U0 mu. This condition further substantiates the fact that the irradiation product produced other than calciferol is not toxisterol, for if the new product were toxisterol, the value for 250 mu. would increase at a greater rate than the values at 230 mu. and 2UO mu. The formation of toxisterol should then be manifested by a large increase in value for 250 mu. four times as great as that for 230 mu. and twice as great as that for 2li0 mu. and 260 mu. Furthermore, if toxisterol were forming in the irradiation mixture there would De no leveling off effect in these areas and no ratio increase in the range 290 mu. and 29l* mu. It was found in the experimental results (Figure 19) the ratio values in­ creased at 260 mu. and 270 mu. until a maximum is reached at 32 minutes E( 1 cn., 1% ) 500 200 lo< i70 £90 V.ave l e n g t h in mu. I it . - »bs.--r; t* „n curvt. of ergosterul, tac:.y sterol, and calciferol ...ixtures c<^leulateo frod extinction values of t..e ; ore c. ..j^o^ios. il; i-rc ergcsteral. (.£) erc-->steroi. 90>} tacMyi tore 1, 1*; calciferol, 9%. (2) ei fusterol, Q0'*| taCi.ysLeroi, 4/t>; calciferol, 16%, 14) ergosterol, 70%; ti-.cnyste.ro i , 9%; calciferol, 11 a , erti'Sterol, 60%; tt-.enysterol, 16%; calciferol,E4>, (6) ergo sterol, 50>; tacnysteroi, £5%; calciferol, 25%, C7) ergosterol, 40%; tachysterol, Z6>; calciferol, £4%,(a) ergosterol, 20%; tachysterol, 49%; calciferol, £1%, (9) ergo sterol, £0%; tachysterol, 54*; calciferol, 10%, (.10) ergosterol, 10%; tocr.ystero 1, t>lx; calciferol, 9%, 1.00 0.60 10 O 80 40 160 T i m e In m i n u t e s Flt 11.- The i^/K. monocnromatic irradiation (4) r a t i o s for e r g o s t e r o l i r r a d i a t e d with l i £ht o f w a v e l e n g t n 1 5 4 mu. time, (l) 130 mu., (l) 140 mu., 160 m u . , (5.) 1 7 0 m u . , (6) 1 7 4 m u . , (7) 194 mu., (y) 106 mu., (10; £00 mu. as a f u n c t i o n of (3) £50 mu., 100 mu. , (61 io 11 OO li r e K i fe. 11.- T h e monocc.roxatlc lrrauiati.u. 14; lOu i.'O mu. mu., 10) 16 d l..u .1. 6 1 r a t i o s for e r g c s t c r o l li^ot o f w a v e time. lb) (1) len^ta 1 6 0 -10 mu., 17- i.u.. v6; : o 4 mu., Cl-) 106 mu. , 111) lj-radlated w i t h rau. as a iuiictlon o f (l) 1.40 mu., 174 mu., • oo ; v7; (l) 150 mu., -76 mu., ? o O mu. (d) 1.20 1-00 -H 0.80 10 11 0.20 O l?.- The raonocnronatic Irradiation C4) 260 200 mu., 80 line 40 ratios li^ut time- ol’ w a v e Cl) U1U-, (.5) 270 (rf) 2 04 mu., l'or e r t o s t e r o l lezi^tn 26 5 mu- irradiated as CIO) (.6) 2 7 4 DU., 296 mu., (7) (.11) a witn function of (2*) 250 nu, f 2^0 iuu. , (!•) 240 m u . , at., 200 160 in minutes 276 £00 n u . , (8) mu. 1.00 HT i o.ao .1 0.60 lO 0.40 11 0.20 O Fife. 14.- 40 The 2^/ □onocnromatlc irrauiution (4) 260 mu., .-0O mu., 80 120 Time In minutes 200 r a t i o s f o r erfcosterol i r r a a l & t e d witti llfent of w a v e time. 160 l e n g t n 275 mu. ^.1) 220 mu., (2) 24U n u . , (?) 250 mu., (5) 270 nu. , (6) 274 mu., (9 ) i.o4 uu., (lO) as a f u n c t i o n o f 2 0 6 mu., (.7) 276 DU., ^.11) 20 0 mu. (8) 1.20 1.00 0.80 0.60 0.40 11 0 Fig. 80 40 120 800 160 Tima In minutes ratios for ergo3terol irradiated with 15.- The monochromatic light of »ave lengtn 181 mu. as a f metIon of the irradiation time. Cl) H O (4) 160 mu., 190 mu., (5 ) 170 mu., (9) 194 mu., mu., (i) ^40 mu., (6) 174 mu., (?) 260 mu., (7) 276 mu., (10) 296 mu., (ll) ZOO mu. (0) 1.00 o.ao 0.60 O Time lit:ion t • i.6.- T h e i c l iro.aat 1 c i E^/h. ^ lit rj-auaut i o n :.60 mu., . t t i m e . 160 60 40 r^tloc ■•r.vc ( 1 ) vJ') - 7u mu., In m i n u t e s for e r r u s t e r o l a e i i L t r 2 £ 0 mu., i r r a d i a t e d wit h Lo.'j mu. a s a function ( i ) 1 4 0 mu. , ( £ ) 174 mu., (.7) i 76 uU., 3c&0 C6) o f mu., 1.00 0.80 0.60 0.40 0.20 11 0 fig. iso 80 Tine In minutes 40 17.- The 160 ratios for ergosterol irraaiated with monochromatic light o f wave length 297 mu. as a function of the Irradiation time, (4) 250 mu., 276 mu., 200 m u . , (l) 220 mu. (5) 260 mu., (9) 290 mu., (2) l>40 mu., (6) 264 mu., (10) 294 mu., (2) 244 mu., (7) 270 mu., (11) 296 mu., (8) (12) 1.00 -o- o.ao 10 0.00 11 -o- i l g . 1H Ti.e c, / £ t - c n y s t e r o l , snQ j'Ci'c e n t 1:0 GO 40 P e r cen t e r g o s t e r o l 100 ratios cfaiclferol e r t Jste r c - 1 . U J for mixtures (Pig. lo.) 1.0 m u . , (.4) iGO mu., ^.6.) 1G 8 :nu. , 176 nu., (1) as 1. f u n c t i o n o f tue 140 mu., (6 ) 1 70 lliu. , (9.)i .‘O r.u. , (.10) 164 -u. , ol' e r g o s t e r o l , t2) ^£>0 mu., (.7) 174 mu., (.11) 100 mi., (8 ) (.11) ?00 rau. 1.20 1.00 0.80 0 0 -M 0.60 0.40 24 36 Ti m e in m i1n u tte s big. 19. - T h e t'j/l'i.ai r atios for e r g o s t e r o l i r r a d i a t e d l i g h t o f w a v e l e n g t h 235 mu. tr.e i r r a d i a t i o n (4) 48 260 uu. , (5) 2 9 4 mu. , (3J ti:.ie. to tne v i s i b l e as a f u n c t i o n (l) 23 0 mu., 270 mu., 296 nu. , (10) wit (2) 240 mu., (6) 2 7 4 mu., 300 mu. (b) 250 m u (7) 290 mu., (3) 1.00 t-l O.BO ro iM 10 0.60 0.40 24 12 T i m e In Fig. 20.- L-^&l r a t i o s for e r g o s t e r o l i r r a d i a t e d w i t h The l i g h t o f w a v e l e n g t h s £33 uu. the I r r a d i a t i o n (4) £60 mu., 290 mu., 60 48 36 a ml inut es (5> time. (1) 230 m u . , (2) ^ 6 4 mu., (.9) 204 mu., to £ 6 5 mu. (10} (6) 170 mu., 296 mu., as a f u n c t i o n of £40 mu., (7) (3) 274 mu., (11) 300 mu. £50 mu. (Q) 1.80 1.00 0.80 ao 0.60 10 O .3201 SO 2 T i m e in m i n u t e s Fig. 21. 50 40 - The Ej /^281 r a t i o s for e r p o s t e r o l I r r a d i a t e d wit h a h i g h p r e s s u r e m e r c u r y a r c a s a f u n c t i o n o f the i r r s a l a t i o n time, (l) £50 mu., (2) 2 4 0 mu., (fa) £ 7 0 m u . , (6) £ 7 4 m u . , £96 mu., (10) 5 0 0 mu. (S) £50 mu., (7) £ 9 0 mu., (4) £60 m u . , (8) 2 9 4 mu ., (9) 1.20 0.90 0.60 o-o- 0.30 0 15 60 45 Per cent calciferol *i«. 22. - The t r . / ratios of calciferol as a function of the per cent calciferol calculated from values obtained from Fig. 19. by assuming the Increase In tne ratio values of curve 3 to be proportional to the calciferol content of the Irradiat­ ed solutions at any given time. (1) 230 n u . , (2) 240 m u ., (3) 250 mu., (4) 260 mu.; (5) 270 mu., (6) 274 mu., (7) 29o ssu., (8) 300 mu. 17 of irradiation with a subsequent decrease in ratio values for longer periods of time . This is characteristic of tactiysterol in which the ratio values at 270 mu. and 27h mu. decrease rapidly with increased con­ centration of tacliysterol, together with smaller ratio decreases at 260 mu., 262 mu., and 2oh mu. and with the continual build-up of ratio values at 280 mu. and 290 mu. It is also observed experimentally that the ratio value at 27 U mu. increases at a greater rate tlian tlie ratio value at wave length 270 mu. thus suostantiating the presence of tachysterol. The presence of lumisterol was not detected in the solutions by this method of analysis due to the similarity of the spectrum of lumisterol to that of ergosterol and the formation of other intermediates in the irradiation mixture which have larger absorption coefficients. It can therefore be concluded tliat the two main products of irradi­ ation under these conditions are calciferol and tac'nysterol. The other region of ultraviolet irradiation to be considered is the short wave length end of the mercury spectrum, wave lengths 2300 A.— 2600 A., wnicii cover the lower end of the ergosterol absorption spectrum. Irradi­ ation in this region was accomplished by the use of a bromine—chlorine filter transmitting only in this region. In the early seconds of irradiation, (.f'iffdre 20) the absorption co­ efficients for the irradiated mixture increased greatly and is manifested at all wave lengths by the increased ratio values indicating the general slope cnaracteristic of the formation of calciferol. There are a few ex­ ceptions wnich must oe pointed out, for example, the ratio curve 252 exui-jits a greater slope than is found for ratio line 260 mu. If it were 18 pure calciferol formed, a plot of ratio lines 250 mu. and 260 mu. would have the same slope. Furthermore , the slope of line 2hO mu. is less than that of 230 mu. which is not characteristic of the concentration build-up of calciferol in ergosterol solutions. Wave lengths 290 mu., 29U mu. arh 296 mu. should show decreasing ratio values with increases in calciferol, yet, the irradiation mixture shows an increase. To explain these variations it is again necessary to assume other intermediates in the solution. If one product is toxisterol the ratio line of 250 mu. should increase at a greater rate than ratio line 260 mu., (.Figure c) . However, if this product were present, there would also be an abnormally large increase in ratio lines 230 mu. and 2liO mu. with 2hO mu. increasing at a greater rate tiian ratio line 230 mu. From the experimental data (Figure 20), a large increase in ratio values for 230 mu. and 2UO mu., was noted, but the rate of increase was greater for ratio line 230 mu. than for ratio line 2h0 mu. which resulted in an intersection of the lines on short irradiation. in addition to ergosterol and calciferol, If tne solution also contained suprasterol I, the increase would be primarily at 230 mu. increasing the rate of increase of this ratio line above the increase of ratio curve line 2hO mu. It can then be concluded that supra— sterol is present on short irradiations. On irradiation of ergosterol for periods longer than one minute, the ratios of all curves show either a leveling off or a sliarp decrease in the ratio values. This indicates another compound being formed or the destruction of the calciferol present at a greater rate than it is being formed, resulting in a complete ciiange in the trend of the ratio lines, it was found that the ratio curves for increased quantities of tachysterol 19 in ergosterol solutions decreases for wave lengths 260 mu., 270 mu., 27h mu., and 276 mu. with no change for 250 mu. and small increases for lines 290 mu., 29U mu., and 296 mu. wtiile 230 mu and 2 U0 mu. are un­ changed by its formation. However, during the time of formation of tachysterol, the calciferol molecules are broken down at a greater rate than calciferol molecules are being formed which would superimpose on the picture not only the ratio change characteristic of the formation of tachysterol in ergosterol solutions , but also the change characteristic of a decrease in calciferol. The series of curves in (Figure 9) indicate the effect on the ratio curves which result with the decrease of calciferol with a build-up of tachysterol showing a very sharp slope decrease in ratio values at 230 mu. and an even greater decrease in ratio values for 2li0 mu., while the values of 260 mu. and 250 mu. run parallel to each other. Thus it is possible to account for the intersection of lines 230 n u . and 2h0 mu. between one minute and two minute irradiations. On very long irradiation, it is found that the ratio values of lines 230 mu., 210 mu., 260 mu., 26Li mu., 270 mu., and 27^ mu. tend toward the common value of unity. This is explained by the complete collapse of absorption at 281 mu., the reference point of all ratios. On considering the effects of irradiation with the entire quartz mercury spectrum (Figure 21) , a composite picture of both types of irradi­ ation just discussed presents itself. It is found that in the early seconds of irradiation, the ratio curves tend to be similar to the short wave length type with very rapid increases in all ratio curves, y e t , the ratio curves do not experience the fast breakdown characteristic of this type of irradiation. 20 Again assuming calciferol as the only intermediate being formed in the solution and making a comparison with the ratio curves in (Figure 5), it is found that wave lengths 230 mu., 2UO mu., 26 U mu., 290 mu., 29h mu., and 300 mu. vary from the assumed curves. By the same argument used above, it can be established that suprasterol causes the observed varia­ tions at 230 mu. and 2h0 mu. The absorption at 26U mu. shows variation from the expected trend as indicated by the fact that the ratio values of 26b mu. do not run parallel to the slope of ratio line 260 mu. At the end of four minutes of irradiation the two ratio curves 2 bU mu. and 260 mu. intersect indi­ cating a large increase in the relative absorption at 260 mu. with respect to 26b mu. This further indicates an abnormal increase in tlie lower wave lengths wliich is characteristic of tiie trends shown by toxisterol. If toxisterol were present, there must also be a general increase in the relative absorption at 230 mu. for all mixture concentrations and a large ratio increase at 25>0 mu. however, this abnormal increase in the ratio values was not manifested in the experimentally determined curves (Figure 21). The explanation of theise phenomena lie in the fact that in the formation of the intermediates from ergosterol, calciferol is formed rapidly at first. However, under these conditions the calciferol lacks stability and is destroyed quickly while the over—all concentration of x-acnysterol continues to increase as the result of its stability in this system. The spectra resulting from a decrease in calciferol and ergosterol with a buila up of tachysterol, (Figure 16) woulc cause a rapid decrease of ratio lines in the shorter wave lengths, 230 mu., 2b0 mu., 2^0 mu., 260 mu., 21 270 mu., and 27k mu., while in the longer wave lengths, 290 mu., 29k mu., 296 mu., and 300 mu. would show tendencies toward leveling off at a maximum value which is manifested in irradiations of longer than 15 minutes. Figure 10 is representative of a series of curves calculated from the spectra of the pure compounds to show the trend of the absorption char­ acteristics of an ergosterol solution in which the calciferol builds up at a rapid rate reaching a maximum concentration of 25 per cent and fades to zero, while the concentration of tacliysterol builds up at a constant rate . The trend of the ratio values under such condition are shown in Figure 18). Since the relative quantity of any intermediate product present in an irradiated solution varies according to the wave length of the energy of activation, several solutions were investigated under monocliromatic conditions to determine the effect such radiation has on ether solutions of ergosterol. When the irradiation was carried out at wave length 25k mu., the ceneral trend was toward a general over—all increase in the ratio values for all wave lengths on short periods of irradiation. The largest in­ creases were manifested in ratio curves 250 mu., 260 mu., 270 mu., and 27 u mu., with small increase in ratio curves at 290 mu., 29k mu., and 29o mu., (Figure 11). However, if the product of irradiation were pure calciferol in an ergosterol solution the relative slope of the ratio curves 250 mu., 260 mu., and 270 mu., and 27k mu., would be greater than the experimentally determined data. Furthermore, ratio curves 290 mu., 29L mu., and 296 mu., would have a tendency to maintain their ratio values 22 or tend to decrease. Since at these wave lengths , the ratio curves experience an increase, it can be concluded that other products are pro­ duced in addition to calciferol. It can further be concluded from the slope of the ratio curves at 250 mu., 260 mu., and 270 mu., that little calciferol is formed under the influence of radiation of this wave length. Toxisterol is excluded as a possible isomer present in the irradiated solution; for if this product were present the ratio curves at wave length 230 mu., 2 U0 mu., 250 mu., and 260 mu., would experience a greater in­ crease in ratio value than is experimentally realized. The presence of tachysterol in the irradiated solutions is character­ ized by a decrease in the ratio curves at 260 mu., 270 mu., 27k mu., with no change in ratio values for wave lengths 230 mu., 2 UO mu., and 260 mu. Experimentally, a decreased ratio was found for ratio curves 270 mu., 27k mu., 290 mu., 29k mu., 296 mu., and 300 mu., with the rate of increase of the ratio curve at 260 mu. leveling off after 60 minutes of irradiation. It was concluded that tachysterol is building up at a very rapid rate oncer these conditions. It was further concluded that little suprasterol I and Ii are present since the relative rate of increase of ratio curve 230 mu. is similar to that of curve 2 l|0 * u . If a similar analysis is made of tlie ratio curves calculated from the aosorption spectra of irradiated ergosterol wliich has been irradiated under similar conditions using other wave length of monochromatic radiation, it is found that at the short wave lengths i.e. 260 mu., Figure 12, and 265 mu, Figure 13, similar reaction products result. At these wave lengths, the primary products appear to be calciferol, tachysterol, and suprasterol 23 varying only in amounts of each produced. For example, at vave length i .-0 mu., the ratio curves indicate the rapid formation of calciferol in the first minutes of irradiation, however, at no time does a large con­ centration exist. At the same time tachysterol is formed at a slower rate, .;ut ouilus up to high concentrations causing a maximum to be formed in ratio curves, Figure 12, at the end of 60 minutes, under conditions of irradiation with radiation of wave length longer i n n 203 mu. there is no tendency for the formation of maxima in the ratio .curves . Thus it can oe seen in Figure lii, that while no maxima are formed, t:.ore is a greater tendency for the curves to level off at 60 minutes of Irradiation time than is manifested in samples irradiated at longer wave I'jM.-'tns . Also , it should be pointed out that at this wave length the •;io; e of the ratio curves at 230 mu., 2uO mu., and 300 mu. is greater than those of Figures 13, 16 and 17, which indicate the presence of greater -^entities of tachysterol oeing former, on irradiation at wave length 275 . ti.a ri at the longer wave lengths. Furthermore , the ratio carves In .1cat-, thv p css isle huila-up in the irradiation solution of 1 ami sterol ri-.arily wit:, irradiations carried out at 261 mu. At these longer wave l-ngtns tne presence of suprasterol I was not detected. 2JU GENERAL DISCUSSION It has been shown that tachysterol is formed in ergosterol solutions when such solutions are irradiated at various wave lengths from 25U mu. to 2y7 mu. however, the quantity of tachysterol present for the same periods of irradiation decreases with increase in wave length of the incident radiation. Windaus and co-workers (18) found by irradiating ergosterol until 60 per cent conversion of ergosterol took place, the tachysterol could be readily isolated from the solutions whereas the separation of calciferol under such conditions was more difficult. As previously pointed out, the tendency during irradiation at short wave lengths of ultraviolet is for the relative concentration of calciferol to build up in the early minutes of irradiation, but to decrease with in­ creased irradiation time. Other investigators have shown the instability of calciferol at short wave lengths of irradiation. Askew et al., (l) irradiated ergosterol with light of 280 mu. to 297 mu., then separated out the ergosterol from solution and again irradiated the converted products with xylene filtered ultraviolet radiation, bromine-clilorine filtered radiation, and unfiltered radiation of a mercury arc lamp. They found little change in the concentration of the preparation in the first case, almost complete destruction in the second case and a concentration con— siderably lowered in the latter case. They also observed, however, an in­ crease in the extinction values of the absorption spectra at 250 mu. aroden and workers (L) irradiated calciferol in ethanol and found that the rate of destruction of calciferol was greatest at wave length 265 mu. 25 This concurs with the results obtained, since it has been found that calciferol is destroyed very rapidly in unfiltered mercury arc light, and that the rate at wiiich calciferol is destroyed is a function of the wave length of tne incident light. Figure 23 is a plot of the change of extinction of the absorption spectra of calciferol at its maximum, 263 mu., wi in respect to time of irradiation indicating that calciferol is destroyed -.ost rapidly on irradiation with radiation 263 mu., 25U mu., 2L1 mu., 2 9b mu. Furthermore, the rate of oreakdown appears greater in diethyl c than in ethanol for any given wave length of irradiation, Figure 2 U . o . Green irradiated calciferol in ethanol and concluded that the rate of breakdown when irradiated with a new mercury arc tube is -renter thian when the tube has had many hours of operation. The trends with regards to suprasterol indicates the formation to oe o function of length of indicant light. Greater yields appeared to be present, in solutions which were irradiated witdi radiation of wave lengths shorter than 2t>5 mu. and with the highest yield oeing at 265 rau. This Is to uo expected since calciferol 'undergoes a greater rate of oreakdown md e r these conditions. The presence of suprasterol was not detected on irradiation with light of wave length 261 mu., 2 65 mu., or 29o mu. Toxisterol reported in the literature to be present as an end product i.e., a decornpositlon product of calciferol , was not detected under the conditions described herein. I W f f a r « c a of axtlnction at 266 ^u. ► *i t-H ( t> Jr *t r * o c h «h *V c*H <♦ >•■!*■►“ a <* •«' • i r» £.66 2.57 0.6 1.8 Time In reclj-lcai minutes Fig. £4. - T.'.e logarithm of tne cnange In extinction o f calclferc- at Its absorption maximum, i£t mu. as a function o f recipical time snowing tne rate of destruction of cal­ ciferol In inflltered quarts mercury arc llgr.t In different solvents. Cl) etnanol, 12) dletnyl etner. 26 SUMMARY (1) The effects of various wave lengtlis of ultraviolet radiation on ether solutions of ergosterol liave been studied and a system devised for the qualitative analysis of the absorption spectra of irradiated solutions, using this system of analysis it has been found that* Ca) Calciferol is formed early in the photochemical reaction at all wave lengths studied, but due to its instability in radiation of short wave length, i.e., wave length 265 mu. and less, the relative concentration of this compound decreases rapidly on extended irradiation. \.o) Tachysterol is formed at all wave lengths, but is found in larg­ est concentrations in solutions wtiich have been irradiated with radiation of wave lengths less than 280 mu. ^c) Lumisterol was detectable only in solutions which had oeen irradiated at wave lengths longer than 2 60 mu. U) The suprasterols appear to form at a greater rate at 265 n:u. and to a lesser extent on short wave lengtr. irradiation. (e) Toxisterol was nut detected in tlie irraaiatec ether solutions under the conditions which the irradiations were carried out. C2) The effect of two different regions of the ultraviolet band on ^o:.cr solutions of ergosterol were studied* (a) 2cu mu. and less. Co) 2 in mu. and longer. It. was founc tu^at the longer wave length radiation tends to form -or.rounds whose aosorotion spectra lie in the snort wave length range and 27 the shorter wave length light tends to form compounds whose absorption spectra lie in the longer wave length range of the ergosterol absorption sand . (3) The effects of various wave lengtlis of ultraviolet radiation on the decomposition of calciferol in ether solutions have been studied and it has been found that the destruction rate increases with decrease in wave length with the exception of wave length 265 mu. which exhibits the greatest rate i.e., 265 m u . > 25^ mu. 7 281 mu. 7 296 mu. (U) The effect of solvent on the rate of breakdown of calciferol has ueen determined and found that the rate is greater in diethyl ether tuan that in ethanol. 26 LITERATURE CITED 1. F. A. Askew, R. B. Bourdillon, H. M. Bruce, R. G. C. Jenkins, and T. A. Webster, Proc. Royl. S o c . (London), 107B. 76, 91(1930). 2. C. E. Bills, E. M. Honeywell, and W. M. Cox, J. Biol. Chem., 92, 601(1931). 3. n. Brockmann, "Ergebnisse Der Vitamin- und Hormon- Forschung," Vol. II, Academic Press Inc., Mew York, ( 1 9 L D . u. F. P. Browden, and C. P. Snow, Nature, 1 2 9 . 720(1932). 5. J. Green, Biochem. J., h9, 232(1951). _. T. ft. Hogness. A. E. Sidwell, and F. P. Zscheile, J. Biol. Chem., 1 2 0 , 239(1937). 7. W. A. Noyes Jr. and P. A. Leighton, "The Photochemistry of* Gases," fteinhold Publishing Corporation, New York, (19Ll). c. n. V. Phillips, Gloelampenfabricken, G. P. 63li,llt6. 9. 10. 0. Rosenheim, Nature, 1 2 1 , T. Ryder, G. Sperti, 1 0 6 , L52(1936). G. P. 570(1926). Goode, andn. G. Cassidy, J. Am. K e d . Assoc. 11. P. Setz, Z. Physiol. Chem., 215 , 163(1933). 12. L. Velluz, and G. Amiard, C. ft. A c d . Sci. (Paris), 226, 692(1969). 13. L. Velluz, G. Amiard, and A. Petiet, Bull. Soc. Chem. (Paris), 16, 501(19^9). lit. T. A. Webster and ft. B. Bourdillon, Biochem. J., _22, 1223 (1926). 1^ . A. Windaus and E. Anhagen, Z. Physiol. Chem., 1 9 6 , 108(1931). Ic . A. Winciaus , K. Ditlimar, and E. F. Fenholz , Ann., It93 . 265 (1932). 17. A . Windaus, J. Guede , J. Koser, and G. Stein, Ibid. Lb3, 17(1930). 16. A. Windaus, F. von Werder, and A. Luttringhaus, Ibid. It99 , 166(1932) . 8 V *8 « ► r .'H P (’*> I #jfT) V t» » ♦ < * r»O’*1 500 200 lOO £70 I i t c iengtn In mu. £90 Absor; tliij Cu: v-. . _i ergo sterol solutions snosring tne ccjkngei In absorf tlor. a t e J t e ;iice on irradiation wltn tbe light of • quartz mercury are filterec witn cniorlne and bromine gas for Taring lamgtr.s of time, (1,/ 0 minutes, (£) 4 minutes, c minutes, (4) 16 minutes, i. t j 51 minutes, v.6.) 61.6 minutes. I late 500 SOO a o 100 £50 £50 £70 live length In mu. £90 Absorption cjrvi 3 of ergosterol solutions shoeing the changes in absorption > ■ .ich take place on Irradiation with Mono­ chromatic lifc.'.L of wave length £54 mu. for earing lengths of time. (1) 0 minutes. (£) 40 minutes, (3) 60 minutes, (4) ISO minutes, (5) ISO minutes. Plate IV 300 ,A a o 100 £70 £90 ftave length I n au. A b sor a - t i o n c u r v e s o f e r t o s t e r u l s o l u t i o n s S h o e i n g t h e c h a n g e s In a b s o r b t i - n .. o n I r r a d i a t e in w i t h too n o cor-zDutic llg.it of * a v a lengt.* 2 6 0 m u . f o r v a r i n g l e n g t h s o f t i m e . (1> 0 m i n u t e s , (2) 60 m i n u t e s , (.2) 120 al..utej, (4) 130 ainutes. Plate V 2Z0 250 270 290 length In mu. . Absorption currti of ergosterol solutions showing ths changes in absorption which take place on Irradiation with mono­ chromatic llfcfit of wavelength 265 au. for waring lengths of tlae. (l) O minutes. (2) 40 alnutes, 60 alnutes, (4) 120 minutes, (5) 180 alnutes. (z) Plate VI 300 200 a u 100 050 VaTe length lr. r u . Absorption curves of ergosterol solutions showing th«change In absorption w h i c h take place on Irradiation w i t h m o nochromatic light of wave length 275 mu. for v a r y ­ ing lengths of time. Cl) 40 minutes, (r) 60 minutes, (4) 120 minutes, (5) ISO minutes. Plata VII 300 200 13 o lOO 230 270 250 Mave length In au. 290 Absorption curves of ergosterol solutions showing the changes In absorption take place on Irradiation with mo n o ­ chromatic llg. ‘ A&ve length 281 mu. for varlng lengths of time. (1) 0 mli,.i.as, (2) 40 minutes, (3) 60 minutes, 14) 120 minutes, (5) 190 minutes. Plate VIII 8001--- 8 lOOfc^ 230 250 Save 1 eng til In au. 290 Absorption curves of ergosterol solutions showing the Chengs In absorption take place on irradiation w i t h sonochroaatic light oi’ ware length 285 au. for waring lengths of tlae. (1) 0 alnutes, (2) 60 alnutes, (s) 120 alnutes, (4) 180 alnutes. Plate XX 300 soo a o —« lOO 870 ■see length In au. Absorption curees or ergosterol solutions shoeing the changes In absorption Ich take place on Irradiation with aonochroaatlc llg-.t. . wave l a n g th 296 au. for earing lengths of tlae. (l) O ni...',3. (2) 60 alnutes, (3) 60 alnutes, V4) 120 alnutes, (bj 180 alnutes. Plate X a 29 II. THE CHROMA.TOCShA.PHIC SEPARATION OF THE IRRADIATION PRODUCTS OF ERGOSTEROL The problem of separating pure vitamins D from the natural sources ana from the irradiated provitamin is one which has been studied ex­ tensively in the past. Angus, Askew, Bourdillion, Bruce, Callow, Fishermann, Philpot and Weoster (l) separated vitamin D 2 from the irradiation mixture oy the precipitation of the unchanged ergosterol with digitonin in ethanol and separated the residues into their components by molecular still distilla­ tion. The advantages of this method has been questioned Decause of the instability of the vitamins and other intermediates to heat (2U). Concurrently, Windaus, Lutringhaus , and Deppe (26) isolated the pure vitamin ticrough a cliemical procedure utilizing toe citraconic and maleic acid ester formation of the intermediates and fractionally crystallizing out the calciferol ester from the isomeric mixture. Askew, oourdillion, oruce, Callow, etc., (2) prepared vitamin D 2 removing the unconverted ergosterol with digitonin, esterifying the calciferol to calciferyl 3, 3, dinitrobenzoate , and fractionally crystal­ lizing out the ester. The vitamin was recovered through hydrolysis of toe ester . More recently ciiromatography has oeen successfully used by various investigators for the isolation of the pro-vitamin and the separation of complex sterol mixtures . Winterstien and Stein (2c) using activated alumina as an adsorbent, separated ergosterol and cholesterol into two colorless bands using oenzene as the developer solvent. 30 Landenburg, Fernholz, and Wallis (18) formed the azobenzene-monocarooxylic acid esters of sterols to produce colored products enabling tneui to follow the separation bands more readily. Benzene was used as the solvent with a mixture of benzene and petroleum ether as the de­ veloper solution. After studying several different sterols, they con­ cluded that only those sterols differing in the number of double bonds could be separated. Several investigators (8) have utilized chromatographic procedures tor the separation of vitamin D a and D a from fish liver oils and fortifiei fish liver oils for analytical purposes. In each case the object jf the investigation was the separation of the vitamin from the carotenoir.s, and sterols which interfere witn their analysis. Miller (19) separated vitamin D from ooth natural oils and irradiated provitamin solutions by adsorbing the vitamin from benzene-ether solutions -sing trisodium phosphate as the adsorcent. DeWitt and Sullivan (10) separated vitamin D 2 from natural oils and irraaiated ergosterols by a chromatographic procedure. The chromatographic column consisted of a 1 * 1 mixture of magnesia and diatomaceous earth as *.n-j adsorcent with petroleum ether as the solvent and developer. In these laboratories Young (29) used a two step cliromatographic r"oeedure developed oy Ewing, Kingsley, crown and iwnmett (11) for the separation of vitamin D 2 from irradiated ergosterol. The irradiated mix­ ture of ergosterol was passed over a nSuperfilterol" column using a cexane—atner—ethanol solvent to remove the pigments, and the sterols in u:,e second step on a column of the same nature using a benzene-Skelly 31 Solvent mixture as developer. He found the results for irradiated ergo­ sterol to be erratic. Baker (3) using a modified procedure of Hage (ll;) investigated the separation of irradiated ergosterol mixtures in c o m oil on "Superfilterol" columns with hexane-ethanol-ether solvent for development. The results ootained were satisfactory for ixigh vitamin oils , but experienced diffi­ culty in the lower concentration range. Trie separation of pure ergosterol and pure calciferol has been stuaiea . Baker (3) used the procedure outlined above ana found that ergosterol was successfully separated from calciferol with trie calciferol fraction piscine trirough the column while ergosterol is retained. Bullard (7) studied the same separation more extensively and claimed tj, uo 9 9 per cent recovery of calciferol from such a system. Chen (9) studied the separation of vitamin D 2 , vitamin D 3 , vitamins A, in: ergosterol on various adsorbents and concluded that vitamin D could .: separatee from vitamin A and ergosterol quantitatively on "Superfilteroi" columns using hexane—ether—ethanol solvent for developers. Powell (20) using a modified chromatographic method of Ewing et a l . Ill) found the separation of vitamin D 2 on "Superfilterol" to be good, yielding consistent results. however, on conversion to larger scale separations, the results varied. The investigator found that the chromato- yrap-.ed materials in the eluate of such columns possessed aosorption ourvos identical to those of calciferol, but tne percentage calciferol of to.. resulting solid calculated from spectrographic data amounted to 32 90 per cent purity. The impurity present was assumed to be a substance whose absorption spectrum appears in the visible region of the spectrum. The cltromato graphic procedure for the separation of calciferol has been successfully worked out on the analytical scale, however, attempts to isolate large quantities of the pure vitamin by this procedure have not been successful. This study was carried out to determine how the or^sence of various irradiation intermediates affect the chromatographic separation of pure calciferol when produced under different conditions of irradiation. 33 PROCEDURE Two types of cliromato graphic columns were used in the study of the separation of irradiated ergosterol. Column I — The chromatographic column consisted of a l.b cm. tube packed with "Superfilterol" to give a column 7 cm. long (12 g.) . The '•SuporfilterolM was packed in the column in tiiree portions of U g . each ander a pressure differential of 10 cm. and each portion tamped before the additional portion was added. Prior to use, this column was washed *rith 50 ml. of the developer solution consisting of 50 volumes Skelly Solvent, "b" , 10 volumes of diethyl ether, and one volume of etiianol. The column was pressurized with carbon dioxide to give a drip rate of one irop per tiuree seconds. Column II — The column consisted of ii g. of alumina in a 10 mm. tube packed under a 10 cm. pressure differential. column was washed with 10 ml. of diethyl ether. Prior to use, the The developer consisted 50 volumes of hexane and 50 volumes of diethyl ether . Several methods were used in the irradiation of the ergosterol solu­ tions in these determinations. Procedure A — Irradiation cell. A 75 nil. quartz erlenneyer flask was used as the Ergosterol was weighed to yield solutions of 0 .01i-i7 g./ -0O ml. of ether and placed in the cell. The flask had been previously swept out with carbon dioxide to insure the absence of air. Prior to Irradiation, the ergosterol solution was bubbled with a stream of carbon iioxlde saturated with ether for a period of four minutes. After the 3k solution had bean irradiated for the desired length of time, the solvent was evaporated off under a partial vacuum in the presence of carbon dioxide. The residue taken up in 25 m l . of an alcohol solvent consisting of 902 volumes of ethanol, U7 volumes of methanol, and U5 volumes of water. (The solubility of pure calciferol was determined to be 0.1 gm/ml at room temperature, 2U°C . , under uncontrolled condition in this solvent) . This solution was then cooled to - 20°C . and allowed to stand over night. The ergosterol was filtered off and washed twice with solvent. was evaporated to dryness under vacuum and carbon dioxide. The filtrate The residue was taken up in tiiree ml. of the (50 volumes of Skelly Solvent MBM , 10 volumes of ether, one volume of ethanol), solvent, and placed on a column, Type I, w:icti had previously been washed with 50 ml. of the developer solution. Toe rate of flow through the column was regulated to one drop per three seconds by pressurizing the column with carbon dioxide. The eluate was collected in five ml. aliquants. The spectrum of each aliquant was determined by diluting the sample sofficiently with developer solution to give extinction values of approxi­ mately 600 extinction units on the Beckmann D. U. spectrophotometer at x-he curve maximum. Procedure B —- A 300 ml. commercial type Hanovia irradiation cell containing a 12 inch, high pressure, A. C., 8.5 Amp., 205-235 Volt, mercury arc Darner was used as tlie irradiation cell. Ergosterol was dissolved in fresiily distilled ether to give a solution of 0.25 g./lOO c c . After the ether solution had been placed in the irradiation cell, ether saturated carbon dioxide was bubbled through the solution for four minutes prior to 35 irradiation to expell any remaining oxygen. The bubbling was continued through the entire irradiation to maintain the inert atmosphere and to insure uniform agitation. After the solution had been irradiated for the desired period of time, it was evaporated under carbon dioxide and a partial vacuum to dryness. Hie residue was taken up in 250 ml. of alcohol mixture ^.Procedure A) and cooled to —20°C . over n i g h t . The crystalline precipi­ tate was filtered off and filtrate evaporated to dryness under partial vacuum and carbon dioxide. The residue taken up in six ml. of developer consisting of 50 volumes of Skelly Solvent "BH , 10 volumes of diethyl ether and one volume of ethanol and chromatographed in the same manner as described under procedure A. Procedure C — 0.25 g. of ergosterol were dissolved in 100 ml. of freshly distilled e t h e r . The solution allowed to flow through a thin quartz cell 0.21 mm. thick, 125 mm. long, 11 mm. wide, by gravity flow ana collected as it passed out of the cell. The time of exposure was varied by recirculating the solution through the cell until the desired time exposure was obtained. The sample was collected and the ether dis­ tilled off under carbon dioxide and partial vacuum. The residue was taken ap in six ml. of developer solution consisting of 10 volumes of diethyl ether , 50 volumes of Skelly Solvent W3M , and one volume of ethanol. This solution was placed on a column previously prepared as outlined under Procedure A and 3. The fractions were collected in one ml. portions. -•vfter 75 portions were taken, the column was eluted with five ml. of ethanol to desorb any remaining sterols. The spectrum of each portion was determined in the Beckmann D. U. spectrophotometer. 36 Procedure D — A five ml. aliquant collected from a MSuperfilterol" column was evaporated to dryness under partial vacuum and carbon dioxide. The residue taken up in three ml. of (50 volumes of Skelly Solvent "3W , and 5>0 volumes of ether) , solvent and placed on an alumina column of Type II which had previously been washed with IO m l . of anhydrous ether. Tne column was then developed with eight m l . of the mixed solvent and eluted with 15 m l . of pure ether. The spectrum of each fraction determin­ ed by diluting in eluate in sufficient solvent (50 Skelly Solvent nB ,*-50 ^ther) to give an absorption density at the curve maximum of approximately cOC density units on the Beckmann D. U. spectrophotometer. 37 DISCUSSION In part I of this thesis , it has been shown that under all condi­ tions of irradiation tachysterol is formed. The rate of build up of tliis compound in an irradiation solution under a given set of conditions is constant for long periods of irradiation time and tachysterol has a tendency to be more stable in the presence of short wave lengths of ultra­ violet radiation than calciferol. Furthermore, the formation of the suprasterols as ooserved on the irradiation of pure calciferol have a tendency to form at a greater rate on irradiation with short wave length radiation, with wave length 265 mu. appearing to be more advantageous for their formation. Lumisterol on the other hand manifests itself to a .greater extent in the longer wave length irradiations. The possibility of other isomers being present in the irradiation mixture must also be considered since evidence has been presented for one presence of protacliysterol (2U) , precalciferol (2 2 ,23 ), and suprasterol (13), under varying conditions of irradiation and aging. Other ao- sorption phenomena as observed by Browien ^t al. (6) and Windaus et al. (2a) have neither been explained nor has a scheme of analysis been de— visea to totally isolate all isomers formed in the complex mixture which results at various stages of activation. The results of this investigation have oeen divided according to t.ne condition under which the irradiation was carried out, i.e., according to the type of cell and the source of light used in activation. various conditions used for activation are classified in Table I. The 36 TABLE I CONDITIONS FOR IRRADIATION OF ETH-cR SOLUTIONS OF ERGOSTEROL Cell Type Solution Thickness nrlenmeyer flask 35 m m. rianovia commercial cell 10 ra m. Light Source Unfiltered , Uviarc D . C . , 12 a m p ., 2iiO volts , mercury lamp p-xylene filtered, Uviarc D. C., 12 amp., 2hO volts, mercury l a m p . pressure , A. C., 8.5 amp., 205-235 v o l t ,mercury arc lamp, p— xylene filtered,high pressure, A. C., 8.5 amp., 205-235 volt, mercury arc lamp. • 3 3 Flat ribbon cell ■ ro Unfiltered ,liigh Unfiltered, high pressure, A. C. 8.5 amp., 205 v o l t , mercury arc lamp. Irradiation in a quartz Erlenmeyer flask offers observation with re­ gard to tiie reaction in which the ultraviolet radiation is impinged upon a thick layer of solution, thus on activation each molecule of sterol is in the optical p a t h , but the path of solution is of sufficient thickness so that the ultraviolet radiation is totally absorbed before the entire solution is traversed. Therefore, each molecule, even though rapid agitation was maintained, would not De evenly irradiated. Trie series of curves of Figure I, show the changes which take place in ether solutions of ergosterol in this type of cell and umiltere d mercury irradiation for varying length of time . When such solutions are ciiromatographed on "Superfiluerol" columns of Type I, they are primarily separated inuo two major bands, those having the ergosterol type structure 39 i.e. the sterol type compound and those whose structure has been con­ verted by the irradiation. The sterol type compounds are retained on the column and are eluted only after extensive development. The con­ verted compounds are eluted from the column as indicated in Figure 2 to figure 7, where each curve is representative of a five ml. fraction of ^luate and the number indicates the order in which the eluate was col­ lected. This system of designation of absorption carves will be used throughout tiiis thesis . Tiiese curves indicate a complete separation otrie ergosterol type compounds as well as the compounds present whose absorption spectra lie in the short wave lengths region. There was incom— plete separation of the calciferol like and tachysterol type compounds. .lOVKver, it was found that oetter separations are achieved with regard 10 calciferol in solutions of least irradiation. In Figure 2 to Figure 5, the fraction of eluate which is enriched in calciferol in each case is fractions ..o. 2, which has its absorption maxima at 26t mu. for solutions o; snort exposure, but as the time of irradiation increases, the maximum o: tne fraction is disolaced to the longer wave lengths until with uo minutes of irradiation the maximum is located at 275 mu. Fraction No. 3, tne first fraction after the passing of the calciferol, has its maxima at 260 mu, which is the tachysterol rich fraction. Fractions no. U show increases in the intensity of the absorbent present indicating the passage of the first major oand. Fraction u o . 5 and n o . o are the first fractions of tne second band containing the ergosterol type compounds which are first manifested by a strong absorption peak at 252 mu. 0 .60 JC *ave lengtr. I Fig. 1. - Absorption cur»«s o f ergosterol Irradiated for T»rlo\,s lengt.-.s o f tlae lr. a c^artr Erlenmeyer flask using u n f i l t e r a d ultrariolet llgr.t. (l> 10 alnutes, \k.) 15 isin«.tes, \.Z) 2 0 aJ.ns.tes, (4 ) 40 ninutes. 15.00 11.25 T f l a s k u r i n r u r f l l t e r e d u l t r a v i o l e t ll»ht. a o -d o a 3 50 4-* o~ "iVf ler.^t F i e . 5. - J b s o r p t l o r . c a r v e s of v a r i o u s f r a c t i o n s of e l u t>-. o b t a i n e d c h roratcurrapnlru' t h e a l c o h o l s o l u b l e f r a c t i o n of ericosterol irradiated f o r 4 C relnutes lr. a q u a r t s Fr! enr.eyer f l a s k u s l n ? u n f i l t e r e d u l t r a v i o l e t li^ht. hO -i.rradia.ted solutions activated in p—xylene filtered ultraviolet radiation, Figure o, the fractions of eluate collected in one ml. Dortions, yields very regular curves of the calciferol type although their maxima lies at 2o6 mu., Figure 7. It is further to oe noted that the first frac­ tions eluted from the ciiromatogram are not those of calciferol, but possesses absorption maxima at 272 mu. and 2&h mu., Figure 7. From the above data it appears that the most difficult separation is that of calciferol and tachysterol. A cell of larger capacity was selected in which the thickness of chs ergosterol solution was reduced to 10 mm., first., to give larger quantities of material and second, to decrease the ranaomness of the irradiation. In such a cell, the concentration of the solution was so selected to utilize most of the energy' in the activation range. It was further felt that with vigorous agitation the amount of energy received y each molecule would oe more uniform. A series of experiments were devised to determine both the effect of time of irradiation and wave length of irradiation on the separation of the resulting intermediates using "Superfilterol" columns of type I acccrc.ing to "Procedure 3." From these data, Figure 8 to Figure 1$ , in which an unfiltered source was used , it is seen that with increased time of irradiation of the ergo­ sterol solution, the tendency is for the extinction coefficient of each portion of eluate to increase:. In fraction . 2 of Figure c, the maxima are located at 2r'3 mu. and 293 mu., however, as the time cf irradiation is increased fraction H o . 2 has a tendency to shift towards shorter 6 0 .0 idctinctijn 45.0 so.o 15.0 230 270 250 Wave length In m u . 290 Fig. 6. - Absorption curve of ergosterol irradiated for 15 minutes In a quartz trlenmyer flask using p- xylene filtered ultraviolet light. IT.00 Jatir.ctl^a O cx o? 250 Vave lamcth In ml. v. - Absorption curves of various fractions of el’iate obtained chroiMto*raphln» the alcohol soluble fraction of ergosterol Irradiated for 15 minutes In a quartz "rlenneyer flask uslrur p- rylene filtered ultraviolet lie.ht. FIs’ . 1.80 1.20 0.6 0 270 250 Wave •i length In mu. 290 trig. 9. - Absorption curr«« o f rirlous fractions of eluate obtained chromatographing the alcohol soluble fraction of ergosterol irradiated 8 minutes in unfiltered ultraviolet light. hxtlnctlon 2.40 60 290 tare lengt.i In au. rig. 1 .. - A ^ a a . - p :lr,n curves o i t a . - I c l s ir&cti'r.s oD'.ainei :,.rjiatograf .'.inj; t:*e alc;..vi sclucli iractiir. oi ergosterol Irradiated li linutes ir. ur.liltereo ultraviolet lig.ot. 230 250 270 290 Wave length in «u. H g . 11. - Absorptl >n curves oi various lractions oi eluate ootalned chromatographing tne alcohol soluble 1'raction or ergosterol Irradiated lb minutes In unl'lltered ultraviolet light. 1-80 I,. +> 3 0.60 270 Mave l e n g t h In oru. 290 Fig. 12. - A bs o r p t i o n curves various J'rictljns or eluate o b t a i n e d cnr o a a t o g r ap h i n g tne a lcohol soluble fraction of e r g o s t e r o l I r r adiated 20 m inutes In unflltered ultraviolet light. o i 40 -o C*(P) L'7. - air. 6 - h e v e - i : s ^ r ; : , ^ r 4 c^-v e : C .T t - » • dp Ti»' t l A r . g t r. j; t ^ f&ri-: -s & >.C 1 'i * ~ . frictl-r s ;f e l ^ t e 5^ C- - 1 T a C t * . ri Da erg; ttir;: irrac-at ei 26 '- Ir.-;tes lr. ur.i ilterej ultraviolet light. 2.40 e x tin c tio n i.ao 1.20 0.60 o o 270 Wave length in mu. 290 llg. 14. - Absorption curves of various fractions of eluate obtained chromatographing the alcohol soluble fraction of ergosterol irradiated S O minutes in unllltered ultraviolet light. 2.40 1.90 51.20 0.60 -o-o250 £70 ■av e l e n g t h In b u . £90 frig. 15. - Abacrj.Li.on curves or T&rlouj fractions ol eluate obtained cnronatO£raf.hing tne alconol soluble Traction oT ergosterol Irradiated 40 minutes in unflltered ultraviolet light. Ul wave lengths as illustrated in Figure 13, where the maximum lies at 275 m u . after a period of 26 minutes of irradiation and proceeds to even shorter wave lengths with further increases in time as in Figure 15. These phenomena are due to increased quantities of calciferol like material in the original solution which causes the adsorption characteristics of the column to change. This effectively floods the column causing various components of the starting mixture to appear in earlier fractions eluted from the column. In subsequent fractions, fraction No. 3 of Figure 8 to Figure 15, the maximum in each case seeks higher extinction values as time of irradiation increases and the maximum in each case lies at 270 mu. except for the iiO minute irradiation, Figure 15, in which the maximum lies at 275 mu. indi­ cating a tendency for this fraction to shift towards longer wave lengths on extensive irradiation. This trend is also true in fractions No. h of Figure 8 to Figure 15, in which the absorption curves of irradiation mix­ tures of short irradiation periods manifest themselves at 270 mu., Figure y, and increase in extinction with a shift of maximum to 2 80 mu. on long irradiation, Figure 15. Fractions No. 5 and No. 6 in Figure 18 have maxima at 252 mu. and 282 mu., however, as the time of irradiation in­ creases the maximum at 252 mu. disappears yielding only a small maximum at 282 mu. When similar experiments are conducted using a p —xylene filtered source, the rate of conversion of the ergosterol into its isomers and Dreakdown products is greatly reduced. This is due to the absorption by the filter of the short wave length end of the mercury spectrum where the energy distriDution of the arc is the greatest and in the range where U2 ergosterol has its highest absorption power. Furthermore, the filter transmits only 85 per cent of the light at 28k mu. and does not transmit IOC per cent below 290 mu. This in itself has a great effect on tlie resultant spectra. The series of graphs in Figure 16 are representative of the change •which takes place in the absorption spectrum of ergosterol with increased time of irradiation under the stated conditions. The irradiation products whose absorption spectra lie in the 230 to 275 mu. range ouild up slowly ana become more pronounced as time of irradiation increases, while the products whose spectra lie in the range above 280 mu. are not observed. Thus it will be noted in the early minutes of irradiation, the absorption maximum at 270 mu. increases with respect to the absorption maximum at 2cl mu., 15 minute irradiation, Figure 16, and by 10 minutes of irradiation the entire short wave length portion of the ergosterol curve has increased greatly with the longer wave length portion decreasing in magnitude, (Curve n o . 2). When such solutions are subjected to ciiromatographic separation on "Superfnlterol" columns of type I, using '’Procedure B ," the absorption curves of each of the fractions, Figure 17 to Figure 23, are similar to those of the unfiltered light irradiation fractions i.e. Figure 8 to Figure 15, with the exception of fraction No. U which lacks the maximum at 280 mu. A maximum which lies at 270 mu. is present, in both fractions No. 3 and uo. k while subsequent fractions No. 5 and No. 6 show maxima at 252 mu. and 280 mu. with a decrease in extinction values with increased time of irradiation. 1.20 0.90 230 250 270 V.ave length In mu. 290 Fig. 16. - Absorption curves of ergosterol Irradiated for various lengths of time in a uanovia quartz cell using pxylene filtered ultraviolet light. 11) 5 minutes, (.2) 10 minutes, 13) 20 minutes, (.4) 29 minutes, (.5) 35 minutes. £.40 1.00 0.60 270 250 Iftavm length In mu. Klg. 17. curves of various fractions of eluate obtained chromatographing the alcohol soluble fraction of ergosterol irradiated 4 minutes in p- xylene filtered ultra­ violet light. £.40 1.80 o 1.80 0.60 850 850 Var* len*th la w . r u - 18. - Absorption curves of var i :uj rrmc •:1 :r.a obtained chromatographing the alcohol soluble r n c c t - c ergosterol Irradiated 3 minutes in p- xyLena filtered ilcravlolet light. 2.40 1.80 O 1.20 0.60 o— 230 270 250 Wave length In mu. Fig. 19. - Absorption curve* of various fraction* of eluate obtained chromatographing the alcohol soluble fraction of ergosterol Irradiated 12 minutes In p- xylene filtered u l tra­ violet light. 2.40 1.80 c o O ti 4-> 3 o- 0 .6' O £70 i.ave lengtn in mu. 290 Fig- £0. - Absorption curves ol' various Tractions oi’ cluate obtained chromatographing tno alcohol soluble Traction of ergost- rol irradiated £0 iiinotes In p- xylene filtered ultra­ violet light. S. 40 1.80 0.80 p- J7U have length In mu. H g . ll. - Absorption curves ol various l r a c t b n s of eluate obtained chromatographing the alcohol soluble fraction ol' ergosterol irradiated b4 minutes in p- xylene liltered ultra­ violet light. 2 .4 0 1.80 c o O 3 3 ** 0.60 _ z -o-o- - O- - 290 270 m k>ve length In iu< “■ F lg • 22. - Absorption curves or various fractions of eluate obtained chromatographing ti.e alcohol soluble fraction of ergosterol Irradiated 30 minutes in p- xylene filtered ultra­ violet light. 260 £.40 1.B0 — a o 4-» wa -4 1.20 - ---4J 3 0.60 270 Viavo length In mu. *0 lg. 2?. - Absorption curves ol various fractions of eluate btained chromatographing tne alcohol soluble traction of ergosterol irradiated 40 minutes in p- xylene filtered ultra­ violet light. 1*3 The absence of the maximum for the isomer characterized by an absorp­ tion maximum at 260 mu. in fraction No. L indicates a low concentration of '.achysterol in the solutions and under such conditions should facilitate the separation of calciferol from the irradiation mixture. Several experiments were carried out with the purpose of isolating calciferol in the pure state. The absorption curves of the calciferol like material resulting from such experiments are shown in Figure 2L. In each case the maximum lies at 266 mu. and the crystalline material possess­ es tne physical characteristics as listed in Table II. TApLE II PHYSICAL PHOPruiTIES OF THn PREPARED CALCIFEROL Experi­ ment .-uir.oer HP. E(,l$t> ,1cm) at 266 mu. Color Activation Source Time of Irradi­ ation 30 92-117 237 White Source filtered oy p-xylene 16 Min. 31 109-117 3 61 White Source filtered oy p—xylene 16 Kin. 30—Lt 115-117 33d White Source filtered dv p-xylene 16 K i n . u7 115-11o 323 White Source filtered by p-xylene 16 Kin. 51 116-11c 371 White Unfiltered mercury source 16 Kin. ;2 115-116 303 White Unfiltered mercury source 16 Kin. Tne melting point of pur e calciferol is lli- 117 ° C . (.26) with an E ;lh, I c m .) of L6L (.15) . The melting points of the prepared compound are in good agreement with the literature values. However, the extinction co­ efficient shows considerable variations from the accepted values. To further characterize the interfering material in each case, a correction was calculated for the calciferol content and applied to each absorption curve . The corrected absorption curves are shown in Figure 25. The curves indicate that the interfering compound in each case has an aosorption maximum at 272 mu. trailing off to a plateau at 260 mu. Furthermore, the presence of this compound does not depend on the irradi­ ation history of the preparation for it is found in all preparations prepared both in unfiltered mercury light and p-xylene filtered light. An attempt was made to separate the two materials on alumina columns of "Type IIW using "Procedure D," the results of which are shown in Figure 27. This separation shows some promise since it appears freon the absorp­ tion spectra that partial separation was achieved and with the proper adjustment of the developer solution with regard to polarity complete separation appears possible . The effect cf oxygen on the irradiation products of ergosterol in the production of calciferol oy irradiation has been discussed by m.ary investigators (t,lfc,2U,27) and in each case they have concluded that the aenrimental effect of ooygen in the process lies in the formation of oxides of the isomers other tlian calciferol, forming a mixture from which tne vitamin is difficult to crystallize. Evidence indicates that the con­ centration of calciferol in the preparations was the same whether the irradiation mixtures were protected from the atmosphere or not. Several experiments were carried out to determine the effect, if any, on the aosorption s p e c t r a when irradiations were carried out in the presence of atmospheric oxygen. The solutions were irradiated according to Procedure C . and cliromatographed on a column of Type I . Under these 8.00 4. 00 3 8.00 8.00 1.00 50 Vave length In mu. (ls , £ 4 . ,.l ^ jr^tlun curves of crystalline "calciferol" iso la tea i :• Irradiated ergosterol solutions by chromatographic sep.rations on "Superfiltero1" columns. Crystalline material Isolated in experiment number: Cl) , (2) Zl, (z) ?1 tnroug.i 45, (4) 47, (5) 51, C6) 52. 0.300 0.150 B4 0.075 Wave length I270 n pu. Fig. 25. - Absorption curves of tr.e impurity present In tne crystalline o terials of Fig. 24, obtained by correcting tne absorption c-.v-a of tnese materials for tneir calciferol content, (ly 3«y, (2) 31, (3) 31 through 4S, (4) 47, (5) 51, (6) 52. 0 .800 hxtinction 0.600 0.400 0.800 870 Wave length in m u . 890 Fig. 86. - Absorption eurvea o f various fractions o f eluate obta i n e d by chromatographing tne eluant, fraction £, o f Fig. 15 o n an alumina oolumn. Extinction 0 .900 0 .300 230 270 250 Wave length in m u 290 Fig. 21. - A b s o r p t i o n curves o f various fractions o f eluate ob t a i n e d by c h r o m atographing the e l u e n t t fraction 2, o f ^lg. 15. o n a n alumina column. U5 conditions, the rate of flow \&s nearly uniform and the thickness of the cell such that each molecule received a uniform exposure. It follows then that the time of exposure can be calculated for each molecule in solution, i.e. in seconds of irradiation per molecule. Thus, if lOO ml. of solution require lO minutes to pass through the cell and the volume capacity of the cell being .29 ml. the cell must then be filled 3U U .8 times in lO minutes or 600 seconds. Each molecule is irradiated &00/3kh.8 or 1.72 seconds. The time of irradiation for each experiment conducted Figure 28 to Figure 33 is given in Table I I I . TABLE III IRRADIATION TIME OF ERGOSTEROL SOLUTION IRRADIATED In THE PRESENCE OF ATMOSPHERIC OXYGEN Figure Number 28 29 30 31 32 Time Seconds /ilolecule 3.5 8.1 11. h 12.2 22.8 For periods of short irradiation as illustrated in Figure 28, it was found that the first fractions to pass from the column were calciferol like curves, for example, fraction No. 16 possesses a true calciferol curve with an absorption maximum at 26$ mu. , but with slight distortions at 276 mu. and at 268 mu. In fraction No. 17, the same type of curve re­ sults, but with an additional inflection in the curve at 290 mu. These Extinction Wa vel e ngt h in mu. 40 fcO 120 80 40 /r 220 260 300 Wov«-l*ngth 2 9. 3*0 in m u . 380 Extinction he observations follow through all fractions up to and including fraction n o . 20. Fraction No. 21 in addition to its decrease in the over-all ex­ tinction values also exhibits a rise in the relative extinction at 28h mu. ■which is distinguishable in all fractions up to fraction No. 29. Fraction No. 22 possesses two maxima, one at 268 mu. and another at 275 mu. with a plateau extending to 290 mu. The maximum at 268 mu. disappears by fraction No. 2h leaving a major maximum at 278 mu. No. 29 are of the Fraction No. 26 to Fraction same general shape with a maximum at 2bh mu. However, starting with fraction No. 30, one major band passed through the column i.e. the irradiated mixture on nSuperfilterolM divides itself up into major bands , the first oand to appear in the eluate is the activated ergosterol group, calciferol, tacliysterol and other materials, and the second major banc is the ergosterol like material which is similarly subdivided into minor bands . Fraction N o . 31 through fraction N o . 70 consists of the ergosterol and ergosterol like materials partially separated on the column. Fraction No. 33 exhibits a large maximum at 252 mu. the same maximum which is present in all fractions tlmough fraction No. 37 varying in the absorption intensity throughout the five fractions and decreasing in intensity with the build up of new maxima at 2cl and 293 mu., the char­ acteristic absorption maxima of ergosterol. With each subsequent fraction the absorption at 252 mu. decreases until the characteristic ergosterol curve emerges in fraction N o . u 2 . no change is ocserved in the ergosterol traction through fraction No. 70 except for the decrease in overall intensity. In all experiments carried out, this same general trend was observed for irradiated solutions receiving irradiation up to 11.h seconds per hi molecule. However, as the time of irradiation increased, the effective­ ness of the column to separate out individual components decreased. For solutions having long irradiation periods , the calciferol fraction was not resolved, but only showed enrichment in the earlier fractions, for example, in Figure 30, the first fraction of eluate did not yield a pure calciferol curve as can be seen in fraction Wo. 13 which possesses a maxi­ mum at 268 mu. and another at 288 mu. The interfering material in each case possesses a maximum at 275 mu. which slowly moves towards the longer wave lengths by the introduction of a new compound having a maximum at 2814. mu. resolved for the first time in fraction wo. 19. The remaining frac­ tions through fraction No. 28 are characterized by the decrease in the maximum at 275 mu. with an increase at 2 8h mu. becoming more predominant. The remaining fractions exhibit the same trends previously outlined under short time irradiation. Many differences in absorption characteristics appear in tlie solu­ tions irradiated in the presence of atmospheric oxygen which were not evident in solutions irradiated in the absence of a i r . The differences are manifested by sharp absorption maxima at 275 mu., 28U mu., and 293 mu. Since the only evidence is the absorption spectrum, it is not possible to further characterize these compounds. -U8 GENERAL DISCUSSION "Superfilterol" chromatographic columns were effective in the resolu­ tion of many of the irradiation products of ergosterol into distinct bands. However, in each case the bands exhibited overlapping and were not completely isolated. To accomplish this it would first be necessary to select one given isomer ana adjust the starting solution for concen­ tration, and correct the developer to the proper polarity before isolation could be effected . Landenberg, et al. (16) studied the separation of sterols on chromato­ graphic columns and from these studies pointed out the necessity for a difference to exist in the bond structure of the molecules before separa­ tion could be perfected. Since ergosterol and lumisterol have bond structures in ring nsn and possess the same number of double bonds it would be expected that these compounds would defy separation. Tachysterol and calciferol also possess the same number of double bonds with but a slight difference appearing in their location. For example, tachysterol iias unsaturation at the 6*7, 6*9, and the 5*10 positions while calciferol nas double bonds in the 5*6, 7*6, and 10*19 positions. However, resolu­ tion was obtained between taciiysterol and calciferol, while lumisterol was not detected in any of the chromatographic fractions . The extent of the resolution appeared to be a function of the concentration of the different components present. Since the starting solution of ergosterol was constant for all experiments, the only variable was the time of irradiation. With the increased time of irradiation or as the wave length h9 of the radiation of activation was varied, the relative concentrations of each component in the irradiated solution varied exhibiting great influence on the effectiveness of the separation. It has been shown ttiat either long periods of irradiation or the use of unfiltered mercury radiation decreases the effectiveness of the column which may be attributed to the rapid build-up of the tacliysterol in the irradiated solution and other Dreakdown products resulting in an effective flooding of the column with these compounds and contamination of other component bands. To study the effect other isomers have on the calciferol fraction, 0.0209 g. of pure calciferol were passed over a "Superfilterolw column of type I under the conditions of the experiment in (50-10-1) developer (50 volumes of Skelly Solvent, 10 volumes of diethyl ether, one volume of ethanol). Figure 33 shows the results of the study of the rate of flow when unhindered oy the presence of like materials. It was found that calciferol passed through the column unchanged in 22 ml. of eluate with the maximum concentration found in fraction ho. 2 h . However, in the presence of other isomers of the irradiation mixture, the calciferol frac­ tion is usually found to appear in the lUth fraction ana extending through higher fractions being overlapped by taclysterol . Another factor influencing the rate at which calciferol passes through the column is the presence of the compound whose absorption spectrum is shown in Figure 25. If it possesses similar adsorption char­ acteristics it would have the effect of spreading out the calciferol band by increasing the concentration of the absorbate, thus increasing the adsorption on the column and requiring more developer to prefect the 240 120 9 CM 20C 80 o I. A I III V I11/ II 160 40 120 80 Tf 40 io 3o 3/ 220 2 80 300 Wo v « - l e n g t h g. 3 3 340 in mu. 3 80 So elution resulting in a spreading of the band. If the combined materials exceeded the capacity of the column, the calciferol fraction would appear more quickly in the eluate and would ultimately result in a diffused b a n d . The foreign material which appears in the calciferol fraction defies separation on the "Superfilterol" column. Varying concentrations of these materials have been recliromatographed over several fresh columns without prefecting a separation. Since this material possesses an absorp­ tion spectrum with a maximum at 212 mu., and 260 mu. the probability of it having a bond structure similar to calciferol is remote. Yet, because of its adsorption characteristics it appears such should be the case. If the two materials isolated were in combination such as found in the formation of vitamin D x i.e. a molecular* compound between lumisterol and calciferol the behavior of the two materials on the column would be the same. Lumisterol possesses a similar absorption spectrum to that of the interfering compound except that the accepted values of the maxima of lumisterol are located at wave lengths 265 mu. and 260 mu. (25 ) however, ■one data is in need of clarification. This observation follows those of Powell (20) who observed that the vitamin fraction separated on a ”Superfilterol" column using (50 -10 -1 ) developer, 50 volumes of Skelly Solvent, 10 volumes diehtyl ether , one volume of ethanol, nad an E (1 cm, 1^) of one-half that accepted for pure calciferol, but attributed the discrepancy to the presence of a com­ pound possessing an absorption band in the visible range. The first fraction of eluate of the second major group of adsorption oands eluted from the 11Superfilterol" columns were characterized by a 51 large absorption band at 252 mu. and a smaller maximum at 28h mu. which appeared to increase in intensity during the early minutes of irradiation, but decrease rapidly on longer periods of irradiation. Kimball (.1?) in­ vestigated the action of ergosterol on "Superfilterol11 columns in the presence of (50-10-1) developer, 50 volumes of Skelly Solvent, 10 volumes of diethyl ether, one volume of ethanol, and found that ergosterol which had been recrystallized from an alcohol mixture and from benzene-ethanol solutions (lii) on passing over such columns, the first fractions to be eluted possessed two maxima one at 252 mu. and one at 28h mu. Kimball collected the fractions of eluate which possessed the pure ergosterol absorption curves and repassed the fractions over new MSuperfilterol1' only to find the first fractions of eluate to asrain iiave absorption maxima at 252 and 2th mu. From this she concluded that a conversion or rearrange­ ment was taking place in the presence of the strong acid adsorbent;, no attempt was made to crystallize or isolate this material. It might bo worthy of note to point out the formation of a white crystalline precipitate in the eluate when such fractions are allowed to stand in the developer solution over night at -20°C. This precipitate was located in those fractions which were cliromatographically enriched in the tachysterol fraction and in the fractions which consisted primarily of the calciferol mixtures. The precipitate is insoluble in alcohol and hexane and only slightly soluble in ether. The absorption spectra were similar to the spectrum of the compound Green (13) called Suprasterol lit. 52 SUMMARY (1) Irradiated ergosterol solutions have been chromatographed on "Superfilterol" and alumina columns. Separation of ergosterol, a compound having maxima at 252 mu. and 288 mu., calciferol, and tachysterol were achieved. The calciferol portion experienced interference from a com­ pound having maxima at 272 mu. and 280 mu. The presence of this compound in the crystalline calciferol fraction is independent of the irradiation history of the irradiated solutions. Alumina used as the adsorbent is more effective in the separation of the calciferol and "compound 27 2 m tiian "Superfilterol." (2) The effect of concentration of irradiation intermediates and photochemical end products on the separation of calciferol from solution have been studied. It has oeen found that either extensive irradiation or the presence of short wave length radiation results in inferior separa­ tions . This is due primarily to the build up in either case of large concentrations of tachysterol in the irradiated solution giving an un­ favorable tacliysterol to calciferol ratio resulting in extensive over­ lapping of the two isomer bands . (3) The effect of atmospheric oxygen on the irradiation products of ergosterol has been studied and the solutions resulting from such irradi­ ations separated on "Superfilterol" columns . It was found that compounds having maxima at 275 mu., 288 mu., and 293 mu. were present under these conditions of irradiation which did not manifest themselves in solution protected from the atmosphere during irradiation. 53 LITERATURE CITED 1. T. C. Angus, F. A. Askew, R. B. Bourdillon, H. M. Bruce, R. K. Callow, C. Fischermann, J. S. T. Philpot, and T. A. Webster, P r o c . R o y l . S o c . (London), 1 0 6 B , (1931). 2. F. A. Askew, R. B. Bourdillon, H. M. Bruce, and R. K. Callow, Proc. Royl. Soc. (London), 109B, (1931). 3. D. H. Baker, Michigan State College, M. S. Thesis (1983). 8. C. E. Bills, E. M. Honeywell, and W. M. Cox, J. Biol. Chem., 80, 557 (1928) . 5. h. Brockmann and A. Busse, Z. Physiol. Chem., 2 5 6 , 252 (1938). 6. F. P. Browden and C. P. Snow, Nature, 129, 720 (1932). 7. L. J. Bullard, Michigan State College, M. S. Thesis (191*5). 8. H. G. Cassidy, J. Am. Chem. Soc., 63, 2628 (1981). 9. 10. F. H. Chen, Michigan State College, M. S. Thesis (1951). J. B. DeWitt, and M. X. Sullivan, ind.Eng. Chem. Anal. Ed.,18, (1986). 117 11. D. T. Ewing, C. V. Kingsley, R. A. Brown, and A. D. Emmett, Ind. Eng. Chem. H.nal. Ed., 15., 301 (1983). 12. D. T. Ewing, K. J. Powell, R. A. Brown, and A. D. Emmett, Ind. Eng. Chem. Anal. Ed., 2_2, (1950). 13. «J.Green, Biochem. «J . , 89, 232 (1951). 18. J. W. Hage , Michigan State College, M. S. Thesis (1983). 15. W. Huber, G. W. Ewing, and J. Kreiger,