without further purification, was converted into the dibromo compound by shaking for 20 hours with a solution of 2.74 g. of bromine in 20 ml. of carbon tetrachloride. The reaction mixture was washcd with aqueous sulfite and then with water, and finally filtered and evaporated to dryness. The residue was taken up in petroleum ether (b.p. 35-60”) and chromatographed on a 40-g. alumina column, using more solvent as eluent. A single solid fraction consisting of 0.706 g. (18.6’%, based 0 1 1 sulfonyl chloride), m.p. 55.0-56.5”, was isolated. Recrystallization from methanol yielded 0.409 g. (11%) of light yellow crystals, m.p. 57.0-57.5”. A mixed melting point with a n authentic specimen of 4-nitro-2,Rclibromotoluene was not depressed, and the infrared ahsorption spectra of the two samples were identical. 4-Nitro-2,6-diiodotoluene.--Five grams (0.015 mole) of 4-nitrotoluene-2,6-disulfonyl chloride was converted to the Iischloromercuri compound as described in the preceding sectioii. The mercury compound was stirred overnight with a suspension of 4.2 g. of iodine in carbon tetrachloride. Excess iodine was destroyed by washing the mixture with aqueous sodium sulfite. The organic layer was filtered and evaporated t o dryness, and the residue was taken up in a 1:4 mixture of benzene and petroleum ether (b.p. 35-60”) and chromatographed on alumina, using benzene-petroleum ether mixtures of increasing benzene content for elution. The only sizable fraction recovered was 0.293 g. (,5.070)of light yellow crystals, m.p. 116’. Anal. Calcd. for C ~ H J ~ N O ZC, : 21.61; €1, 1.80; I , 62.26. Found: C , 21.61; H , 1.84; I , 6 5 . 3 5 . PROCTER A N D GAMBLE Co. MIAMII‘ALI E Y LABORATORIES CIXCIXS.4TI :< 1, OHIO
Quantum Mechanical Calculations of Electrical Effects of Substituents in kuru-Substituted Anilines I3u
JOHZ
nitrogen in para-substituted anilines is, at least theoretically, dependent on the character of the substituent, calctilations wcrc iri:ide by the simple ~tiolecularorbital method’ of thc charges on the amino nitrogen in such compounds where the parnsubstituent is: (1 ) -X@bearing a unit positive charge (analogous to -N(CH:3):j@) ; ( 2 1 --Y carrying ari empty p-orbital (analogous to --RR!); and (;3) --C:=-Z where 2 (analogous to oxygen) is more electronegative than carbon and the polari7:Ltioii of
the -C-Z bond results in an incoln1)letely filled ],-orbital on carbon. The substituent -X\i’ is regarded as being formally uiinble to “conjugntc” , ~
with a ,bnrci-aniino group, whereas ----\aiid C -7, are both able to take part in such conjugation.
I --.I’ and -C=Z
differ from each other only i n the degree of unsaturation of the porbitals (on Y and C, respectively) and in the electron affinities of 1and 2. The condition was imposed t h a t systems IV, Lr arid VI have nearly the same charge ( e . g . , +0.05 of an electron charge*) in the rr-electrons a t C4 before introduction of a pura-amino group, so that the only important variable in the para-substituted ariilines is the electron-accepting power of the substittlellt.
D. ROBERTS A N D DOROTHY A SEMEVOW RECEIVEDOCTOBER 28, 1954
Rase strength measurements with substituted anilines indicate direct conjugation between the amino groups and electron-attracting unsaturated etc. (I). substituent groups such as --NOz, -C=N, Such conjugation delocalizes the unshared electron pair of nitrogen and confers a fornial (+) charge on the amino group. Somewhat analogous behavior might be anticipated for substituents which do not carry multiple bonds but which, like tercovalent l)oron, could accept anelectron pair (11)or,likeapositively charged trimethylammonium group, could 5tabilize an adjacent negatively charged center as 111 the “ylides” (111). Resonance forms may be written which symbolize each of these situations (1-111). SH2
@k(CH,),
6I+ CITJ)3 (
111
T n order to determine whether the extent of delocalization of the unshared pair of electrons on
This was achieved by assigning appropriate couSP, to atoms having electron lomb integrals, a affinities different from a “normal” carbon atom with coulomb integral a ; 6 > 0 for larger arid 6 < 0 for smaller electron affinities than a “normal” carbon atom. The 6 values used for atoms adjacent to the source of polarization3 ( i . e . ,C1 and Cs in IV, C1in V and C7in VI) were one-tenth the 6 values for the atom at the source ( i . e , ,CLin 1lr,I’ i n V and 2 i n VI) ; coulomb integrals for which 6 was less thaii 0.05 were set equal to a. Kesonance integrals between non-adjacent atoms were neglected and, in general, those between adjacent atoms were assigned a value of p. Coulson4 has stated that the deviation from P of resonance integrals involving
+
(1) (a) E. Hiickel, Z . Physik, TO, 204 (1931): “Grundziige d e r Theorie ungesattiger und aromatischer Verbindungen,” Verlag Chemie, Berlin, 1938, p p . 77-85; ( h ) C. A Couson a n d H. C. Longuet-Higgins. PYOC. R o y . Soc. (London), A l S l , 39 (1947); (c) H. Eyring, J . Walter a n d G. E.Kimbal1, “ Q u a n t u m Chemistry,” John Wiley and Sons, I n c , A’ew York, N. Y., 1944, C h a p XI11 (2) T h e charge a t Ca was chosen t o be equal t o 0 05 of a n elertron charge since t h e n a substance possessing t h e over-all charge distribution of V I would h a v e a resonance moment of 0 4 D which is of t h e s a m e order of magnitude as t h e experimental resonance moments, 0 3 a n d 0.6 D ,for benzaldehyde and benzonitrile, respectively. (3) G. W.Wheland a n d I. Panling, THISJOURNAL,67, 2086 (1935). ( 4 ) C . 4 . Coiilwm, ‘ ‘ \ ‘ : i I w ~ r ~Oxford ,’’ IJniversity Press. I . ~ i i i ~ I o n I % i 2 , p 242.
Julie 3, 1933
:i153
NOTES
hetero-atoms is small and can be shown4a5to exert only a second-order effect on charge distribution. Since the orbitals of the central atom of the -X@ group in IV (like N in -N(CH3)a@)are used in four sp3-bonds, the a-electron resonance integral for the C1-X@ bond was assumed to be zero. Thus, the only electrical effect considered for -X@was inductive generation of a positive charge on Cl, increasing the electron affinity of that atom and to a lesser extent the electron affinities of C2 and ( 2 6 . The values of the parameters used are summarized in Table I. The resulting r-electron charge distributions in the /)arts-ainino derivatives are shown in VII, VI11 and IX.
on Cq as postulated previously.' The failure of pamino- and p-dimethylaminophenyltrimethylainmonium salts to be weak bases relative to the corresponding m-isomers is thus regarded as strong experimental evidence against important preferential induction by the trimethylammonium group of charged centers in the 2- and 3 - p 0 s i t i o n s . ~ ~ > ~ (7) (a) J. D. Roberts, E. A. McElhill, R . A. Armstrong. 1 ~ 1 s JOURNAL,71, 408 (1960); (b) J . D Roberts, R . A . Clement a n d J J . Drysdale, ibid., 75, 2181 (1951). ( 8 ) A somewhat different interpretation recently has heeii exlxessed b y C. K. Ingold, "Structure a n d Mechanism in Organic Chemistry." , 234, i 3 2 . Cornell University Press, I t h a c a . S . Y . pp.
CONTRIBUTIOX S o . 1953 GATESA N D CRELLISLABORATORIES CALIFORNIA IXSTITUTEOF TECHNOLOGY PASADEXA 4, CALIFORNIA
Reaction of Fumaric Acid with Cysteine
m
-c>-
Q-Y +0.264 -0.031 f0.055
+0.106
BY LEOPISE ASD CARLL. PEACOCK RECEIVED FEBRUARY 7, 1955
,--,-0.0391fJ=z
+0.046
H2N
rn -0.424
Ix TABLEI ~ ( I I J L O MINTEGRALS ~ U S E D Ih' CALCULAI'ICJSS
System
I \'
alld
\.I1
\. atid \-I11
\.I alld
1s
\ . I l , \.I11
,llltl
Coulomb integral
CI
=
Y
= =
c 1
=
cp
+ 0.3Op c, = + (I.U$ CY
cy
- 1.40p 01 - 0.14/3
CY
Z =
cy
c,
cy
IS s
= =
CY
c1 =
CY
+ 0.80p + 0.088 + 2.008" + 0.20p
1\--1x
Carbon atoms not otherwise tlesignated = CY Thc w l u c chosen for the coulomb integral of nitrogen 20 conimonly is used, alis not critical; the value of CY though it is probably not the best value for n i t r ~ g e n . ~
+
The calculated charges on the amino nitrogens for VI I-IX are essentially the same.6 Furthermore, the magnitudes of the calculated resonance energies for IrII-IX and the corresponding anilinium ions indicate t h a t compounds with charge distributions like VII-IX should have practically equal base
I
are calcustrengths. Thus, -X@, --I7or -C=Z lated to be equally effective in causing delocalization of para-amino electrons provided that C4 initially has the same charge in each case. It is concluded, therefore, that so far as the LCXO method is concerned, the relative basicities of amino groups para located with respect to electron-attracting substituents provide a reasonable measure of the charge ( 5 ) G. W. Wheland, THISJ O U R N A L , 64, 900 (1942).
I
( 0 ) T i t h e charges O I I C ? , C , .tnd Cti in t h e -X@ a n d -C-Z syqtern< .+re iiiitiallv t(l 1 . i irihte.id < > l+ O 0: i a n rlertrtrn charge, t h e culrtilitrrl charar-: 1)11 a Piii.,i-alninc, g r o u p are aKaiii i,ractically equal i , i i i i , r ~ , ~ i i i i : i li ~ - 0l ~. 1 - I ( t i U I I electr,,ii c h a r r r ) A c o i ~ i t ~ u r a l ~calcule I:I~I~III II:~L. n , > tlxrii in:idr i < >trh e -Ysgstriii. 1,111 tlirrr i b I I O re V Y I X ' L . I ( 1 1 tt 1 1 1 ~ . rr\i111 \\o111,1 llr rlitrrrent.
In recent studies of growth requirements of the yeast phase of Histoplasma capsulatum in liquid snedia,' it was observed that the presence of fumarate in a cysteine-containing medium did not affect growth of the fungus when the medium was sterilized by filtration. When the medium was autoclaved, however, growth was completely inhibited. Substitution of pyruvic acid for fumarate completely inhibited growth whether the medium was sterilized by heat or by filtration. These results appeared to demonstrate a requirement of the fungus for -SH groups in the medium, since the reaction of pyruvate with cysteine in the cold' and the reaction of niercaptanswithunsaturated compounds3 are known. Although Morgan and Friedniann4 have reported the reaction of maleic acid with cysteine a t low temperatures, fumaric acid did not react with cysteine under their conditions. The amorphous addition product of maleic acid and cysteine was isolated by these workers and was identified as S-cysteinosuccinic acid. Because fumaric acid and cysteine are used in the preparation of certain culture media and in many biological systems their reaction a t elevated temperatures is reported here. -4 simplified procedure for the isolation of S-cysteinosuccinic acid as a crystalline salt of acetic acid asid some of the properties of this compound are also given. Experimental Ten grams of L-cysteine (free base) and 11.3 g. of fuinaric acid were added t o 200 ml. of 95% ethanol. The mixture was heated t o near boiling until the reactants dissolved and the presence of -SH groups could no longer be detected with nitroprusside and strong alkali. The time required for the reaction t o go to completion was approximately two hours. The ethanol lost by evaporation during this time was replaced with water. The solution was placed in a refrigerator a t 5" overnight. The following morning, any cystine or fumaric acid which crvstallized from solution was filtered off and the volume of the filtrate was reduced to approximately 25 ml. by vacuum distillation. One hundred ml. of distilled water was added, the solution was cooled to ;io, arid the e y c t w fuiri:iric a c . i t l iv1iic.h crystallized from solu-
