Sept. 20, 1954
NOTES
ZI3-dimethoxyacridine. The bulk of the trials gave products which contained minor amounts of arsenic, and repetition failed to yield substances having constant ratios of nitrogen to arsenic content. 6-Chloro- 9 - (4-diethylaminobutylamino)- 2 methoxyacridine (11) yielded a yellow substance which had the approximate formula (II).BHCl. H3As03; it contained no ionic arsenic. No structure has been formulated for either of the products containing non-ionic arsenic. A number of unsuccessful attempts were made to prepare arsenic (111) complexes of several amines (4 - diethylamino - 1 - methylbutylamine, 4 - diethylaminobutylamine, 3 - diethylamino - 2 - hydroxypropylamine and N- (2-hydroxyethy1)-ethylenediamine) and also certain 4-aminoquinolines (chloroquine and its 3-methyl analog, and also 7-chloro-4- [2-(2'-hydroxyethy1amino)-ethylamino ]-quinoline).
Regarding the Inherent Order of Electron Release by Alkyl Groups Attached to Electron Demanding Unsaturated Systems
4625
BY W. A. SWEENEY AND W. M . SCHUBERT MAY17, 1954 RECEIVED
We have evidence that the inherent order of electron release by alkyl attached to electron demanding unsaturated systems may not be the same as t h a t postulated for predominant C-H hyperconjugation.' I n Table I are values for the E-band of compounds of type I. This band results from excitation to a state in which there is a greater contribution of dipolar structures (Ia, Ib).2 Y is electron attracting to enhance differences in electron release of p-alkyls. Y
Y3
Experimental Arsenic(II1) Complex of 9-(3-Diethylamino-Z-hydroxypropylamino)-2,3-dimethoxy-6-nitroacridine.-A solution of 1.1 g. of 9-(3-diethylamino-2-hydroxypropylamino)-2,3-dimethoxy-6-nitroacridine dihydrochloridei in 5 cc. of water R I R Ia Re I b was treated with a solution prepared by dissolving 0.2 g. of TABLE I arsenic(II1) oxide in aqueous sodium hydroxide (0.16 g . in Compounda Xmax mfi (4 3 cc. of water). At first, a precipitate formed, then dissolved t o produce a clear garnet solution. No solid sepa- Acetophenone 242.0 13,100 rated on chilling, however the addition of ca. 50 cc. of diox- p-Methylacetophenone 253.0 14,108 ane caused a reddish solid to precipitate. The crude sub253.5 15,600 stance was crystallized twice from methanol to yield 0.7 g. p-Isopropylacetophenone 253.0 16,600 of reddish-orange microcrystals, m.p. 197-199" dec. (cor.). p-t-Butylacetophenone It dissolved in water to 1% ~1 . / v . ; the solution had a p H Nitrobenzene 260.0 8,160 6.25 and gave a precipitate a t p H 6.30 when 0.1 N NaOH @-Methylnitrobenzene 275.0 9,400 was added. No ionic arsenic was present in a solution of the 13,200 6-Acetylindane 258.0 substance.6 258.5 13,400 Anal. Calcd. for 3C22HliN4Oo.As2O3.H20: iY, 11.22; 7-Acetyltetrdlin As, 10.01; H20, 1.20. Found': N, 11.16; As, 9.92, 9.99; 2-Acetyl tetrahydrocyclohepH20,1.28. tabenzene 258.5 14,300 Alteration of the ratio of the acridine to arsenic over a 2-Acetylhexahydrocyclooctareasonable range produced the same substance. When the benzene 258.5 13,400 original reaction mixture was diluted with water, or conConjugate acid of: centrated in ZIUCUO, the base I , was obtained in most cases. It was undesirable t o reflux the reaction mixtures, for other- Acetophenonec 295.5 21,500 wise the base was the only product. This latter fact does p-t-Butylnitrobenzene 275.0 9,950 not lend much support to the possibility t h a t ring arsenation Benzoic acidb 228.0 11,830 occurs in the reaction. 236.5 14,620 Arsenic(II1) Complex of 6-Chloro-9-(4-diethylaminobutyl- p-Methylbenzoic acidb amino)-2-methoxvacridine Dihvdroch1oride.-Arsenic(II1) p-t-Butylbenzoic acidb 237.5 16,290 oxide '(0.7 g.) was dissolved in -aqueous sodium hydroxide p-Methylacetophenone' 312.5 24,500 (0.61 g. in 3 cc. of water); this solution of sodium arsenite 315 0 24,500 was added t o 5.13 g. of 6-chloro-9-(4-diethylaminobutyl- p-Isopropylacetophenone' 315.5 27,100 amino)-2-methoxyacridine dihydrochloride trihydrates in 6 p-t-Butylacetophenone" cc. of water at 60". A reddish oil separated, then a carmine p-Methylnitrobenzened 375.5 solution resulted when some dioxane was added; however, fi-t-Butylnitrobenzened 380.5 dilution with dioxane t o a total volume of 250 cc. gave an All spectra in 9590 ethanol except as noted. Beckmati orange precipitate (3.9 g.) . The product was crystallized DU used. Solution about 0.01 N in HC1. Spectrum in twice from methanol-dioxane, from which it separated as concentrated HzSOa. Measured in 101% HzS04 in which yellow microcrystals (after drying a t 100' (1 mm.) it weighed nitro compounds are known to fully ionize to conjugate acid. 3.2 g.), m.p. 222-223" dec. (cor., immersed a t 210'). An Extrapolated to zero time (slow sulfonation was occurring) aqueous solution failed to give ii test for ionic arsenic.6 without an exact measure of the extinction coefficient being Anal. Calcd. for C ~ I H ~ p C 1 N : ~ 0 . 2 H C 1 ~ H : ~S. A s7.18: O~: obtained. C1 (ionic), 12.12; As, f2.81. Found?: K, "7.291 C1 (ionic); 11.70; As, 11.30. The trend, both in Amax and E, may be opposite Repetition of the experiment gave a product which in order to the number of a-hydrogens, and the melted at 226-227' dec., but contained only 9.33% arsenic. spread in Amax is greatest when Y is most strongly STERLING-WINTHROP RESEARCH INSTITUTE electron demanding. Furthermore, no ring size RENSSELAER, NEWYORK effect in the order of steric inhibition of C-H hyper( 5 ) Prepared by a method essentially a s reported b y C. S. Miller and C. A. Wagner, J. Ovg. Chem., 13, 891 (1948). (5) As indicated b y lack of uptake of iodine in usual conditions with sodium hydrogen carbonate present. T h e original solution of sodium arsenite responded t o this test as expected. (7) Analyses b y Mr. M. E. Auerbach and staff of t h e Analytical Division of this Institute. (8) W.Huber, R. K . Bair and S C. Laskowski, THISJ O U R N A L , 67, l t i l 9 (1945).
con.jugation is observed. $ , l o The ionization potentials of alkylbenzene~~ also are opposite in order to t h a t expected for predominant C-H hyperconjugation, despite a large demand for electron (1) Cf. J. W. Raker, "Hyperconjugation," Oxford University Press, London, 1 9 X (2) See, e . g , K . Ruwden and E. A. Braude. J. Chem. Suc., IO(j8 (1952). (3) W. C Price, C h e m Reaiews.41, 262 (1947).
release by alkyl iri the excited pmitive ion. Other physical measurements (study, e.g., rei. 1 ), while knding to show a definite release effect by alkyl, have not unequivocally dciiioiistratcd the C-I3 hyperconjugative ordev of release.4 \Ye suggest that at least part of the apparent coritradiction between spectral and ionization potential data and rate data (for reactions in which there is a large demand for electron release in the transition state) may be explained by steric hindrance to solvation. This effect has been invoked by C. C. Price to account for reactivity orders in the saponification of i ~ and p-substituted alkyl benzoates,ia,b Price also suggested this factor may explain the similar rates of broiiiiria titm of t-butyl- and neopentylbeiizene.'" I n the Sx i reaction of bcnzhydryl solvation ( I [ tlic intermediate "ion" a t the partially positive 4tc.s in the neighborhood of the alkyl group wi)ul(l 1 ) ~ hindered inore by t-1311 thaii I I e and this solvatioii iactor should contribute to increasing ,E.\ uizd A.5' values iii the ordcr Are < I-BLI (cmipare Table I1 1 . The rate data of Dacldcley'Jcan bc. accounted for hy either steric inhibition to C-1.1 hyperconjug a t 1011 ' or solvation. The sigiiificaiit 1s" changes are consistent with the latter. .!mold's k clata o i l similar systeins"] fit either explanation, but there 14) Changes in Lhe I , ) \ Y intensit>- I3-band spec?ra ( o x r i t a t i , l i i :ii hoinopdar iurms! of alkyll,enze tib-e order of electron rZleace ( I< I,. Chutlke, Chevz. Reoif;.s, 4 71) liiive Isccn give:] ii rlitlcrerit interpretatiun i i y sevei-al au ( 5 ) V. J . I l a m i n o n d , \Y. C . P r i c e , J . 1'. Teegari aiiii .\ 11 L V c d ~ l i , ss. F i i r ~ d ~ .\ory , 9 , 53 (19.701 J. R . Plntt and I I . 13. K l e v e n s , < i , c i , i . : i ; ('11 C. C . Price a n d 1 ) . C I.;; ( l ! t , i l ) ; '1)) C C . Price and \V. J eelAi
is no clear trend in L A aiid AS". 'l'he solvation factor also could play a role i r i the ritte differences found in the bromination of C~H61i"~12 and C6IZjCH2R.1' 'FABLE ~ ~ Y D R O L Y S IOh' S
It =
(0") E.i(kcn1.) 106
kl
AS7
11
/ I - K C ~ ~ ~ ~ C IlX ~ C8(J"& I $ I A Q U E O U S AcE'roXe I1
9 82 71.11 -8.9
rile
13 t
83 5 189
6 2 ,6
,
11a.4
-8.6
-5J.8
l'r
1 - 1311
4 6 . 95
3.7 $1 70.03 -.7 li
E-
1!)X
-7.7
111light absorptioii, ewitation occurs too raI)idly
allow reorientation of solvent inolecules. l 3 SAvation of the ground state of the neutral molecules of Table I should be important only around the fuiictional group and hence play a very ininor role i i i cleterniiiiing differences in excitation energy. I n the coriiugate acids, appreciable positive charge is relayed to the ring, hence steric hindrance of solvation may contribute to the larger spread in A,,, values. The evidence presented indicates other modes of interaction, particularly C-C hyperconiugation, or perhaps induction aiid internal dispersion forccs,l4 may be more important than C-H hyperconiugation even when alkyl is attached t o an electron demanding system. t.1
(11) ( a j E . Berliner and F Bondhus, i b b . , 68, 2335 ( 1 S l O ) : ( 1 1 ) 13. Berliner and F.Berliner, i h i i i . , 71, 1195 (1949); (c) E. Berliner and F. Berliiier. i b i L , 72, 222 (19.50). 112) P. L, S3. de la Mare and P. W. Rubertson, J. Chem. SOC., 279 (IOt3). .sumption i 4 iitvolved in the commonly invoked Franckple; see, e.g , N. S . Bayliss, J , Cliem. Phys., 18, 2 3 2 (IOJO). ( I & ) \\'.T. S i m p s o n , T ~ r l s J O ~ ~ h . . 3 ~ ~M , 7(IU31), 3
USIVERSITY01: WASHINGTON 3 , i\'.4SIiI?;CTOS
SEAT'l'LE
~ C O ~ T K I IJ A~ U P KII IU C O I ~ < J I?IUER .; S ~ c r IrN , L'. S. I . ) H P ~ ..54).
b c ~ . t l .I,.
l
~
. i i i~< i
1.b J
Crcely, J . i'tiiy,iicu
. \ ~ i, 13, i!)J
