880
COMMUNICATIONS TO THE EDITOR
theory of DNA structure12 is provided by the constancy of molecular weight throughout the structural changes which we have described. In addition, however, we are led to wonder whether the observed phenomena do not point to the participation of water in stabilizing the structure of DNA. With this end in view, we are investigating the behavior of DNA in a variety of organic solvents with different hydrogen-bonding characteristics.
Vol. 78
refining, x is the distance that the zone has traveled along the tube, and 1 is the zone length) which indicated that certain impurities can be efficiently removed by this technique. In the case of boron where the concentration level was below the limit of spectrographic detectability, a mathematical extrapolation was employed based on the minimum spectrographically detectable concentration found. Using this value in conjunction with the expression :
cz
(12) C A. Dekker and H K Schachman, P r o c N o t Acad Scr , 40,
=
K[Co
+ (C,-l/K
-
cz-dl
where C, is the concentration of an impurity frozen 89 (1954). into the zth zone, C,-l is the concentration of the STERLING CHEMISTRY LABORATORY impurity frozen into the previously frozen zone, CONTRIBUTION 1346 YALEUNIVERSITY E P GEIDUSCHEKk is the segregation coefficient of the impurity, NEW H A V E N , CONNECTICUT and COis as described above, the maximum value foI the k of boron was determined. The values of the DEPARTMENT OF BIOCHEMISTRY ARMYMEDICAL SERVICE GRADUATE SCHOOLIRVICG GRAY k's determined for various metallic impurity species WASHINGTON 12, D C. in the Si14 matrix are: boron, 0.16 f 0.07, alumiRECEIVED DECEMBER 28, 1955 num, 0.88 =k 0.04, sodium, 0.07 0.01, magnesium, 0.58 0.06, copper 0.63 f 0.05. I t is apparent from the above that the impurities SEGREGATION COEFFICIENTS OF VARIOUS IMPURITIES I N A SILICON TETRAIODIDE MATRIX listed can be removed to levels below the one part per million range by a suitable number of passes. Sir: Work on this project, and the method and/or The demand for transistors and other semi- methods of preparing pure silicon tetraiodide, as conductor devices has stimulated considerable well as the preparation of the elemental silicon, is research in the preparation of ultra-pure silicon. continuing. It was found that zone-refining techniques' were CAMBRIDGE RESEARCH CENTER inadequate for the removal of certain impurities, AIR FORCE ELECTRONICS RESEARCH DIRECTORATE e.g., boron, in a silicon matrix,2 and it was decided LAURENCE BERNARD RUBIN G. HANSCOM FIELD to approach the problem by preparing a suitable BEDFORD,MASSACHUSETTS GUYH. MOATES compound of silicon, subjecting this to such puriRECEIVED JANUARY 9, 1956 fication techniques as recrystallization, sublimation, and zone purification and ultimately decomposing i t to elemental silicon. Thermodynamic GLUCOSIDURONIC ACID SYNTHESIS BY 8-GLUCUcalculations indicated that of the four tetrahalides, RONIDASE IN A TRANSFER REACTION silicon tetraiodide decomposed most readily and Sir: that i t lent itself best to zone-melting techniques Certain biological phenomena have been correbecause of its relatively high melting point (121.5- lated with the activity of the enzyme p-glucuroni122.5'). Furthermore, i t could be expected that dase. Among these are glucuronidogenesis,l action the segregation coefficients for the various impuri- of gonadal hormone^,^^^^^^^ human c a n ~ e r , ~ , ~ ~ ~ ties in silicon tetraiodide would differ from those genetic control in mouse t i s s ~ e s ,and ~ * effects ~ ~ ~ ~of found in silicon. pituitary interstitial cell stimulating hormone." Silicon tetraiodide was prepared by passing In attempts to arrive a t an interpretation of the iodine vapor over Coleman and Bell Company function of the enzyme in vivo, we have found it ninety-eight per cent. pure silicon a t 800" and the difficult to explain the findings on the basis of product was then crystallized from toluene. This a purely hydrolytic action of the enzyme or its material was used to fill Pyrex ampoules nine milli- simple reversal. It was postulated that p-glumeters in diameter and thirty centimeters long. curonidase participates as a member of a multiIt was then densified, the tube sealed, and zone- component system concerned with glucosiduronic purification was effected by vertical passage a t the (1) W . H . Fishman, J . Bioi.Chem., 136, 229 (1940). rate of five centimeters an hour. The zone width (2) W. H . Fishman, ibid., 159, 7 (1947). was two and one-half centimeters, and the tempera(3) E'. H . Fishman and Y I H . Farmelant, E n d o c v i d o g y , 5 2 , 530 ture of the molten zone was about 135". h small (1953). (4) A. L. Beyler a n d C. M .Szego, i b i d . , S 4 , 3 2 3 , 3 3 4 (1954). molten zone was passed through the charge only ( 5 ) W. H . Fishman, i n "Vitamins a n d Hormones," Vol. I X , Acaonce in order to maintain impurities a t spectro- demic Press, X e w York, N. Y., 1951, p. 213. graphically detectable levels. Spectrographic an(6) W. H . Fishman, A. J . Anlyan a n d E . Gordon, Cancer Research, 7 , alyses of the successive two and one-half centi- 808 (1947). ( 7 ) W. H. Fishman and R. Bigelow, J . S a t i . Cancer I n s t . , 10, 1115 meter zones permitted preliminary calculations to be made based on Pfann's original equation.' The (1950). ( 8 ) W. H. Fishman, S . Green, F. Homburger, S. C. Kasdon, H. E. results of these calculations gave plots of C/CO Nieburgs, G . McInnis and E . R . P u n d , Cancer, 7 , 729 (1954). versus x / l (where C is the concentration of an (9) A. G . Morrow, E. hl. Greenspan and D. M. Carroll, J . Nall. impurity in a solid frozen from a mother zone, Co is Cancer I n s f . ,10, 657 (1949). L. W. L a w , A. G. Morrow and E. M. Greenspan, i b i d , 12, the mean concentration of the impurity before zone 900(10) (1952).
*
(1) W.G. Pfann, Trans. AIME, J. Metals, 194, 747 (July, 1962). (2) J. A. Burton, Physica, 90, 845 (Nov., 1864).
*
(11) W . H. Fishman, G. Benjamin and S . Green, J. Clin. Endocrincl., 16, 876 (1966).
