COMMUNICATIONS TO THE EDITOR
3724
Vol. 83
+
ing. of I1 gave these fractions: (l), 86 mg. of B A ~ cwhere , [ h f ] A c N e +43,300 is the molecular neosamine C, dihydrochloride, [ C X ] * ~ +65’ D (c 1.0, rotation of 11, [ M ] A c D e is the rotational contribution water) [w3[CrIz3D +69’ (c 0.87, water)], Rf of the 4-substituted-N,Nf-diacetyldeoxystrep(BAW 221) 0.17 (lit.I2 0.17); (2), an intermediate tamine moiety, and A A ~ and C B A ~have C the usual fraction, whose components were identified by meanings’’ for a glycoside of N,N’-diacetylneospapergrams and ion exchange,“ and which was amine, C, yields A A ~ = C +24,000 or +31,800 (decalculated from rotation to contain 145 mg. of pending on the sign of [ M ] A c D e ) . 2 0 I n either case neosamine C and 72 mg. of deoxystreptamine hy- the high positive value agrees well with that exdrochloride; (3), 160 mg. of optically inactive deoxy- pected for an a-linked glycoside (as shown i n streptamine hydrochloride; (4), 430 mg. of re- I).18t20’This conclusion may also Le reached, covered neamine, hydrochloride [CY]“D +82’ ( c 1.0, though less quantitatively, from the high molecular water) [lit.2 [ c Y ] ~ ~+83’ D (c 1.0, water). rotation of neamine hydrochloride ( [M]D+38,800)‘ Conclusive evidence for the identity of the relative to the corresponding value for neosamine hydrolysis product and neosamine C was provided c hydrochloride ( [ M I D +16,800).16 Except for the by the N-a~etylation’~ of the former to N,X‘- assignment of absolute stereochemistry to the subdiacetylneosamine C, m.p. 209-221 O (undepressed stituted deoxystreptamine (;.e., 4- ns. 6-substituon mixture with an authentic sample’j melting 210- tion), the present data establish the structure for 214’), [CZ]~’D +38” (c, 0.6, water) [w5 [cY]~’D neamine. 4-37” (c 0.8, water), RNAG(BAW 415) 1.35 (lit.I5 Acknowledgment.-This investigation was 1.32). Neosamine was earlier assigned D-glUCOSe supported in part by research grants, No. E-61s stereochemistry16 from degradative and rotation and KO. E-1278, from the National Institute of evidence; the assignment has been confirmed re- Allergy and Infectious Diseases, Public Health cently by comparison of the diarninohexose with Service, and by a Fellowship from the Upjohn Conlsynthetic 2,6-diamino-2,6-dideoxy-~-glucose.l5 pany. We also wish to express our thanks to the Both N,N’,N”,N”’-tetraacetylneamine(11) and Upjohii Company for generous quantities of neoN,N’,N”,N”’-tetrabenzoylneamine (111) consume mycin sanlples. selectively two moles of periodate when treated a t (20) Recent work elsewhere (Dr. C . P. Schaffner, personal communiroom temperature with 0.1 N periodate. Since a cation) has showed t h a t t h e 4-O-rnethyl-N,N’-diacetyldeoxystrepobtained by N-acetylation, 0-methylation, and hydrolysis o f maximum of one mole of periodate may be con- tatuine neomycin B has (.he same rotation (positive) a s t h e same compounil sumed in vic-glycol cleavage of the deoxystrep- from t h e analogous sequence on paromomycin.‘¶ Hence, [?d]AoDe is tamine moiety and a like amount in the neosamine -8,900 and A A ~ Cis +31,800. C moiety in I1 and 111, these results establish a 4- DEPARTMENT OF CHEMISTRY H E R B E R T E. C A R T h R J O H N R. DYER AND CHEMICAL E S G I N E E R I N G (rather than 5-)glycoside linkage on deoxystrepPAULD. SHAW OF ILLINOIS tamine and a pyranose (rather than furanose) struc- UXIVERSITY KEXNETH L. RINEHART, JK. ~JRBANA, I L I ~ O I S ture for neosamine C in neamine. Confirmation of MARTISHICHENS the attachment of neosamine C a t C-4 in deoxyi m E I V p : t ) .rIiL\; ;;I, 1961 streptamine (providing a vzc glycol for cleavage) is provided by the observation that no deoxystrep- A STUDY OF THE REACTIONS OF VARIOUS TIN(I1) tamine may be detected after the two-molar COMPOUNDS WITH CALCIUM HYDROXYLAPATITE periodate oxidation of IT, with subsequent hydrol- Sir, ysis with refluxing 48y0 hydrobromic acid (cf. We wish to report the results of a series of iuabove). vestigations carried out in aqueous media involving From the rotation of methyl N,N’-diacetyl-a- the reaction between calcium hydroxylapatite, neosaminide C,I6 [MID +32,800, and the value of Ca10(P04)G(OH)2, and these tin(I1) salts: Snl’?, 1 G for methyl N-acetyl-~-glucosnininides, SnClZ.2Hz0, SnS04, SnClF, and SnzCIFa. The +17,390, one may derive by Hudson’s rules1*the reactions led in each case, except for that involving value B A ~ C15,400 for N,N’-diacetylneosaminides SnC12.2H20, t o the formation of a basic tin(I1) C, while the rotation of a 4-substituted N,N’- phosphate having a molar Sn/P04 ratio of 2.0, diacetyldeoxystreptamine is known from [M]~I,, to which we have assigned the empirical formula +3,900 for 4-O-methyl-N,N’-diacetyldeoxystrepSn4(P04)2(OH)z.Hz0 (calculated : Sn, 66.2; PO4, tamine.lg Combination of these values and the 26.3. Found: Sn,63.8; PO4, 26.5). The reaction iiiolecular rotation of N-tetraacetylneamine (11) involving SnCl2,2H20,however, resulted in tlie into the equation [ h f ] A c X e = [ATlAcDe AACC formation of Sn3(P04)2. Experimental.-Calcium hydroxylapatite for (13) K . L. Rinehart, Jr., and P. W. K. Woo, J . Am. Chem. Soc., 83, ( i 1 3 (1001). use in these studies was prepared according to the (14) S. Roseman and J. Ludowieg, i h i d . , 7 6 , 301 (1034). method of Hayek and Stad1mann.l Stannous (15) K. L. Rinehart, Jr., M Hichens, K . Striegler, K. R. Rover, fluoride was obtained by treating SnO with HF.2 T. P. Culbertson, S. Tatsuoka, S. Horii, T. Yamaguchi, H . Hitomi a n d A. Misake. ibid.. 83. 2964 (1961). The mixed chlorofluorides were prepared accord(10) K . L. Rinehart, Jr., P. W. K . Woo a n d A. D . ArEoudelis, ihid., ing to procedures previously outlined in this 80. G4G1 (1958). . . lab0ratory.3.~ Commercial preparations of both (17) P. W. Kent and hl. 1%’. Whitehouse, “Biochemistry of the SnC1~2H20and SnS04were employed. Aminosugars.” Eutterworths Scientific Publications, London, 1R55,
+
+
p. 234. (18) C. S. Hudson. J . A m . C h e m . Soc., Si, G O (1909); cf. W. Pigman, in “The Carbohydrates,” Academic Press, Inc.. New York, N. Y., I D j i ‘ , p . 70. (19) T. H. Hnskell, J. C. French and I?. R. Bartz, J . A m . Chcm. SOC. 81, 3482 (1959).
(1) E. Hayek and W. Stadlmann, Anpew. Chem., 67, 327 (1955). (2) W. H. Nebergall, J. C. Muhler and H . G. Day. J . A m . Chrri?. S o c . , 74, 1I;O-l (19.52). (3) W.11. Nebergall, G . Bii‘,cjgic~;ind J . C . >Iiihlcr, i b i d . , 7 6 , 53X3 (1954).
(4) W. H. Nebergall, U. S. Patent 2,882,204.
Sept. 5 , l O G 1
COMXPNIC.\TIONS TO THE
The reactions were studied in molar proportions such that there existed initially a 1:1 ratio of Ca to Sn atoms. The amounts of reactants and the volume of water used in each case were adjusted such that the initial concentration of tin(I1) was in the range 0.03-0.05 M for each reaction studied. A solid phase persisted throughout the duration of each reaction. Initially this solid was simply calcium hydroxylapatite, then a mixture of partially reacted materials, and finally the completed reaction products. The reactions were carried out a t reflux temperature in an inert atmosphere of nitrogen gas from which all traces of oxygen had been removed by bubbling the gas through a train of wash bottles containing amalgamated zinc and solutions of vanadium(I1) ions6 This precaution also was employed during filtration of the reaction mixtures in order to prevent any air oxidation of the tin(I1) ion. All products were dried either in vacuo over P4010 or by use of a conventional drying pistol. Identifications were made by means of X-ray diffraction powder patterns when possible, along with chemical analyses. il modification of the Martin-Doty method6 was used for the colorimetric determination of phosphate; tin(I1) was determined by an accurate procedure which we devised in this laboratory and which soon will be published elsewhere. Results and Discussion.-A summary of the reactions studied is given in Table I. TABLE I PRODUCTS OBTAINED BY AQUEOUS TREATMENT OF Caio(PO4)e(OH)zWITH TIN(II) SALTS Tin(1I) salt used
Reaction time, hr.
Final p H of soh.
SnF2
48
3.00
SnFz
72
2.85
SnFz SnSO4
96 48
2.75 2.40
SnClF SnzClFa SnClz2H20
48 60 48
2.65 2.80 2.50
Insoluble products
C ~ I O ( P O ~ ) ~ ( OSn4(POdtH)Z, (OH)z.H*O,CaFz C~I~(PO~)~(O sn~(PO4)zH)Z, (OH)2.H20, CaFz Sn4(P0&(OH)z.Hz0, CaFz Sn4(PO4)~(OH)~.HzO, CaS042H20 Sn4(PO&(OH)~.H20, CaFz Sn4(P0&(0H)2.H20,CaFz Sn3(P04)2
The PH of these solutions was affected both by the hydrolysis of the tin(I1) ion as represented by the equation Sn+2 f HzO -+ SnOHf Hf, and by the relative basicities of the individual anions. The 1:l ratio of Sn:Ca was employed on the basis of the assumption that a simple cationic exchange would take place and lead to the formation of tin(11) hydroxylapatite, Snlo(PO4)6 (OH)2. However, the absence of a tin(I1) phosphate with molar Sn/P04 ratio of 1.67 among any of the products eliminated this possibility. In order to identify the basic tin(I1) phosphate, which was formed in all but one of the reactions studied, a sample of it had to be separated from the accompanying insoluble by-products present in each of the mixtures obtained. This was necessary because X-ray diffraction studies of
+
( 5 ) L. Meites and T. Meites, Anal. Chem., 20, 984 (1948). (61 J. B. .\Ia.tin and D. X. Doty, ibid., 21, O C 5 (1949).
EDITOR
3725
these various mixtures revealed that the pattern of the basic tin(I1) phosphate did not agree with any published for known phosphate compounds of tin(I1) and hence it could not be identified simply by comparison. All attempts a t isolating this material from mixtures containing the very insoluble CaF2 were unsuccessful. In strong acid solutions CaFz exhibits sufficient solubility to permit its removal, but such conditions also cause a change in the structure of the basic tin(I1) salt. Attempted extractions of the calcium with chelating agents such as EDTA were also inefficient under the mildly acidic conditions required to prevent alteration of the structure of Sn4(P04)2(0H)2. H 2 0 . However, isolation of this compound was effected by treating the mixture containing CaS04. 2Hz0 with a buffer solution of p H 4; gypsum is readily soluble under these conditions even a t room temperature and the basic tin(I1) phosphate remains unchanged. Later, in succeeding studies, Sn4(P04)2(OH)z. H20 also was obtained directly as a hydrolysis product of SnHP04. The complete results of our investigations regarding the hydrolysis of SnHP04 will be discussed in a later publication. The assignment of the empirical formula Sn4(PO~)Z(OH)~.H*O to this basic salt was based on its composition as determined by chemical analyses. The same procedure was used for the identification of Sn3(P04)2 as the product obtained from the reaction between Calo(P04)~(OH)2and SnC12.2HzO. The X-ray diffraction pattern of our tribasic tin(I1) phosphate did not agree with that published for this compound in the ASTM X-ray card file. Finally, we have found that SnF2, SnzP207, and Calo(P04)~(OH)zin the molar ratio of 10:1:1, respectively, react to form Sn3(P0& and CaFz. DEPARTMENT O F CHEMISTRY RONALD COLLINS IXDIANA UNIVERSITY WILLIAMNEBERCALL INDIANA HORSTLANCER BLOOMINGTON, JULY 10, 1961 RECEIVED THE REDUCTION OF OLEFINS BY MEANS OF AZODICARBOXYLIC ACID in situ Sir :
Certain considerations1 led us to the belief that the readily available2 salts of azodicarboxylic acid might be a source of HzNz (diimide, diazene), which, although a t present a poorly defined species,a should act in situ as a potent reducing agent. We have now found that the aforementioned salts do (1) Chemical reductions of olefins bv means of hydrazine (c.g., F. Aylward and M. Sawistowska, Chem. a n d I n d . , 404 (1961), and previous papers) has been shown t o be oxygen dependent ( i b i d . , 433 (1961) and by independent observation in this Laboratory). Also, in the acidification of potassium azodicarboxylate (J. Thiele, Ann., 271. 127 (1892)) nitrogen and hydrazine are formed in addition t o carbon dioxide. For this decomposition Thiele suggested diimide as a n intermediate which would disproportionate into nitrogen and hydrazine. In the hydrazine reductions i t was considered possible t h a t the reaction with olefins was preceded by air oxidation t o diimide, the active reducing agent and the possible irtermediate in the azodicarboxylic acid decomposition. ( 2 ) Prepared by hydrolysis of azodicarboxamide (Thiele, ref. l), commercially available from Aldrich Chemical Company, Milwaukee, Wisconsin (3) (a) L. F. Audrieth and B. A. Ogg, “ T h e Chemistry of Hy1951, Chapter drazine,” John Wiley and Sons, Inc., New York, N. Y., 6; (b) S. N. Foner and X. L. Hudson, J . Chem. P h y s . , 28, 719 (1958); (c) D. A. Dows, G. C. Pimentel and E. Whittle, ibid., 23, 1606 (1955).
