3 36

A D A M C I K AND ? V I I K L A S I E R I C Z

VOL.

28

to a 2,4-dinitrophenylhydrazone, m.p. 221-222' (reported,1%21 m.p. 224-225"), which showed no depression in melting point upon admixture with an authentic samplelz of the hydroxy ketone 2,4-dinitrophenylhydrazonederivative. The aqueous acid solution obtained from the above hydrolysis was made basic and extracted with ether. The ether extract was dried over anhydrous potassium carbonate, the ether removed under reduced pressure (aspirator-steam bath) and the residue was fractionated in vacuo. The amino ketone X I I I , 2.22 g. (57YC), obtained as an oil, b.p. 82-86' (8 mm.), n32~ was 1.4611, d 5 0.9481. The infrared spectra indicated a single carbonyl ab24 sorption band a t 5.85 b. Anal. Calcd. for CloH18NO: C, 70.96; H , 11.31. Found: C, 70.66; H , 11.55. The hydrochloride salt was recrystallized from methanol-ether, m.p. 158.5-160".
(21) A. Bell, T. H. Stickland, and G. F. Wright, J . OrQ. Chem., 16, 1742 (1961); K. R. Bharucha, H. L. Cohen, and G. F. Wright, ibid., 19, 1097 (1954).

Anal. Calcd. for CloHzoClNO: C, 58.38; H , 9.80. Found: C, 58.64; H , 9.84. Oxidation of Amino Ketone XI11 to Geronic Acid.-Twotenths gram (0.0012 mole) of XI11 was dissolved in 24 ml. of 2 N sodium metaperiodate solution. The reaction mixture was allowed to stand a t room temperature for 20 hr. At the end of this period the periodate solution was made strongly alkaline, extracted with ether, and the ether extract discarded. The aqueous portion was acidified, and extracted twice with ether and once with chloroform. The extracts were combined, dried over anhydrous sodium sulfate, and the solvent removed under reduced pressure. The residue was treated with 2,4-dinitrophenylhydrazine to yield 0.26 g. (62%) of the derivative, m.p. 135.5137" (reported m.p. 135.5-137'2a or 139.5-14012). A depression in melting point was not observed upon admixture with an authentic sample12 of geronic acid, 2,4-dinitrophenylhydrazone.

(22) H. H. Strain, J . A m . Chem. Soc., 67, 758 (1935).

Cyanoethylation.

I. Weakly Basic Catalysts in the Reaction of Acrylonitrile with

Active Methylene Compounds1
JOEA. ADAMCIK EDWARDMIKLASIEWICZ AKD J.
Department of Chemistry, Texas Technological College, Lubbock, Texas Received August 6 , 1961

Cyanoethylation of some active methylene compounds has been shown to proceed in the presence of triethylamine, often in high yields. With this catalyst a t 25" the cyanoethylation of acetylacetone is greatly affected by the nature of the solvent. Little or no reaction occurs in tetrahydrofuran or dioxane, and the rate of the reaction varies with the nature of the solvent in this manner: t-butyl alcohol < isopropyl alcohol < 95YCethanol < "50% ethanol." The cyanoethylation of a-cyanoacetamide in the presence of triethylamine gave the expected product a t 20', but an isomer (VIII) a t 65-70'. Studies on VI11 have enabled us to formulate it as 3cyano-3-(2'-cyanoethyl)-6-imino-2-piperidone.

Strong bases, such as benzyltrimethylammonium hydroxide and potassium hydroxide, are ordinarily used as catalysts in the cyanoethylation reactions2 However, the use of cyclohexylamine in the cyanoethylation of ethyl cyanoacetate has been reported.3 Wakamatsu and Shimo have recently reported liquid ammoniacatalyzed Michael additions of nitroparaffins4 and of derivatives of malonic and cyanoacetic acid@ to acrylonitrile and other acceptors. Nevertheless, the use of such catalysts has not generally been exploited in synthesis. Since it might be expected to have the advantage of giving a more easily controllable reaction, the employment of triethylamine in the cyanoethylation of some active methylene compounds has been investigated. The results obtained are reported in Table I. The yields can be quite high and the method is very convenient. It is necessary only t o allow a solution of the active methylene compound, acrylonitrile, and triethylamine in a suitable solvent to stand a t room temperature whereupon the product, if solid, separates from the reaction mixture. I n the course of the work, it was noted that the nature of the solvent had a great effect on the rate of the reaction and on the ultimate yields obtainable. In nonhydroxylic solvents such as dioxane and tetrahydro(1) The support of this work b y the Robert . . IWelch Foundation is gratefully acknowledged. (2) H. A . Bruson, Org. Reactions, 6, 81 (1949). (3) G. Wiest and H. Glaser. U.S. Patent 2,396,626 (March 12, 1946). (4) S.K a k a m a t s u and K. Shimo, J . Org. Chem., 27, 1609 (1962). (5) S Wakamatsu, tbzd., 27, 1285 (1062). . (6) K. Shimo and S. Wakamatsu, %bad., 26, 3788 (1061).

furan the reaction took place extremely slowly if at, all while it did proceed a t a reasonable rate in alcohols. This factor was therefore investigated more thoroughly by conducting the cyanoethylation of acetylacetone in various solvents a t 25'. The ultimate yield of product was determined, and, as a rough indicator of the reaction rate, the time required to obtain one half of the ultimate yield. The results are shown in Table 11. That the reaction rate increases with the solvating ability of the solvent is shown by the order of reactivity in the solvents studied.
5001, CzHsOH-50% H20 > CzHsOH > (CHs)*CHOH > (CHIhCOH

The following explanation can be given for our results. The monocyanoethylation of acetylacetone in the presence of triethylamine proceeds as follows:
(CH3CO)&Hz (CHZCO)&H-

+ (CzH6)3N +(CHzC0)zCH- + (CnHs)rNH' + CHz=CHCN +

( CH,C0)2CH--CH2CH--CS

VI

where HA is an acidic species such as acetylacctone, triethylammonium ion, or possibly even solvent.' Since the reaction involves formation of the ionic species VI and triethylammonium from neutral molecules, the observed dependence of the rate on the solvent is expected.
(7) This mechanism is a modification of the one given for the hydroxide ion-catalyzed cyanoethylation of acetylacetone by Y . Ogata, &I Okano. .. Y. Furuya. a n d I. Tabushi, J . Am. Chem. Soc.. 7 8 , 6426 (1956). I t is essentially the same as t h a t given by Wakamatsu (ref. 5 ) .

FEB~WAHY, 1963

CYANOETHYLATION. I
TABLE I
CYANOETHYLATIONS
M.P.,
Lit. m . p , OC. Yield,

337

Starting niatcrial

Solvent a n d catalyst

Time

Product

4C.

Lit. yield, 70

Acetylacetonc (0.1 mole) t-BuOH H20 Et3N Methyl acetoacetate i-PrOH (0.1 mole) HzO

Et3X
Ethyl acetoacetate (0.1 mole) i-PrOH H20 Etas Malononitrile (0.2 mola) EtOH

EtaN
a-Cyanos.cetamide

(45 ml.) (15ml.) (lOml.) (30 ml.) (30ml.) (lOml.) (30 ml.) (30ml.) (10 ml.) (60 ml.) ( 5ml.)

2 days

r,r-Diacetylpimelonitrila ( I )

185-166

182-184"

77

76"

1day
1 day

-,-Acetyl-r-carbomethoxypimelonitrile (11)

152-154

154h

88

5Oh

yAcetyl-r-carbet'hoxypimelonitrile (111) s,~-Dicyanopimelonitrilc (IV)

s1-s3
S9-90

S2h

77
57

50"

2 hr.
c

02d

Wd

r-Carbamg 1- ycyanopimelo117.0-llS.5 118' 63 74' nitrile ( V ) H. A. Bruson and T. W. Riener, J . ,41~. C'hem. SOC., 2850 (1942). 64, a A. B. Boese, Jr., U. S. Patent, 2,438,061 (April 6, 1948). G. R. Zellars and R Levine, J. Org. Chem., 13, 911 (1948). Ref. 10. e When conducted at 20' by the procedure described in the Experimental. Ref. 8.

TABLE I1 CYANOETHYLATION OF ACETYLACETOKE VARIOUS IN SOLVENTS AT 25"

Ultimate yield, Solvent

CNCHiCONnp

CH,:CH CN E13N. 2 0 *

CHICHICN CN-C-CONH) CH2 CHICN VI1

EI,N

CNCHICHI~ ,CHICHI,

65-10.

,,/".,N/C."
O H

Time required to obtain half of ultimate yield, hr.

Vlll

"50% Ethanol"" 50 95% Ethanol 57 Isopropyl alcohol 67 &Butyl alcohol 54 a Equal volumes of 95y0 and water.

2 23 89 106

The triethylamine-catalyzed cyanoethylation of acyanoacetamide gave anomalous but interesting results. When the cyanoethylation was carried out a t 20', a 63% yield of y-carbamyl-y-cyanonitrile (VII) was obtained. Bruson and Rieners reported obtaining VI1 from the reaction of a-cyanoacetamide with acrylonitrile in the presence of benzyltrimethylammonium hydroxide a t 35-40', Also, Wakamatsu6 prepared this compound by cyanoethylation of a-cyanoacetamide in liquid ammonia at -50'. At 65-70' the sole product was VIII, an isomer of VII. A mixture of VI1 and VI11 was obtained a t intermediate temperatures; separation was effected by utilizing the insolubility of VI11 in tetrahydrofuran. When VI1 was treated with triethylamine in aqueous alcohol solution at 60-70', it was converted in good yield to VIII. Hydrolysis of VI11 in 6 N hydrochloric acid a t 50-60' gave IX, formula C9H9N302. Hydrolysis of VII, VIII, or IX in refluxing concentrated hydrochloric acid gave X, formula CgHIoK204. Titration of IX with aqueous sodium hydroxide showed it to be a weak acid with pKa' of roughly 8.6. Titration of X showed that it is a somewhat weaker acid than IX, although it was not possible to obtain a definite value. Reduction of IX with lithium aluminum hydride gave a diamine, XI, formula CgHlsNz. A consideration of these facts, the infrared spectra of the compounds involved, and the nuclear magnetic resonance spectrum of XI enabled us to formulate these transformations as shown in Figure 1. The spiro-diimide (X) has been previously reported
( 8 ) 13. A . Bruson a n d T.

by Sury and Hoffmanng and by Mariella, Clutter, and Ebner.Io The first-mentioned workers obtained it by cyclization of diethyl y, y-dicyanopimelate with sulfuric and acetic acids; the latter by cyclization of y,ydicyanopimelonitrile, y-carbamyl-y-cyanopimelonitrile or y, y-dicarbamylpimelonitrile with concentrated hydrochloric acid. The melting points reported by these two groups disagree significantly; our value agrees closely with that of Mariella, Clutter, and Ebner. The spiro-diamine (XI) has been obtained by Sury and Hoffmanng from lithium aluminum hydride reduction of X, as well as from reduction of related starting materials. The decomposition point of the picrate which we observed differed significantly from their value. l1 However, the nuclear magnetic resonance spectrum and the other data reported here leave little doubt as to the structure of XI. The interesting cyclization of VI1 to VI11 is not without precedent. Birch and Kon12 reported the synthesis of XI1 and XI11 from the reaction of A'-cyNH "
CHt-CHz

CHz

/ \
CHz-CHZ

' /
/ \
II

CHz-C

I/

CH2-CHz

CH-CO

/ NH

CH2-CH2

CHZ-C ''

\\

/ \

CH-CO
I

/SH

cpi
XI1
XI11

&N

(9) E. Sury and K. Hoffmann, Hela. Chim. Acta, 36, 1815 (1953).

W Riener. J . A m . Chem. Soc., 6 6 . 2 3 (1943). .

(10) R. P. Mariella, R. Clutter, a n d H. G. Ebner. J . Ore. Chem., 20, 1702 (1955). (11) Sury a n d Hoffmann did not correct their melting points. (12) S. F.Birch and G . A . R. Icon, J . Chem. Soc., 1 2 S , 2440 (1923).

338

ADAMCIK D MIKLASIXWICZ AN

VOL. 28

clohexenylacetonitrile and Al-cyclopentenylacetonitrile, respectively, with cyanoacetamide. Kandiah and Linstead13 showed that these products resulted from the isomeric a,@-unsaturated nitrile formed by isomerization in the alkaline medium.

Experimental
All melting points are corrected. Microanalyses were performed by Schwarzkopf Microanalytical Laboratories and Geller Microanalytical Laboratories. Infrared spectra were obtained with a Perkin-Elmer Infracord, Model 137. Nuclear magnetic resonance spectra were recorded with a Varian Associates Model A-60 spectrometer. Cyanoethylations.-The general procedure for cyanoethylations involved allowing a solution of the active hydrogen compound and an equivalent quantity of acrylonitrile in a suitable solvent-catalyst mixture to stand in a water bath a t 25". Reaction of malononitrile was so vigorous that ice-bath cooling was utilized initially. The product was then removed by filtration, washed thoroughly with ethanol, and dried. Cyanoethylation of a-Cyanoacetamide.-A solution of 8.41 g. (0.10 mole) of a-cyanoacetamide, 20 ml. of triethylamine, and 10.6 g. (0.2 mole) of acrylonitrile in 30 ml. of 9570 ethanol and 30 ml. of water was placed in a water bath maintained at the desired temperature for the time indicated. The product wa8 isolated by filtration and washed with ethanol. When the water-bath temperature was maintained between 65 and 70", the reaction time was 3 hr. and 14.75 g. or 777, of 3cyano-3-(2'-cyanoethyl)-6-imino-2-piperidone(VIII), m.p. 221222", was obtained. After recrystallization from ethanol-water, the product melted a t 223-224'. Anal. Calcd. for CpHIoNdO: C, 56.83: H , 5.30; N,29.46. Found: C, 57.09; H , 5.42; N, 29.68. When the water-bath temperature was 20', the reaction time was 8 hr. and the yield of ycarbamyl-7-cyanopimelonitrile (VII) was 11.9 g. (63%), m.p. 117.0-118.5"; lit.,s 118'. The product was completely soluble in tetrahydrofuran. When the water-bath temperature was maintained between 40 and 50', the reaction time was 2.5 hr. and the yield of product was 12.10 g. (64%). When the product was treated with tetrahydrofuran, 9.55 g. (50%) of VI11 remained undissolved. The tetrahydrofuran was removed from the filtrate by distillation, whereupon 2.5 g. (13%) of VI1 remained as a residue. The infrared spectrum of VI1 (KBr disk) showed the following significant bands: 3450 cm.-l(s), 3200crn.-'(s), 2260cm.-l(m), 1700 em.-l(s), 1630cm.-'(m). Thatof VI11 (KBrdisk)showed: 3400 cm.-1 (s), 2270 em.-' ( m ) , 1680 em.-' (m-shoulder), 1650 em.-' (9). Transformation of VI1 to VII1.-A mixture of 20 g. of VII, 135 ml. of %yo ethanol, and 45 ml. of water was prepared and placed in a water bath held between 60 and 70' for 5 min. Then 40 ml. of triethylamine was added and heating was continued for 1.25 hr. The mixture was filtered and the product (VIII) washed with ethanol; m.p. 221-222'. The yield was 15.45 g. (77y0). Hydrolysis o VI11 in 6 N Hydrochloric Acid.-Ten f grams (0.053 mole) of VI11 in 150 ml. of 6 N hydrochloric acid was heated in a water bath held between 50 and 60' for 45 min. The mixture was placed in a refrigerator overnight, filtered, and the product, 3-cyano-3-(2'-cyanoethyl)-2,6-piperidmedione (IX), washed with cold water. The yield was 5.65 g. (5670); m.p. 134135'. The analytical sample was obtained by recrystallization from ethanol-water: m.p. 134-135'. Anal. Calcd. for CeH9N302: C, 56.54; H , 4.74; N, 21.98. Found: C, 56.53; H, 4.72; N , 21.88. Significant band in the infrared spectrum of I X (KBr disk) were: 3400 em.-' (w), 3200 cm.-' (s), 3100 em.-' (m), 2260 em.-' (m), 1720 em.-' (s). Hydrolysis of VI1 in Concentrated Hydrochloric Acid .-A mixture of 24 g. (0.126 mole) of VI1 and 150 ml. of concentrated hydrochloric acid was refluxed until a white solid precipitated and for a few minutes thereafter. The mixture was cooled to room temperature, filtered, and the product, 2,8-diazaspiro[5.5] -undecane-l,3,7,9-tetrone (X), washed with ethanol. It weighed 10.0 g. (38y0). On repeated recrystallization from water, it melted a t 283-284".
(13) A. Kandiah and R. P. Linstead, J . Chem. Soc., 2138 (1929).

Sury and Hoffmanngreported m.p. 268-270', Mariella, Clatter, and Ebnerloreported m.p. 283'. Anal. Calcd. for CsHl0N2O4: C,51.42; H , 4.80; N , 13.33. Found: C, 51.47,51.63; H,5.09,4.95; N, 13.30, 13.54. The infrared spectrum of X (KBr disk) showed the following significant bands: 3250 cm.-l (s), 1720 em.-' (s), 1680 em.-' (8). There was no band near 2260 em.-'. Hydrolysis of VI11 in Concentrated Hydrochloric Acid.-The hydrolysis of 20 g. (0.105 mole) of VIII in 100 ml. of concentrated hydrochloric acid was carried out in a similar manner to the hydrolysis of VII. The product ( X ) weighed 11.0 g. (42%). After repeated recrystallization from water it melted a t 283.0284.5'. Anal. Calcd. for COHloN2O4: 51.42; H , 4.80; N, 13.33. C, Found: C, 51.23; H , 4.73; N, 13.54. The identity of the product obtained from VI11 with that obtained from VI1 was further shown by the identity of the X-ray patterns14J6and infrared spectra. Hydrolysis of IX in Concentrated Hydrochloric Acid.-The hydrolysis of 6.3 g. (0.033 mole) of I X in 50 ml. of concentrated hydrochloric acid was carried out in a similar manner t o the hydrolysis of VII. The product ( X ) weighed 3.0 g. (43%) and The infrared spectrum (KBr disk) was melted at 283-284'. identical to that of the products obtained from VI1 and VIII. Potentiometric Titrations.-A Beckman Model H pH meter with glass and calomel electrodes was used for the titration with approximately tenth normal sodium hydroxide. The pK,' (pH at half-neutralized point) of I X was 8.6. Titration of X showed that it was an acid weaker than I X , but it was not possible to obtain a definite value due, probably, both to the weakness of X a s an acid and to the fact that it is diprotic. Reduction of X with Lithium Aluminum Hydride.-To a 2000ml. three-necked flask fitted with a mechanical stirrer was a t tached a Soxhlet extractor containing a glass thimble with sintered glass bottom. In the flask were placed 15.2 g. (0.4 mole) of lithium aluminum hydride and 1200 ml. of tetrahydrofuran (dried over sodium hydride and distilled). The thimble contained 21.0 g. (0.1 mole) of X . The mixture was refluxed, with constant stirring, for 26 hr. This period of time was more than sufficient for complete extraction of X from the thimble. The reaction mixture was then cooled in an ice bath and 100 ml. of water was added slowly. Stirring was continued for 20 rnin. after addition of the water. The mixture was filtered, the solid washed with tetrahydrofuran, and the combined filtrate and washings acidified to congo red with concentrated hydrochloric acid. The lower (aqueous) layer was separated and evaporated to dryness a t reduced pressure. The residue was dissolved in 25 ml. of water and basified with solid potassium hydroxide. EX; traction with four portions of chloroform followed by removd of the chloroform a t reduced pressure yielded a residue which was distilled twice i n vacuo through a 350 X 14 mm. Vigreux column to give 2,8-diazaspiro[5.5]undecane(XI). The fraction boiling a t 73.5-74.5'/0.8 mm. weighed 4.60 g. (30%); lit.,9 b.p. 117119'/11 mm. Anal. Calcd. for CwHlsNz: C, 70.07; H, 11.77; N , 18.16. Found: C.69.98: , H. 11.82: N,18.00. . The picrate was formed in methanol and recrystallized from water. It decomposed a t 251'. Sury and Hoffmanns reported decomposition a t 233-234". Anal. Calcd. for CZ~HZ~N~OM: C , 41.18; H, 3.95; N, 18.30. Found: C,41.22; H,4.21; N, 18.00. The infrared spectrum (liquid film) showed maxima a t 3350 em.-' and 1650 em.-' (w). The nuclear magnetic resonance spectrum (102 mg./ml. in carbon tetrachloride) showed chemical shifts (p.p.m. downfield from tetramethylsilane as internal standard) of 2.64 (a-protons), 1.39 (p- and ?-protons), and 1.14 (N-protons); intensity ratios 4:4: 1. For comparison, the spectrum of redistilled piperidine was determined a t a similar concentration; the chemical shifts
(14) The X-ray patterns were kindly obtained by Dr. Arthur L. Draper and Mr. Richard L. Garner using a Phillips X-ray diffraction unit with Cr K a radiation. We desire to express our gratitude for these, deterrninaI tions. (15) There was ,some diffiqulty oonnected with this. Most commonly, the infrared spectrhm of the product showed a strong band at 795 om.-' (a-form). Occasionally, a product lacking this band, and having a different X-ray pattern, was obtained @-form). These were shown to be crystal modifications by the fact upon seeding a supersaturated solution of the a-form (maintained a t 40") with the &form, the latter form crystallized.

FEBRUARY, 1963

MICHAEL-TYPE REACTIONS ~,CY,W,W-TETRANITROALKANES WITH 339

ture was filtered shortly before and shortly after the estimated time required for half-completion af the reaction (as determined by preliminary experiments). The product ( I ) from each filtration was washed with ethanol, dried, and weighed. The filtrate was maintained a t 25" until the reaction was essentially complete and the product removed by filtration, washed, dried, and weighed to obtain the ultimate yield. A graphical interpolation was then used to determine the time required to obtain half of the ultimate yield.

were 2.72, 1.48, and 1.18; intensity ratios 4:6: 1. The chemical shift of the N-proton was very dependent on concentration; a t 3 about 400 mg./ml. it was superimposed on the j- and y-proton resonances in the spectra of both compounds. Effect of Solvent on the Cyanoethylation of Acety1acetone.A mixture of 9.8 g. (0.1 mole) of acetylacetone, 10.7 g. (0.2 mole) of acrylonitrile, and 60 ml. of the solvent was placed in a water bath thermostatted a t 25.0 f 0.1'. After thermal equilibrium had been reached, 10 ml. of triethylamine was added. The mix-

Michael-type Reactions with a,cu,w,w-Tetranitroalkanesl

HENRY FEUER, GERDLESTON,~ ROBERT MILLER,^
Receiioed September 17,1962
Michael-type reactions of 1,1,4,4tetranitrobutane (T), 1,1,5,5-tetranitropentane(11), and I,l,b,g-tetranitrohexane (111)with a molar excess of acceptor, such m acrylic acid, methyl acrylate, acrylonitrile, acrolein, methyl vinyl ketone, methyl vinyl sulfone, and nitroethylene, afford a,w-bisadducts in each case. Either the free a,a,o,w-tetranitroalkanes their a,w-bismethylol derivatives can be employed, the latter being demethylolated or in situ. The reaction of the disodium salt of 1 1 with 2-bromo-2,2-dinitroethyl acetate and with 2-nitrobutyl 1 acetate leads respectively t o 1,1,3,3,8,8,10,10-octanitrodec~ne X ) and 3,5,5,10,10,12-hexanitrotet,radecane (X (XXII). The dibasic acids 4,4,7,7-tetranitrodecanedioic acid (XI) and 4,4,8i8-tetranitroundecanedioic acid (XV) are converted, consecutively into their respective acid chloride, diazide, diisocyanate, methyl carbamate, diamine, and diacetamido derivatives.
AND

ARNOLD NIELSEN T.

Department of Chemistry, Purdue University, L a k q e t t e , Indiana

Michael addition reactions of mononitroalkanes have been studied extensively and numerous examples are known.a Fewer examples of additions of 1,l-dinitroalkanes have been r e p ~ r t e d , ~ - ~ the scope of this rebut action appears potentially equal to that employing mononitroalkanes. The addition of methyl acrylate to dipotassium 1,1,3,3-tetranitropropanesappears to be the only reported example of the Michael-type reaction with an a,a,w,w-tetranitroalkane. In this instance, only monoaddition took place and methyl 3,3,5,5-tetranitropentanoate was obtained. I n continuation of our study of the Michael-type addition with nitroalkane~,"J-~3 are now reporting our we such findings with cu,cu,w,w-tetranitroa1kanesi4 as 1,1,4,4(11), and tetranitrobutane (I),1,1,5,5-tetranitropentane 1.1,6.6-tetranitrohexane(111), leading to a,w-disubstituted products. Two procedures were used with such Michael acceptors as methyl acrylate, acrylic acid, acrylonitrile, acrolein, methvl vinyl ketone, methyl vinyl sulfone, and nitroethylene. Procedure A.-The a,w-bismethylol-a,a,w,w-tetranitro alkane^'^ were employed, the tetranitroalkanes being generated in situ by demethylolation in the presence of catalytic amounts of sodium hydroxide. In the
(1) Presented before the Division of Organio Ghemistry at the 142nd National Meeting of the American Chemical Society, Atlantic City, N. J., September, 1962. (2) Taken, in part, from the Ph.D. theses of Gerd Leston (1952) and Robert Miller (1958), Purdue University. (3) For a thorough survey of the literature, reference is made to E. D. Bergman, D. Ginsburg, and R. Pappo, Org. Reactions, 10, 179 (1959). (4) L. Hersog, M. H. Gold, and R. D . Geckler, J . A m . Chem. Soc., 73,749 (1951) (6) K. Klager, J . Ore. Chem., 16, 161 (1951). (6) H. Sheohter and L. Zeldin, J . A m . Chem. Soc., 73, 1276 (1951). (7) M. B. Frankel. J . Org. Chem., 23,813 (1958). (8) S. 9. Novikov I. S. Korsakova, and Rf. A. Yatskovskaya, Dokl. Akad. N a u k , S S S R , 118,954 (1958); Chem. Abstr., 62, 12792 (1958). (9) K. Klager, J. P . Kispersky, and E. Hamel, J . Ore. Chem., 26, 4368 (196 ). (10) H. Feuer and C. N. Aguilar, J . Ore. Chem., 23, 607 (1958). (11) H. Feuer and R. Harmetz, ibid.. 26, 1061 (1961). (12) H. Feuer and R. Miller, ibid.. 26, 1348 (1961). (13) H. Feuer, E. H. White, and 9. M. Pier, i b i d . , 26, 1639 (1961). (14) H. Feuer, E. Colwell, G . Leston, and A. T. Nielsen, ibid., 27, 3598 (1962).

equation below, Procedure A is illustrated by the reactions of 2,2,5,5-tetranitro-1,6-hexanediol (IV) and 2,2,6,6-tetranitro-l,7-heptanediol with methyl acry(V) late to yield dimethyl 4,4,7,7-tetranitrodecanedioate (VI) and dimethyl 4,4,8,&tetranitroundecanedioate (VII), respectively.
KO2
HOCH&(

x2 0
I

NOS NO2 IV. n = 2 V. n = 3 NO2

CHZ),-&CHZOH

+ 2 CH2=CHC02CH3 --+
NO*

CH~OZCCHZCHZ~-(CK).-&-CH~CHZCOZCH~ 2 C H 2 0 I
NO2

NO2

VI. n = 2 VII. n = 3 XIV. n = 4

Procedure A

Procedure B, which was very effective with more reactive or base sensithe acceptors such as methyl vinyl ketone, nitroethylene, or acrolein, employed the a,w,w-tetranitroalkanes with a trace of basic free a, catalyst. Procedure B is illustrated by reactions of 1,1,4,4-tetranitrobutane(I) and 1,1,6,6-tetranitrohexNO2 H -(CH2)nNO2 I. n = 2 11. n = 3 111. n = 4

AI

NO2

KO*

bH + 2 CHz=CHCOCHs --+ 1
I CH.&OCHzCH?C-( I
NO2

NOz
CHZ),-C-CH~CH~COCH~ NO2

NO2 VIII. n = 2 IX. n = 4

Procedure B

I I