2446
LOUISSCHMERLING, B. S. FRIEDMAN AND V. N. IPATIEFF
this anion might then combine with more indium ion to form In [In(C204)2(Hz0)z] 3.14Hz0, or 2Inz(Cd&)3’10H20. This would suggest that the compound referred to as the normal oxalate is dimeric with three-fourths of the indium in the complex and thus really a dioxalatoindate. Desiccation of this compound over anhydrous calcium chloride a t 2.5’ yields a product corresponding to Inz(Cz04)3.4Hz0,or In [In (Cz04)2(Hz0)z] 3.2HzO. The dioxalatoindate complex is not extremely stable, for solutions of the compounds prepared give nearly complete precipitation of indium hydroxide when treated with ammonium hydroxide. Furthermore, on digestion with boiling water in quantity insufficient to dissolve it completely, the potassium compdund gave a product analyzing for In, 50.17; CzO4, 27.72; G 0 4 / I n , 0.72, and probably consisting of a mixture of basic salts resulting from hydrolysis. A true potassium trioxalatoindate apparently does not result under the conditions employed. The formation of a material of indefinite crystalline structure corresponding to KzCz04.2K31n(C2o4)3.7H2Oseems to be a function of the experimental conditions as already pointed out. This substance readily yields the dioxalato salt, for recrystallization of Sample 70 (Table 11) gave a product analyzing for In, 29.10; Cz04, 44.01; C2O4/In, 1.975, values closely approaching those calculated for potassium dioxalatoindate. If any
[CONTRIBUTION FROM THE
VOl. 62
trioxalatoindate were present, the reaction could be formulated [In(CzO&I-
+ 2H20 + [ I ~ ( C Z O ~ ) Z ( H ~+O ) ~ ] -
Furthermore, when a solution of Sample S4 (Table 11) was treated with a solution of indium sulfate, a precipitate formed which analyzed for In, 28.49; cz04, 43.80; Cz04/ln, 2.001, values which agree almost exactly with those calculated for potassium dioxalatoindate.
Summary 1. Indium sulfate solution reacts with solutions of sodium, potassium and ammonium oxalates precipitating octahedral crystals of Na[In(C204)~(H~0)21 *H2O, K [ I ~ ( C Z O ~ ) ~ ( H.2H20 ZO)~I and N H ~ [ I ~ ( C ~ O ~ ) Z ( H ~ O ) ~ I . 2 . With oxalic acid, the precipitate corresponds to In~(C~04)3.10H~0, or perhaps In [In(CZO~)~(HZO)*] 3’14HzO. On drying over anhydrous calcium chloride, this becomes Inz(C204)3.4Hz0,or In [In(C204)2(HzO)~] 3.2Hz0. Dimeric formulas are given as possible explanations for the compositions of the materials. 3. Attempts to prepare potassium trioxalatoindate have led to an indefinite material approximating KzCz04.2K3[In(C204)3]-7HzO. 4. Use of the term ‘‘neutral” as applied to simple indium salt solutions is seriously questioned. URBANA,ILLINOIS
RESEARCH LABORATORIES OF THE
RECEIVED JULY 8, 1940
UNIVERSAL O I L PRODUCTS COMPANY ]
Notes on the Preparation and Properties of Some Aliphatic Hydrocarbons BY LOUISSCHMERLING, B. S. FRIEDMAN AND V. N. IPATIEFF Several paraffinic hydrocarbons have been prepared for use in connection with studies in this Laboratory. Their physical constants are presented along with some improved simple methods of preparation. Anti-knock values are also given. 2-Methylpentane was prepared via 2-methylpentanol-4 which was obtained by hydrogenating mesityl oxide, a compound which is more readily available than 2-methylpentanol-2 which has been used for the same purp0se.l An attempt to use diacetont alcohol as the
starting material for the synthesis of the hydrocarbon was unsuccessful. Hydrogenation of the keto-alcohol to the diol was accompanied by hydrogenolysis to isopropyl alcohol. The hydrogenolysis was decreased substantially by use of a lower temperature and of isopropyl alcohol as solvent. However, dehydration of the 2-methylpentanediol-2,4 over activated alumina did not yield methylpentadiene as expected ; instead, the product consisted chiefly of acetaldehyde and isobutene. This result may be explained readily as
PREPARATION AND PROPERTIES OF SOME ALIPHATIC HYDROCARBONS
Sept., 1940
2447
successful and pinacolyl alcohol was recovered quantitatively. On the other hand, a t 230') in CHsC-C--C-CH3 I l l -+ AlrOs the presence of a nickel-copper catalyst, hydrogenI l l CH3 H H ation was complete and the theoretical amounts of hydrocarbon layer and water were obtained. \O/ CH3-C-CH2-C-CH3 + H20 However, only about 15% of the hydrocarbon was I I the desired 2,2-dimethylbutane ; the remainder CH3 I H was the rearrangement product, 2,3' 0 dimethylbutane. \ CHaC---CH~--C--CH3 ---f (CH3)?C=CH2 CH3CHO (1) 2,3-Dimethylbutane was synthesized 1 by two methods. The first consisted H in the hydrogenation of the dimethyl/ butadiene obtained by dehydrating \ CH3C-CH2-C-C~, ---f ( C H 3 ) 2+ ~C ~ H ~ C H = C H- ~ (2) . , pinacol over alumina a t 427'. PinacoI I lone was a by-product of this reaction. L CHI H A The second method consisted of (a) the Reaction ( 2 ) occurs to a smaller extent. Analy- reduction of pinacolone to pinacolyl alcohol, (b) sis of the hydrocarbon gases by Podbielniak'" low- the dehydration of the latter by means of oxalic temperature fractional distillation and by sulfuric acid7 to a mixture of tetramethylethylene and acid absorptionzb showed that the yield of pro- unsym-methylisopropylethylene, and (c) hydropene was about 5% of that of isobutene. Acetone genation of these olefins to the desired hexane. An attempt was made to hydrogenate pinacol was not isolated since i t was present in too small in the presence of nickel directly to the 2,3-diamount to be separated from the aldehyde. Nor could 2-methylpentane be obtained by the methylbutane. When a nickel-kieselguhr catadirect hydrogenation of the diol in the presence of lyst was used a t 250°, the product was a mixture a nickel-copper catalyst a t 200'. Complete of 2,2- and 2,3-dimethylbutane, the latter being conversion t o isopropyl alcohol occurred. A present in greater amount. Similarly, the use of similar reaction in the presence of copper-chro- a nickel-copper catalyst resulted chiefly in 2,3,mium oxide catalyst was observed by ad kin^.^ dimethylbutane along with some of the 2,2-iso2J-Dimethylbutane was synthesized by the hy- mer and either pinacolyl alcohol or dimethylisodrogenation of t-butylethylene which was ob- propylcarbinol. tained by the pyrolysis of pinacolyl a ~ e t a t e . ~ The carbinols which were intermediates for the The intermediate pinacolyl alcohol (t-butyl- preparation of 2,2,3-trimethylbutane, 3,4-dimethmethylcarbinol) was prepared in practically theo- ylhexane, 2,2,3-trimethylpentane and 2,3,4-triretical yield by the hydrogenation of pinacolone methylpentane were synthesized by means of the in the presence of nickel a t 200°.5 This is more Grignard reaction. They were converted into convenient and efficient than the sodium-wet paraffins (a) by dehydration over alumina to form ether method.6 olefins, followed by hydrogenation, or (b) by siThe direct replacement of the hydroxyl group multaneous dehydration and hydrogenation over of the pinacolyl alcohol by hydrogen offers a pos- alumina and nickel by treatment with hydrogen sible scheme for obtaining 2,2-dimethylbutane a t 250-260'. from the alcohol in one step. Hydrogenation in Experimental the presence of nickel on kieselguhr, of black nickel Data for the preparation of the intermediates involved in oxide or of nickel-alumina mixtures a t 200' un- the synthesis of the paraffins are summarized in Table I. der 135 atmospheres pressure of hydrogen was un- The final steps in the syntheses as well as the physical OH
/[
H
OH
lH3
1
+
1
(2) (a) Podhielniak, I n d . Eng. Chem., Anal. E d . , 6, 172 (1933); Cf.Matuszak, ibid., 10, 354 (1938). (3) Adkins, "Reactions of Hydrogen with Organic Compounds over Copper-Chromium Oxide and Nickel Catalysts," University of Wisconsin Press, Madison, Wisconsin, 1937, p. 81. (4) Cramer and Mulligan, 106. c i i . We wish to thank Dr. Cramer for a private communication, detailing t h e preparation and the pyrolysis of pinacolyl acetate. (6) Adkins and H. I. Crarner, 'I'liIS JLIIJHNAL, 62, 4,'W (l93ll). ( 6 ) Cramer and Glasebrook, ibid., 61, 23U (1939). (b)
properties of the paraffins are given in Table 11. The hydrogenations were carried out in a rotating Ipatieff type bomb in the presence of about 10% by weight of a nickel-kieselguhrs catalyst under 100 kg /sq. cm initial hydrogen pressure. The alcohols and glycols were dehydrated by passage over activated alumina. (7) Whitmore aud Rothrock, ibid , 65, 1109 (1933) (8) Ipabefl and Corson, lrid h,#g Chew , SO, 1039 (1838).
LOUISSCHMERLING, B. S.FRIEDMAN AND V. N. IPATIEFF
2448
1701. 62
TABLE I INTERMEDIATES Method Yield, % ' (CHI)ZC=CHCOCHI H Za t 150°Q 7ub 2-Methylpentenes I A1203 a t 427"' (100) 2-Methylpentanediol-2,4 -tHHz (CH~)ZCOHCH~COC ~ a t loood 75 Pinacolyl alcohol (CHa)2CCOCH3 H P a t 2OOoe 100 Pinacolyl acetate CHtCOCl' IV 92 3,3-Dimethylbutene-l Pyrolysis of V a t 450"' 90 2,3-Dimethylbutadiene-1,3 iA1203 at 58-70 (CH3),C(OH)C(OH)(CH3)~ 427"O 2,3-Dimethylbutene-l and -2 IV ( C O O H ) ~a t 110-120"~ 55 2,2,3-Trimethylbutanol-3~ (CHs)3CCOCH3 CH9MgBr 85 3,4-Dimethylhexanediol-3,4 CH3COC2H5 Mgj 23 2,2,3-Trimethylpentan01-3~ (CHa)3CCOCH3 C2H5MgBr 52 2,3,4-Trimethylpentanol-3 (i-C3H1)2C0 CH3MgBr' 81 Compound
I 2-Methylpentanol-4 I1 I11 Iv V VI VI1 VI11 IX
X XI XI1
+
+
XI1
128-131 55-56 194-195 119-120 139-140 41-42 68-i0
+
+
XI11 2,3,4-Trimethylpentene-2'
B. p . , OC.
+
+ Ala03a t 325"
.... 1.4153 I . 4040 1.3768 1.4364
64-75 128-129 94-95 a t 10 mm. 68-70 a t 35 mm. 73 a t 30 inm.
+ + + +
1I2OD
1.414~ 1.3860
114.5-114.9 a t 749.5 mm.
.... .... 1.4540 1.4352 1.43501.4356 1.4269
a Adkins, Zoc. cit., p. 52. There was also obtained, in 28% yield, methyl isobutyl ketone, b. p. 115-117'; ~ * O D 1.3964 The methyl isobutyl ketone was not removed prior to the dehydration. Isopropyl alcohol (25% by weight of the diacetone alcohol) was used as solvent in order to decrease hydrogenolysis to isopropyl alcohol. I n the absence of solvent and a t 125" the 2-methylpentanediol-2,4 was obtained in only 57% yield while isopropyl alcohol was isolated in 38% yield. Adkins and H . I . Cramer, loc. cit., report a quantitative yield of the diol a t 150"; they did not obtain any isopropyl alcohol. E Adkins and H . I . Cramer, Zoc. cit. P. L. Cramer, private communication. German Patent 235,311 (1910). Whitinore and Rothrock, loc. cil. Edgar and Calingaert, There was also obtained pinacolone in 25-39% yield. THISJOURNAL, 51, 1486 (1929). McCallum and Whitby, T r a n s . R o y . Soc. Chenz., 22, 39 (1928). Whitmore and Laughlin, THISJOURNAL, 54,4012 (1932). Idem., p. 4392. "' 450 cc. isolated by distillation through a 100-plate column (footnote d , Table 11) of the 1510 cc. of the octenes resulting from the dehydration of X I I : d2040.7428; daor 0.7260; dD/dt, 0.00084; octane number, 83. e
'
'
'
'
TABLE IIa
SYNTHETIC PARAFFINS Name
%Met hylpentane 2,2-Dimethylbutane 2,X-Dimethylbutane 2,2,3-Trirnethylbutanc 3,4-Dimethylhexane 3,2,3-Trimethylpentane 2,3,1-TriInethylpenLane
By t h e hydrogenation of
I 1 a t 50" VI a t 50" VI1 and VI11 IX
X XI XI11
B. p . , O C .
59.4-59.6 49 .'4-49.5 57.4-57.5 80.5-81 . O 118-118.3 110--110.5c t 12.3-112.4~
Press., mm.
750 753 745 748 i50 749 73ti
dQi
d204
dl04
dD/dt
0.6705 0.6529 . . . . 0.00088 ,6663 ,6485 . , . . ,00089 ,6785 ,6618 .... ,00084 .... ,6895 0.6727 ,00084 .... ,7193 . X i 5 ,00079 .
...
....
,...
....,
,718;
,7027
,000XO
ittO~
Octane numberb
1.3716 7 1 . 5 1 ,3688 93 1.3751 94 1.3894 100 1.4041 81.5 1 ,4030 100 1 ,4050 94. 5
'' Cj. Egloff, "Physical Constants of Hydrocarbons," Reinliold Publishing Cornpany, New York, N . Y., 1939, Vol. I . C . F. R. motor method; "A. S. T . M . Standards on Petroleum Products and Lubricants," American Society fot-Testing Separated by careful fractionation of the mixture of octanes resulting from ?\laterials, Baltimore, Md., 1938, 11, 153. the simultaneous dehydration-hydrogenation of XI. This purified fraction amounted to 48,t570 by volutne of the total 100-plate bubble-cap still, Bruun and Faulconer, Ind. Eng. Chem., Anal. E d . , 9, 192 (1937); b. p. determined octanes. in modified Cottrell reboiler, Bruun and Hicks-Bruun, Bur. Standurds J . Reseurch, 6, 871 (1931). As a n example of the simultaneous dehydration and hydrogenation of the alcohols ( I X and X I ) and the glycol (XI, the formation of 3,4-dimethylhexane from ( X ) is described. Three hundred cc. of the diol, 50 cc. of isopentane, 2 g. of powdered activated alumina and 18 g. of nickelic oxide was heated for ten hours a t 250-260' in a &liter rotating bomb under 100 kg./sq. cm. initial hydrogen pressure. The hydrocarbon layer was then separated from the resulting water and the catalyst, dried over sodium and fractionated Dehydration of 2-M ethylpentane diol-2,4 (111).-Forty grams of the diol was passed during forty minutes over 50 cc. of activated alumina in a tube a t 427O. There was
obtained i cc. of water, 6 cc. of a viscous upper layer which was not further investigated, and 36 cc. of material condensed a t -78". The latter was redistilled by warming t o room temperature, bubbling the gas through water into a trap immersed in dry-ice-acetone mixture. A low temperature Podbielniak" distillation gave the following as the composition of the water in5oluble condensate : C3H6, 4.1%; C3H8(?), 0.7%; i-C4Ha, 81.0%; n-CdHa, 6.9%; C4Hlo(?),1.2%; C j f , 1.5%. There was also found 4.6% of material boiling a t -23 to -20". The aqueous solution contained acetaldehyde, which was identified as the p-nitrophenylhydrazone, in. p. 128°.q (9) Hyde, B e y . , 32. 1813 (1899).
Sept., 1940
AROMATICAMINESA N D 3-NITRO-6-FLUORONITROSTYRENE Summary
The synthesis of three hexanes, one heptane and three octanes is described. Their boiling points, refractive indices, densities, and octane numbers are given. 2,2,3-Trimethylbutanol-3, 3,4-dimethylhexane-
[CONTRIBUTION FROM
THE
2449
diol-3,4 and 2,2,3-trimethylpentanol-3may be simultaneously dehydrated and hydrogenated to yield the corresponding paraffin. Dehydration of 2-methylpentanediol-2,4 Over activated alumina yields chiefly acetaldehyde and isobutene. RIVERSIDE,ILLINOIS
RECEIVED J U N E 22,
PEARSON MEMORIAL LABORATORY OF TUFTSCOLLEGE ]
Aromatic Amines and 3-Nitro-6-fluoronitrostyrene BY DAVIDE. WORRALL AND HENRYT. WOLOSINSKI The relatively weak tendency of nitrostyrene to form addition compounds with certain organic bases is slightly strengthened by the introduction of a nitro group into the aromatic nucleus.1 It is also true that the additive tendency toward other bases, hydroxylamine or aniline for example, is lost, perhaps from an increased tendency for destructive oxidation-reduction reactions. The first effect is increasingly enhanced by the presence of chlorine, bromine, and iodine in the nitrated aromatic ring,2 while the inhibitory factor is correspondingly diminished. The present communication completes the series of halogen derivatives in which i t seems clear that fluorine is the least effective in promoting these addition reactions. Experimental
amine, also with benzidine, the latter separating as orange crystals, m. p. 134-135". Anal. Calcd. for C28HZ4F2N401:C, 64.9; H, 4.6. Found: C, 64.8; H, 5.1,
a-Nitro-p-(6-fluoro-3-nitrophenyl'i-ethylene (II).-A small portion of I was nitrated a t ordinary temperatures using fuming acid. The product, precipitated by means of ice water, crystallized from alcohol as slender pale yellow needles, m. p. 142-143". Anal. Calcd. for CsH5FN204: C, 45.3; H, 2.8. Found: C, 45.4;H , 2.7. I t has been assumed that the nitro group has the same position in the ring as with the derivative? previously studied. The same procedure previously used was followed in preparing the addition compounds, which were slow in forming.
TABLE I P-DERIVATIVES O F a-NITRO-~-(&FLUORO-3-NITROPHENYL)ETHANE
~u-Nitro-p-(2-fluorophenyl)-ethylene(I).-A
mixture of fluorobenzaldehyde prepared by hydrolysis of o-fluorobenzalbromide with an equivalent amount of nitromethane containing a small portion of triethylamine1f2after cooling and standing for several hours was steam distilled in the presence of hydrochloric acid; yield, approx. 60%. (I) separated from ligroin as narrow yellow plates, m. p. 56.557.5O. A n d . Calcd. for C R H B F N O ~C,: 57.5;H, 3.6. Found: C, 57.3;H , 3.7. a - Bromo a nitro - p - ( 2 - fluorophenyl) ethylene,A small portion of I wm brominated in hot chloroform solution and the residue treated with an equivalent amount of alcoholic potassium acetate, avoiding the application of heat. After pouring the mixture into water, the residue crystallized from alcohol as yellow prismatic needles, m. p.
- -
89-90", Anal. Calcd. for CsHSBrFNOz: Found: C, 38.6; H , 2.4.
-
C, 39.0;
H, 2.0.
It formed a n addition product with p-phenylene-di(1) Worrall and Benington, THISJOURNAL, 60,2844 (1938). (2) Worrall and Benington, i b i d . , 62, 493 (1940).
M,. p . , Formula C. CiaHnFN304 134-135 CiaHiiFNaOa 106-106 CirHiaFNsOa 116-117 CIIHIJFNCOP 103-104
Substance Anilino m-Toluidino 9-Toluidino Phenylhydrazino
--Analyses, Calcd. C H
55.1 56.4 56.4 52.5
3.9 4.4 4.4 4.1
%-
Found C H
54.5 56.7 56.1 52.8
4.0 4.7 4.6 4.6
Negative results were obtained with o-toluidine, the anisidines, hydroxylamine, P-tolylhydrazine, and ammonia. N,N' (a+' Di - (6 - fluoro - 3 nitrophenyl) p,p'dinitrodiethy1)-benzidine.-Yellow granular crystals appeared eventually on mixing a n alcohol solution of the nitro derivative of I with benzidine, m. p. 139.5-140.5'. Anal. Calcd. for C2aH22F2NeOa: C, 55.3; H, 3.6. Found: C, 55.0; H, 4.0.
-
-
-
-
Summary 3-Nitro-6-fluoronitrostyrene has been prepared and its behavior examined toward aromatic amines. I t is less active in forming addition compounds than analogous derivatives previously examined. MEDFORD, MASS.
RECEIVED JULY 2, 1940
