and chloroacetate buffers fur tlic raiige 2.0 to l3.5. Tlle ionic strength was maintained constant a t 0.02 M by addition of sodium chloride, the final concentration of the amine being 0.001 M. Spectral absorption curves were determined with a Cary recording spectrophotometer (model 11), the ratio RSHaC/RNH2 being determined a t intervals of 5 mp over suitable regions of the ahsorption curves, and the cell compartment being held a t 25 i 0.2’. p H measurements of the buffer solutions were determined with a Cambridge p H meter, the glass electrode being initially tested for linearity, and calibrated with a saturated solution of potassium hydrogen tartrate ( 3 . 5 7 ) . Readings were taken on the buffer solutions at 25’ before and after the spectrophotometric measurements with repeated recalibrations. Precision of p K , Results.-In this paper relating t o substituted anilines, and in subsequent papers dealing with iubstituted naphth5-lamines and heterocyclic bases, the technique of PK, measurement was standarized, and an rstimate of the precision can be made from consideration of the equation used, aiz., pK, = p H log RXHar/R1;H2. The quantity log R9H3’/KKHz was obtained from the relation RYHl’/KNH2 = (Ab - A / ( + - .-le) where A,, .Ab and il are the absorbances of the acid, base and buffered .;elutions. Matched 1-cm. cells tvere used, the blank differing from the absorbing solution only in the absence of amine. Spectrophotometric measurements were taken irt the range 250-400 mp in which for most systems there is a t least one isobestic point, and in which absorbances of the acid species are frequently negligible. In the case of aniline the values of log KSH3’/RSH2 for the buffers 4.74 and 4.25 were -0.077 i 0.005 and 0.400 =t0.009 (10 results) giving pk’, values of 4.66 and 4.65. IVith a correction of -0.06 for ionic strength, the figure of 4.59: is obtained with a precision which, allowing 1 0 . 0 2 for the pH measurements and kO.01 for the log ratios, mal- be given as Irithin 3 ~ 0 . 0 3 unit. Measurements were rejected unless log ratios were consistent and pK, values were within 0.03. Infrared measurements were made with a PerkinElmer model 112 recording infrared spectrophotometer using a LiF prism. Spectra were determined in approxiniately 0.005 -Ifsolutions of the amines in dried and freshly distilled carbon tetrachloride, gaseous ammonia spectra Ixing interpolated between the records of successive amines. .-\s4gnrnent of frequencies was made by reference to the
+
____-__ [ C O S T R I B U T I O S FROM THE S C H O O L O F
ainnionia bands 3396 and 3172 ciii.-l, the accuracy bciiig considered t o be 3~1 cm. - l .
Acknowledgments.-The author is indebted to LMr.I. Reece for taking the ultraviolet and infrared spectral absorption curves, and to Drs. M. S.Carpenter, B. M, Wepster and Messrs. Light and Co. for gifts of chemicals. Appendix.--The probability that the sets of data for the -11 and +31 groups of substituents lie on separate regression lines can be determined by evaluating the variance ratio F = s?, + b / s 2 where s p is the residual variance for all 16 substituents with 14 degrees of freedom, and S2a+b is the combined residual variance of the two separate groups calculated from the equation
where the number of degrees of freedom is 10. The residual variances are calculated by the equation s2
=
(I -
e
(ZAPK)? 8(APK*)2 n - 2
-
11
Y being the correlation coefficient and n the sample number.20 Values of 0.00277 and 0.00409 for s? and s2,+b give F I O , = I ~ sa+b/s2 = 1.5 corresponding to a significance level of about 0.25. Such a high level indicates that there is little difference between the probabilities of the ApK, values lying on a single line or being accommodated on two separate lines.
(20) K. A . Brownlee, “ I n d u s t r i a l E x p e r i m e n t a t i o n , ” H. M . Stat i o n a r y Office, L o n d o n , pp. 32,59. C . A . B e n n e t t a n d N. L. F r a n k l i u . “ S t a t i s t i c a l Analysis in C h e m i s t r y a n d t h e Chemical I n d u s t r y . ” John Wiley a n d Sons, I n c . , S e w York, X. Y . , 1950, p. 928.
.-
CHEMISTRY, ~ N I V E R S I T YO F
A7E\Y SOUTH \$:ALES,
S Y D N E Y , AESTRALIA]
The Effects of Substituents on the pK, Values and N-H Stretching Frequencies of 1and 2-Naphthylamines BY A. BRYSON RECEIVED OCTOBER 14, 1959 The pK. values and X-H stretching frequencies of 44 iia~~litli!~lariiiiies substituted in various ring positions by the groups NO?, CN, COOCHI, GI, Br, I, OH, OCHa and 9 H a have beeu measured, interest being devoted chiefly to the compounds in which the groups are in non-qulnonoid positions. The results reveal characteristic differences between the effects of substituents in vz-substituted anilines, 3-substituted 1-naphthylamines and 4-substituted 2-naphthylamines, the most noteworthy feature being shown by the 3-substituted 1-naphthylamines where substituents group themselves into two classes depending on their mesomeric +M or -hZ characteristics. Similar but smaller effects are observable in the 4-substituted 2-naphthylamines. The U-values derived satisfy equations of the type u = aui b u where ~ u / and U R are the Taft ~ 0 1 1 stants, and the extent of the polar and mesonleric contributions to the substituent effect in nz-anilines, 3-substituted l-naphthylamines and 4-substituted %naphthylamines can be inferred, m d the observed behavior explained Ai1 attempt is riiadc, to correlate t h e results with electronic structure. Relations for compounds with the substituents i i i the second ring are le55 clearly defined, but the distinction between the 7-substituted 1-naphthylamines and the k u b s t i t u t e d I-naphthylamines is interpreted t o indicate smaller intergroup conjugation in the latter. Unexpected relaticins are shown between pK,, valuei and S-H frequencies, theqe being characterized l>!- t h r non-conformity of the parent :imine\ with the lines given by the various positional series.
+
Previous communications‘ have described the effects of the groups 302 and SOa- on the pk’, Values of 1- and ?-naphthylamines substituted in the available ring positions, and a tentative interpretation has been offered2 for these effects in terms of structural changes in the base components of the ( 1 ) A R r y s o n , T~~~~~pnvnilay . c o r , 46, 237 1 2 ’ ,\ Tlr-i;son, i h i i i , 4 7 , 328 (IS331)
4 7 , 5,29 ( l c t 3 1 )
systems. This theory as applied to the 3-substituted 1-naphthylamines and 4-substituted 2naphthylamines, for instance, assumed that the inductive effect of the substituent NO2 on the NH3+ pole of the acid was the same as in the similarly constituted m-nitroaniline, and that the burden of changes in substitutional effects lay in influences affecting the delocalization of the electrons in the
Sept. 20, l!KiO
PIC,
AND
N-H STRETCHING IS K,IPIITHYL.OIINES
amino group of the base. I t was clear that this contention could only be maintained on much more experimental evidence, and a more systematic investigation therefore has been carried out, using in addition to the KO2 group, the substituents C1, Br, I, OH, OCHa, CN, COOCH3, CH3 and YH2 placed a t suitable ring positions. The pKa values and K-H stretching frequencies of these substituted naphthylamines have been measured, and it is now possible to compare the behavior of the various positional series over an adequate range of substituents. From the picture graphically revealed by Hammett plots, it now has become clear that the previous contention no longer can be maintained, and that the acid-base relations in the various series are distinctive, and appear to be determined by influences which reflect the importance of the mesomeric moment of the substituent. The relations between pKa values and N-H frequencies are also distinctive, and reveal a pattern which is difficult to interpret.
Experimental Melting points are not corrected. Substituted naplithylamines were prepared in most cases from the appropriate iiitronaphthylamines by the diazo reaction. Preparation of the mononitronaphthylamines closely followed the literature instructions ( a summary of methods has been publ i ~ h e darid , ~ no account will be given of these methods except in the case of 3-nitro-l-naphth>-lamine and 4-nitro-2naphthylamine where a modified procedure gives a n improved separation of these isomers). 3-Nitro-1-naphthylamine and 4-nitro-2-naphthylamine were prepared by the reduction of 1,3-dinitronaphthalene with sodium sulfide in boiling methanol as described by Hodgson and Birtwell,* and the filtered solution on cooling deposited practically pure 3-nitro-1-naphthylamine. T h e filtrate was diluted with water and the mixture of 3-nitro-lnaphthylamine and I-nitro-2-naphthylamine removed next day. Hodgson aiid Birtwel14recorded the difficulty of separating these isomers, but the following method based on the relative insolubility of 4-nitro-2-naphthylammoniurn nitrate proved satisfactor)-. To the ution of the mised added one-third of amines in 0.1 S hydrochloric acid the volume of saturated potassiuni nitrate solution whereupon a cream precipitate of 4-nitro-2-naphthylammonium nitrate was produced. Re-solution in dilute hydrochloric acid and precipitation with further potassium nitrate followed by basification gave 4-nitro-2-naphthylaniine which after cq-stallization from aqueous alcohol had 1n.p. 98' ilit.'m.p. 98.6"). Chloronaphthylamines were prepared from the appropriate iiitronaphthylamiries by diazotization a t 5' with sodium nitrite and pouring into cuprous chloride solution. After purification, reduction of the chloronitronaphthalene was effected with stannous chloride or by hydrogen over Raney nickel. Bromo and iodo compounds were prepared in malogous fashion. Hydroxy- and methoxy-naphthylamines .-The author is indebted t o Imperial Chemical Industries Ltd. for the gift of a number of these dyestuff intermediates. Compounds not available from this source were prepared as illustrated for I-methoxy-2-naphthylamine. 3-Xitro-1-naphthol ( 5 g. from 3-nitro-I-naphthylamine) was dissolved in lO(,.; sodium hydroxide (15 mi.) and dimethyl sulfate (10 g . ) added. The mixture was warmed on a mater-bath for 30 minutes when yellow crystals of I-methoxy-2-nitronaphthalene separated. Crystallization from ethanol gave yellow needles, m.p. 105', and reduction with stannous chloride gave 4niethoxy-2-naphthylamine, m.p. 59-60' (lit.5 m.p. 57.55 8 . 5 O ). (3) H . H . Hodgson a n d D. E . H a t h n a y , J . D y e r s n ~ Coi., d 63, 231 (1947). (4) H. H . Hodgson a n d S . Birtwell, J . Chem Soc., 75 (1914). f.7) n. H . R o s e n b l a t t , RI AI. S a c h l a s a n d A . 11 S r l i p m a n , THIS JoIiRN ii ,
80 2 4 f X l I 1938).
-LS(i:j
0-Methylation of Aminonaphtho1s.-The aminonaphthol ( 2 g.) was dissolved iu pyridine ( 5 ml.) and acetic anhydride ( 3 nil.) added. The mixture was heated on the waterbath for 15 minutes, producing a precipitate of the acetaminonaphthol acetate. After crystallization from alcohol, the substance was added t o 5 N sodium hydroxide (20 ml. 1 , and warmed to hydrolyze the acetate group as shown b!. complete solubility. T o this warm solution a t 50" w a s added dimethyl sulfate (2 ml.) after which the acetaminomethoxynaphthalene separated. The mixture was heated for a further 15 minutes, water added and the precipitate removed. Hydrolysis of the acetamino group mas effected by refluxing with hydrochloric acid and alcohol for 30 minutes. 3-Methyl-1-naphthylamine was prepared as described b y Marion and McRae.6 4-Amino-2-naphthol was prepared by the Raney reduction of 4-nitro-2-naphthol. The filtered alcoholic solution was evaporated in a stream of nitrogen and the residue crystallized from water in a n atmosphere of nitrogeu giving white needles rapidly darkening in air. The melting point determined in a n evacuated capillary mas 207'. Literature values are conflicting; e.g., 18j0,' 198'* and 211°,9 the last being in CO?. 4-Amino-2-naphthonitrile was supplied by the courtesy of Professor J . Cason. The compound was hydrolyzed to I-amino-2-naphthoic acid as described by this author,'" and converted t o the methyl ester by refluxing a methanolic solution in the presence of hydrochloric acid. Methyl 4-amino-Z-naphthoate, not previously described, forms colorless needles of m.p. 87" from petroleum ether. Anal. Calcd. for ClaHoT02: C, 71.6; H, 5.51; S,6.96. Found: C, 71.75; H, 5.41; N,6.90. 3-Amino-l-naphthonitrile, not previously described, was prepared by the reduction with iron and acetic acid of 3nitro-I-naphthonitrile m.p. l Z O ,prepared by the Sandmeyer reaction on 3-nitro-I-naphthylamine. I t crl-stallizes from cyclohexane in pale yellow needles of m . p . 116-117'. Anal. Calcd. for C I , H ~ N ? :C, 78.55: H, -1.78; N, 16.67. Found: C, 78.35; H, 4.53; N, 16.15. 7-Methoxy-l-naphthylamine, not previously reported, is prepared by the methylation of 7-hydroxy-1-naphthylamine and has m.p. 81O . 3-Amino- 1-naphthoic acid" was esterified by refluxing a methanolic solution through which hydrogen chloride mas passed. Methyl 3-amino-I-naphthoate has m.p. 72' (lit .ll m.p. 72-73' ). 1- and 2-naphthylacetic acids were commercial products and were purified by recrystallization from aqueous alcohol. p K , Determinations .-\Vith the esception of those nitronaphthylamines with pKa values less than 1.O, all measurements were made by the spectrophotometric method previously described for nz-substituted anilines, the concentrations of the amines being 0.0005to 0.001 X a t a n ionic strength of 0.02. As in the case of the substitutcd anilines, the accuracy and precision of the results arc dependent on the errors involved in the measurements of the pH of the buffer solutions used, the ratios RXH3'/RP;H?, and the correction term for ionic strength. I n general, for any given expefiment, two buffers were used close t o the expected p K , values, and measurements were rejected and the exvalues were not within periment repeated if the two Ph-,' 0.03 unit. Replicate experiments were carried out on a number of halogen substituted naphthylamines with the following results (after correction for ionic strength): 3-chloro-l-naphthylamine, 2.69,2.69; 2.67.2.66; 2.64, 2.64; mean 2.666, std. dev. 0.02; 3-chloro-I-naphthylamine (I.C.I. sample), 2.70, 2.72, mean 2.71; 3-bromo-l-naphthylamine, 2.71, 2.68; 2.66, 265, mean 2.676, std. dev. 0.03; 4-chloro-2naphthylamine, 3.41, 3.39; 3.37, 3.39; 3.36, 3.36; mean, 3.38, std. dev. 0.03; 4-bromo-2-naphthylamine, 3.41,3.42, 3.42; 3.39,. 3.38; mean, 3.40, std. dev. 0.02;4-iodo-2naphthylamine, 3.41, 3.40; 3.38, 3.39; mean, 3.39&, std. dev. 0.01. These figures were obtained with a Uvispek spectrophotometer, mark 11, in which the range of absorbance measurements is inferior to t h a t possible with the Cary (6) L. Marion a n d J. A. M c R a e , Can. J . R e s e n r c h , 18B, 267 (1940). (7) P. F r i e d l a n d e r , B e y . , 2 8 , 1948 (1895). ( 8 ) H . Challenor a n d C. K . I n g o l d , J . Chem. S o c . , 113, 2066 (1923) (9) D. Pummerer, et al., A n n . , 563. 40 (1938). ( I O ) J. C a s o n , THISJ O U R N A L , 63, 828 (1911). (11) G. J I.cuck, R . P P e r k i n s a n d I;. C I Y h i t m u r r , i h i , / 5 1 , 18:jI (l(J2!l),
A . BRYSON
.IS64
-.
PA. \
VOl. 8-3
TABLE I 7
~(IT20 ~
A .?
2~ io VTNI,ESS ~ ~ SPECIFIED)ASD N-Ii STRETCHING FREQL-ENCIES OF SURSTITLTED NAPHTHYLAMIVES (TOGETHER 1 v I n i 1- A S D ~-NAPHTHYLACETIC .ICII)S) --hI.p., Exptl.
oc.----.
I,i t .
Substituent X
pKs
APKs
N-H stretching frequencies
4.23 .. 4.30 .. .. 3.92" 3396 3477 4.16< 3399 3485 ' 135.5 2,07(2,22)' NO2 3405 1.85 3489 123-124 2.26 CN 3405 1.66 3489 68-69$ 2.66 C1 1.26 3403 3486 71 13r 2.67 1.25 3402 3486 2.82 88-89$ I 1.10 3401 3484 I , l-NH2, 3-X 87 3.12 .. COOCHI 0.80 3398 3481 .. . .f OCHI 3 26 3400 .66 3482 207 OH 3.30 211 . .k .62 . .k 51 3.96 51-52 - . 0 4 3395 3477 CHI 2.71 1.21 62 c1 3402 3486 2.43(2.63)" 98.5 NO2 1.73 3403 3493 1 50 2.66 3402 3491 CN 69 c1 3.38 0 78 3400 3487 68 1 I , 2-SH2, 4-X .76 Br 3.40 3400 72 3487 .75 I 3.41 7'6 3399 3486 .78 72-73 3.38 COOCHj 3399 3484 4.05 58-59 OCHs 3395 3481 .11 2.89(3.15)" 1.03 3400 3482 167 NO2 c1 0 44 3.48 63-64 3398 3480 111, 1-SH2, 6-X 73-74 3.90 OC?& 3396 3476 .02 OH - .05 3395 3475 3.97 190.5 3.01(3.13)" 1.15 3403 3495 "2 146 OH 4.07 0.09 3398 3481 199 3.10(3.13)" 3405 NO2 1.06 3494 184 3401 c1 3.71(23') 0.43 120-121 3494 V,2-KH2, 7-X OCH, 4.19 - .03 3397 3484 143 OH - .09 3397 3484 4.25 201 - .,Fo 339*5 3483 4 . 66b(18') NH2 161 1.19 2.73( 2.80)' 3388 3480 NO2 119 0 . F8 3399 3480 3.3'4 c 1 85 VI, 1-KHz, 5-X - .04 3397 3479 OH 3.96 Dec. - .29 3398 3479 4 . 21b SH2 189 2.55(2.83)" 1.37 3406 3489 so2 134 3477 3.48 0.44 3396 c 1 46 Y I I , l-KH2, 7-X 4.07 - .15 3387 3464 OCH3 . .0 - .28 4.20 3385 3464 OH 206 3.501 2.62(2.75)" 1.54 3408 KO2 203 YIII, 2-SH2, 6-X 3474 -9.48 4.64 3390 OCHi 139 3308 3500 2.73(2.86)" 1.43 I X , 2-NHz, 8-X so2 104-105 3109 3.38 3419 3 0 2 . . ( 0 54)" 191 X , l-iL'Hz, 4-X 3400 3.21 0.71 3482 Br 102 3522 -1.74( - 1 . 6 ) " 3365 X I , l - S H z , 2-X 5.66 NO2 144 3 54 1 X I I , 1-X, 2-NH2 - 0.85( - 1.0)" 4.95 3390 127 NO2 3. 1i 5 3407 1.48 2.62 X I I I , 2-SH2, 3-X 11fj-116h so2 3456 1.13 3347 XIV, l-KH2, 8-X . , (2.79)" KO2 97 * Dissociation exponents p K , and pK.r for a Values determined by Spekker absorptiometer and reported previouslv.1 symmetrical naphthylenediamines in 5OYc methanol have been published,13 only the results for the 1,s- and 2,i-diamines being of interest in the present context. T h e values are: 1-Naphthylacetic acid 2-Naphthylacetic acid 1-Naphthylamine t'-Pidphthylamine
134 142
134-135 142 5u 112 136 12 5- 126 62 71.5 84
P Kz
P KI
1.5-Naphthylenediamine 4 . 0 7 (20') 1 . 7 4 (20") 2 . 1 8 (18') 2.7-Naphthylenediamine 4 . 5 5 (18') Log 2 must be subtracted from p K 2 and added to p K 1 t o eliminate the statistical effect of the two h*H, groups, and a correction of -0.50 unit applied t o convert the values from 50y0methanol t o water.l4*'6*'R Using the temperature correction recommended by Hall and Sprinkle'' we obtain final pK2 values of 4 21 and 4.66 for 1,5- and 2,7-naphthylenediamine a t 25'. e The experimental values for 1- and 2-naphthylamine agree with those previously published, viz., 3 921'.18 and 4.11," 4.27," 3-iodo-1-naphthylamine and 4-iOdOrespectively. Literature values for the melting points of 3-chloro-l-naphthylamine, 2-naphthylamine are appreciably lower than those found in the present investigation. e A sample of 3-chloro-1-naphthylamine supplied by Imperial Chemical Industries confirmed the previous measurements of melting point, pK, value and N-H frequencies. f 3-Methoxy-1-naphthylamine was characterized by its acetyl derivative, m.p. 178" (1it.O m.p. 179'). * Sample supplied by Professor P. E. Verkade. Sample too insoluble in carbon tetrachloride.
Sept. 20, 1960
pKa AND N-H STRBTCHING IN NAPHTHYLAMINES
recording instrument used for most of the remaining determinations. The precision of the pK,’ values can be taken as of the order of k 0 . 0 3 . A standard correction of 0.06 for ionic strength is therefore justified. The values for the nitronaphthylamines are lower by amounts of 0.05 to 0.20 than those previously published,‘ and can be regarded as more accurate, for the latter were obtained in unbuffered solutions, and the ratios KNHa+/RNHz, measured with a filter photometer, are subject to greater uncertainty than those obtained with a spectrophotometer. The very weak bases 2-nitro-1-naphthylamine and 1nitro-2-naphthylamine required the following special method illustrated for the former compound. Solutions containing 8670, 30.3Yo and 0.1 N sulfuric acid were prepared, and to equal volumes of these were added equal weights of the amine to give a final concentration of 20 mg. per liter, the assumption being made t h a t in 8675 sulfuric acid the amine is entirely in the acid form, while in 0.1 N acid it is present as the free base. The absorbances of the solutions a t wave lengths 430, 440 and 450 mp were measured, from which the ratio log RNH,+/RNH,in the 30.3y0 sulfuric acid was determined as -0.20. The Hammett acidity function HO for acid of this concentration is - 1.54, hence the (apparent) pK. value for 2-nitro-1-naphthylamine is - 1.74. I n similar fashion the value for 1-nitro-2-naphthylamine was found to be -0.85. The PK, values of the 1- and 2-naphthylacetic acids were determined by potentiometric titration a t concentrations of 0.002 M in boiled-out distilled water, 0.05 N alkali being added from a microburet. The pK. values were calculated by t h e equation .-
where C, is the total acid concentration, C N ~is+ the added sodium hydroxide and CH+is the estimated hydrogen ion concentration. The ionic correction term was taken as +0.02. Infrared Measurements.-N-H stretching frequencies were determined in 0.001 to 0.005 M solutions of the amines in freshly distilled carbon tetrachloride as in the previous article. The author is indebted to Dr. R . L. Werner for the determination of the frequencies of a considerable number of the compounds and to M r . I . Reece for recording the spectra of the remainder.
Experimental Results A summary of the experimental results together with relevant data from the literature is shown in Table I . For economy of space no detailed references are given for the recorded melting points of the compounds. All data and references are, however, incfided in Elsevier’s “Encyclopaedia of Organic Chemistry, Vol. 12B, where these summaries are to be found: halogenonaphthylamines, p . 708; mononitronaphthylamines, p. 715; naphthylenediamines, p. 805; aminonaphthols, p. 1633; aminonaphthoic acids, p . 4183.
Discussion The Inductive Effect of the Naphthalene Ring.As part of the general survey of the effects of substituents in the naphthalene ring, the pKa values of 1- and 2-naphthylacetic acids were determined, since in these compounds the aromatic ring is isolated from the carboxylic group by a methylene group and may be expected to act purel.. inductively. In the absence of mesomeric interaction with the side chain, the inductive effect of the ring a t the 1and 2-positions should be determined by the electronegativities of the ring carbon atoms a t these positions. The values found, viz., 4.23 (4.24) for 1-naphthylacetic acid and 4.30 (4.26) for 2-naph(12) J. F. J . Dippy, S. R . C. Hughes and J. W. Laxton, J . Chem. SOL.,4102 (1954). (13) R. Kuhn and A. Wassermann, Helv. Chim. A d a , 11, 79 (1928). (14) G.Thompson, J . Chem. Soc., 1113 (1946). (15) A. L. Bacarella, E. Grunwald, H. P. Marshall and E. L. Purlee, J. O i g . Chem., 20, 747 (1955). (16) W. Q. Davis and H. W. Addis, J . Chem. SOL.,1622 (1937). (17) N. F.Hall and M . R.Sprinkle, THIS JOURNAL, 64, 3468 (1932). (18) R.C.Farmer and F. J. Warth, J. Ckcm. SOC.,85, 1713 (1904).
48135
.d /
,/
1 6.
1 4.
12 APK,
ID
oa. 061
Fig. l.--Ia, 3-substituted-1-naphthylamines +M; lb, 3-substituted-1-naphthylamines - M ; 2, m-substituted anilines.
thylacetic acid, agree with those (in brackets) previously published, l 2 and indicate that the inductive effects a t the 1- and 2-positions differ only slightly from one another and from that of the phenyl group in phenylacetic acid (4.31). In agreement with the above, it is found that the dissociation constants Ka are in the same order as the electronegativities derived from molecular orbital considerations by JaffC,lB and ‘indices of free valence’ defined by Daudel and Pullman20and Coulson.*l Graphical Representation of Results.--Graphical representations of the effects of substituents in various ring positions are shown in Figs. 1, 2, 3 and 4 by plotting A p K a values ( A p K a = log K/Ko where KOis the constant for the parent naphthylamine) against the appropriate Hammett m-substituent constants. The interpretation of this presentation, viz., that substituent constants a t all ring positions have the same U-value, and that reaction constants p vary from series to series is scarcely plausible, and it is more realistic to consider that the reaction constant for all positions is equal to that for m-substituted anilines. The appropriate substituent constants u then can be readily evaluated by drawing intercepts to the aniline line. The values so found are shown in Table 11. This information is supplementary to the table of substituent constants recently published by Wells and Ward,22and enables a comparison with the figures evaluated from other reactions. Analysis of ApK,-um Graphs.-Apart from the question of interpretation, Figs. 1, 2, 3 and 4 (19) H.H.JaffC, J . Chem. P h y s . , 20, 778 (1952). (20) A. Pullman and B. Pullman, “Les theories electronique de la chimie organique,” Mason et Cie, Paris, 1952, p. 338. (21) C. A. Coulson, “Valence,” Oxford University Press, New York, N.Y., 1952, p. 253. (22) P. R.Wells and E. R. Ward, Chcmisfry b Industry, 528 (1958).
0.6
"m
.
Fig. 2.--2a, ~-substituted-2-nspiitl1~-1~~~11iiies --AI ; %b, J-sul)stituted-2-Iiapli~h~latriines+lI; 2c, n2-substitutctl anilinei.
-0.4 -0.
0
0.1
0 . 2 &X
'
4CA
2 Ob 0.7
0.5
Fig 4.- - 1 , . i - ~ u l i s t i t u t e d - I - t i a p h t l i ~ ~ l a i ~ i i i ~2, e . : ;;-sub.tituted-l-:ial,litli! 1:iiiiiiiei. 3, C,-suh.;tituted-2-ii~j)htli!.laiiiiiicG
stituerit groups are in the same relative positions as in nz-substituted anilines. JIesomeric interactioii between groups should therefore be a t a minimuni, and the substituent effect predominantly inductive. It could therefore be expected that the Hainmett graphs wouId show regression lines of similar slopes passing through the origin. Figure 1, however, re\-eals an unexpected pattern for the h u b s t i t u t e d 1nalththylamines in which the substituents fall into two classes. those with - I, - character such as KO?. C S , COOCEI:; which appear to fall 011 a line through the origin, and those with -1. +lL character such as the halogens, O C H I and OH which lie on a h e passing ;tpproxiiriately through the point representing the S O ? group, and cutting the axis a t a positive value uf LpK,. The point for ;+CH., appears to lie slightly above the former line. Figure 2 shows the behavior of the +substituted 2n:iphthylamines and in this case all the substituents t.xcept 1-COOCH: appear to lie on a line approximately O.X.5 unit lower than the aniline line. However. it is ccjually plausible to regard the points rei)resenting the XO,, C X and CGOCH:, groups :LS coliiiear with a line passing through thc origin, the substituents Cl, R r . I, OCIS,Ilying oii the lower lincs. Such behavior is consistent with that of the : h u b and this i~~terprc~tation stituted 1 -naphth~~lamir~es will be atlopted. Coiisiderablr clarification of this behavior becoiiies possible if we apply niodifications of the equation proposed for ni -substituted bmzenc, cwiiipouncls hy Taft.'3 :,iz. U # ,
= u,
52,'j) I< 1%'. T x f t , J r . , "Steric \ I S . Sei\-niaii, J o h n \Vilr?. a n d
