4684
VOL. 26
NOTES
be known. In order to determine this, very dilute sodium hydroxide solutions of known concentrations were d e up in the solvent medium, and the pH measured. At the midpoint in the titration, when (RS-) = 0, Equation (1) reduces to pH = p K , as is well known. The experhental value of the pH at the midpoint in the t i h t i o n was 11.9. For numy reasons, Equation 1 is not exactly applicable in t h e e circumstances and the dif€erence of 1.5 pK Units comprises not only the intrinsic d3erence between the acid strengths in water and 44% ethanol, but also the quantitative consequences of several corrections and approximations. In any case, the difference of 1.5 pK units may wefully be regarded as a correction factor by which the p K values determind in the mixed aqueous organic medium can be transposed to wa&r. To check the applicability of the correction, ethanethiol was titrated both in water and in 44.4% alcohol, and the pK values found at the midpoint of the titration were 10.0 and 12.1; the difference is 1.5. The pK value for e t h e t h i d itl water acmrds well with that found by Yabroff' and Danehy.a The value for hexanetbiol is a little lower than that estimated from the data of Yabroff.
phines. The present work extends deoxygenation
by phosphines to aromatic d t r o s o derivatives. Triphenylphospbine oxide (11. R = C a b ) is obtained in good yield from equimolar quantities of triphenylphosphine and o-dinitrosobenzene in a hydrocarbon solvent. Benzofurazan (I), presumably formed Simultsneously, waa not isolated; however, after StSam distillation of a similar reaction mixture in ethanol both products are isolated in very good yield and exem triphenylphosphine,if present, is recovered. Initial attack by a phosphine at oxygen rather than nitrogen of the nitroso group is not established but presents a plausible explanation for the formation of the product. Intermediate colored materials may be phosphonium zwitterions and/or quaternary bases similar to those postulated for the reaction of other polarizable double bonds and ph~sphines.~
r
+
L
r
l+
1
AcknolpEedgment. This research was supported by the Petroleum Research Fund administered by the American Chemical Society. Grateful acknowledgment is hereby made to the donors of said fund. a grant from
I
I1
Substitution of tri-n-butylphosphine or triethyl phosphite for triphenylphosphine affords benzoDEPARTMENT OF CEEMISTRY furazan in comparable yields. Apparently tri-nOKLAHOMA STATE UNIVERSITY butylphosphine is a more powerful reagent than STILLWATER, Om. triphenylphosphine since the former but not the latter converts, 1,2ainitromnaphthalene to naphthofurazan. In a similar reaction 4,7diphenylfuraeano[d]pyrazine (111) is obtained from 3,6diDeosygeaation of Aromatic o-Dinitroeo phenyl~,5-diai~sopyridaaine. Derivatives by Phasphines' In the reaction of o-dinitrobeneene and triphenylphosphine neither o-dinitrosobenmne nor J. H.BQYEB* AND S. E.ELLZEY, JB.* bermfumean has been detected; however, good yield of triphenylphosphine oxide and a trace of an unidentified solid are isolated in addition to &oxy derivatives,' amine oxides,' C-nitroso starting materials. o-Dinitrobenzene and triethylderivatives,' nitrones,' isonitrones: nitrous oxide: phosphine are reported to form an unstable 1 : l and nitrogen dioxide' have been partially or com- adduct, m.p. 140" dec., in anhydrous solvents.* pletely deoxygenated by trialkyl- or triarylphos- Color formation takes place in the attempted re-
.$$) CBHI
CeH5
(1) Financial support by the O a c e of Ordnance Research, United States Army, under contract No. DA-01-0040RD699 is gratefully acknowledged. (2) Preaent address: American Chemical Society, Petroleum Research Fund, Washington 6, D.C. (3) Present adddress: Southern Regional Research Laboratory, New Orleans, La. (4) L. Horner and H. Hoffmann, Angm. C h n . , 68, 473 (1956). (5) L. Horner and E. Jurgens, Chetn. Ber., 90,2184 (1957). (6) H. Staudinger and E. Hauser, Helv. Chint. Acta, 4, 861 (1921). (7) L. Horner and W. Jurgeleit, Ann., 591,138 (1955).
N\
"
+
(n-C,Hs)3P0
NO CsHs
I11
RC-CR'
/ / \
N
R3P
.,
RC-CR' ~
1 %
N 0 ''
IV ( 8 ) L. Horner and K. Xlupfel, Ann., 591, 60 (1955).
NOVEMBER
1961
4685
KOTES
action of nitrobenzene and triphenylphosphine, but only starting materials are recovered, in sgreement with previous work8 which employed triethylphosphine. In contrast to the reactivity of dinitroso compounds, furoxans (IV) are unreactive toward tri-nbutylphosphine. EXPERIMENTAL'
yields of product
togetner with recovered starting materinl.
WSH recovered snd an unidentified product in low yield, m.p. !B36-%8" dec., and triphcnylphosphine oxide (4870) were obtained on refluxing odinitrobenzene and triphenylphosphine in benzene under nitrogen for 1 hr. Nitrobenzene was unreactive toward triphenylphosphine in boiling ethanol for 18 hr. although the solution b e black. t Diphenylfuroxan and bisp,p'-dichlorobenshydrylfuroxan were each unreactive toward tri-n-butylphosphine in boiling ethanol for 24 hr. and were recovered (90% and 57%, respectively).
Starting material
Benzofurman. On mixing 1.36 g. (10.0 mmoles) of m C~IUXSTE DEPAETMFINT Y dinitrosobenzene and 2.86 g. (10.9 mmoles) of triphenylTULANE UNXVERSITY phmphine in 25 ml. of 9570 ethanol the solution became NEWO R ~ A N18, S LA. warm and turned red. After refluxing for 30 min., steam distillation afforded 0.68 g. of benzofurazan, m.p. and mixture m.p. 52-53',10 insoluble in the first 200 ml. of distillate. An additional 0.13 g., m.p. 47-48", obtained on extracting the distillate with ether and evaporation of the ether layer, Condensation of Mesityl Oxide and Methyl gave a total of 0.81 g. (6870) of b e n z o f m a n which absorbed Vinyl Ketone in the infrared from a potassium bromide diRk at YOT)] : 6.14 (56.3), 6.54 (43.0), 6.89 (51.9), 6.97 (52.7), 7.26 (56.0), 7.36 (44.0),8.75 (47.2), 9.56 (70.91, 9.82 (51.51, 9.91 (39.01, EBNSTD. BERGUNNAND P.BRACEA 11 41 (31.11, 11.83 (40.9), 13.24 (18.3), 13.51 (14.8), 14.37 (43.9). An authentic sample of benzofman, m.p. 53-64', Received February 10,1961 from o-dinitrosobensene and hydroxylamine gave identical absorption except for a band in the region of 14.37 p. A concentrated benzene extract of the distillation residue In a recent communication' we have reported was treated with ligroin to precipitate triphenylphosphine that the condenation of mesityl oxide and methyl oxide, map. and mixture m.p. 156-158',11 2.65 g. (91.7%). It8 infrared absorption from a potassium bromide disk was vinyl ketone in the presence of sodium bpentoxide identical with that obtained from a similar examination of an leads to piperitenone {I). Although the physical authentic specimen. In an experiment with 10.0 moles of u-dinitrosobeneene, 21.8 mmolea of triphenylphosphine, and r d u x time of 5 hr., 65% of the exceea (11.8 mmolea) phosphine was recovered and triphenylphosphine oxide w w obtained quantitatively. y o -0 *c A reaction in anhydrous petroleum ether (b.p. SCrsS0) I I1 I11 IV instead of 95% ethanol gave a 66% yield of triphenylphosphine oxide after 1.5 hr. a t room temperature followed by heating on a steam bath for 10 min. A tarry residue with the constants of the product showed some deviations odor of benzofurazan remained after evaporation of the fil- from the values reported for natural piperitenone by NavesZ and we had based our contrate. Substitution of tri-n-butylphosphine for triphenylphoa- clusions on two observations which appeared to phine afforded benzofurazan in 6270 yield after equimolar be unambiguous: the catalytic transformation to quantities of phosphine and o-dinitrosobensenewere refluxed in 95% ethanol under nitrogen for 1 hr. In a similar experi- thymol (11) and the fission to acetone and 3ment benzofurazan waa obtained in 58% yield from odinitro- mpthylcyclohex-%none (111). sohenaene trented with a molar exceas of triethyl phosphite Naves and Conid were not able to reproduce in refluxing ethanol under nitrogen for 12 hr. our results and claimed that the only product NaphWlofuraunt, m.p. and mixture m.p. 77-78'," WM formed under the conditions described was the obtained in 65y0 yield from l,%dinitrosonaphthalene and tri-a-butylphosphine in equimolar quantitiea in refiuxing well known' self-condensation product of mesityl ox95% ethanol under nitrogen for 2 hr. Triphenylphoephine ide, viz. isoxylitone (IV). In a later communication,* f d e d to deoxygenate l,%dinitrosonaphthalene in refluxing the French authors, reported that after a private ethanol for 5.5 hr. communication from Dr. Ueda: they reinvesti4 , 7 - D i p h m y l j u r a z o [ dlpyridatinc, m.p. and mixture m.p. 193-195', %as obtainedin 360/oyield from3,6-diphenyl4,5-dinitrosopyridazine and tri-n-butylphaephine (molar (1) E. D. Bergmann and P. Bracha, J. Org. Chem.,24, excess) in refluxing 95% ethanol under nitrogen for 1 hr. 994 (1959). The crude product, isolated by filtering the reaction mixture (2) Y.-R. Naves, Helu. Chim. Acta, 25, 732 (1942). Y.-R. diluted with water, waa recrystallized from aqueous acetone. Navea and G. Papazian, Helu. Chirn. A&, 25, 1032 (1942). Equimolar quantities of reactants after 6 hr. gave lower (3) W. Kuhn and H. Schinz, H e b . Chim. A&, 36, 161 (1953). (4) T. Suga,Bull. Chem. SOC. Japan, 31,569 (1958). (9) Melting pointa are uncorrected. We are indebted to (5) Ch. Balant, Ch. A. Vodoa, H. Kappeler, and H. Mr. R. T. O'Connor, Southern Regional Research LaboraSchinz, Helv. Chim. A&, 34, 722 (1951). tory, New Orleans, La., for infrared data. (6) Y.-R. Naves and J.-M. Conia, BuU. soc. chim. France, (10) T. Zmcke and P. Schwarz, Ann., 307,28 (1899). (11) M. A. Greenbaum, D. B. Denney, and A. K. Hoff- 378 (1960). J.-M. Conia, and Y.-R. Navea, Compt. rad., 250,356 (1960). mann, J. Am. Chem.SOC.,78,2563 (1956). (7) J.-M. Conia, Compf.Rend., 240, 1545 (1955). (12) A. G. Green and F. M. Rowe, J . C h a . Soc., 111, (8) Y.-R. Navea and J.-M. Conia, Compf. rend., 251, 612 (1917). 1130 (1960). (13) E. Durio, Gam. chim. ifd.,,61,589 (1931).
p
