(AWN nu ...u .u AWN. .. inc. any a»... mi _.....H.... mm mm“. i. .YV 0“ Join“ N UK fin- It m. m N! (I n o All «MU $V~ a . Q‘DII lb!“ 8 “HR...” 3 1a ‘ "n an. mm Iv .1... mus “mm A“... W.‘ P u uh!» A5 H:5:Egg;:15.2:: 8 '\ 1’3 L IBRA R Y THES‘S . - . A Mchnng ‘tatc Unix-“city CYCLOBUTADIENE META-QUINONES by Anthony D3\Wolf A THESIS Submitted to: Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1968 Abstract Cyclobutadiene Meta—Quinones by Anthony D. Wolf Some approaches to the synthesis of several cyclobutadiene meta— quinones are considered. The synthesis of 2,4-di-pfmethoxyphenylcyclobutadiene meta-quinone (XXVIII) is described and, a structure proof based on both physical and chemical evidence is presented. XXVIII ACKNOWLEDGEMENT The author would like to express both his thanks and appreciation to Dr. Donald G. Farnum for his guidance and friendship throughout the course of this project. to my parents Table of Contents Acknowledgement Introduction Results and Discussion Summary Experimental 1. Preparation of 2,4-diphenyl—3-phenylacetoxycyclobutenone (XXX) 2. Conversion of 2,4-diphenyl-3-phenylacetoxycyclobutenone (XXX) to 2-bromo-2,4—diphenyl-3—hydroxycyclobutenone (XXXI) 3. Conversion of 2-bromo-2,4—diphenyl—3—hydroxycyclobutenone 10. ll. (XXXI) to 2,3-dihydroxy-2,4-diphenylcyclobutenone (XXXV) . Reaction of 2-bromo—2,4-diphenyl-3—hydroxycyclobutenone (XXXI) with triethylamine . Preparation of pr,N—dimethylaminOphenylacetic acid (XXXVIII) . Conversion of pr,N-dimethylaminOphenylacetic acid (XXXVIII) to pr,N—dimethylaminOphenylacetyl chloride hydrochloride (XXXIX) . Condensation of pr,N—dimethylaminophenylacetyl chloride hydrochloride (XXXIX) with triethylamine in ether . Condensation of "squaric acid" with N,N—dimethylaniline . Preparation of the Purple Product Reaction of the "Purple Product" with trOpilidene Reaction of the "Purple Product" with 2,4-difipfmethoxyphenyl- 3-hydroxycyclobutenone (LIII) Page 25 26 26 27 27 27 28 29 30 30 31 32 33 Table of Contents (Cont.) Page 12. Preparation of 2-bromo-2,4-dimethyl—3—hydroxycyclobutenone (LVII) 33 13. Reaction of 2—bromo-2,4-dimethyl-3-hydroxycyclobutenone (LVII) with trimethylamine 34 Bibliography 35 Introduction During the past ten years intensive investigation into the synthesis and isolation of cyclobutadiene and its derivatives has been conducted. The challenge to chemists arose out of the corollary to the Hfickel prediction that systems of 4n n electrons (n=0, 1, 2 etc., on a fully conjugated monocyclic n electron system) would possess reactivity on a par with or perhaps greater than their Open chain analogues. This prediction seems to have been realized in the instance of cyclobutadiene (I) and its derivatives.1 H H I I \ H H Cyclobutadiene itself would be expected to be more reactive than the Open chain analogue 1,3-butadiene. This is reasonable since according to Hfickel theory it should possess no special electronic stability, while at the same time, however, the molecule possesses considerable strain energy, as a result of constraining four sp2 orbitals, which normally have bond angles of 120°, to a situation in which they have a strain of 30% per carbon. EXperiment has born out these expectations. Neither cyclobutadiene nor its derivatives has been amenable to synthesis with classical diene methods. Only with new synthetic methods could these systems be approached. Pettit and coworkers have suggested that cyclobutadiene is a highly reactive intermediate of finite lifetime.2 More recently Freedman and Sandel have presented a detailed physical study, in which tetraphenylcyclobutadiene (II) was postulated as an intermediate in the pyrolyses reactions of (4-bromo- l,2,3,4-tetraphenyl-cis,cis-l,3—butadienyl) dimethyltin bromide (III).3' C H C H C H C H 6 5\\\ 6 5 6 5 6 5 l l / \ C H c H CH3 I Br II CH3 III Also of interest is the dipositive ion cyclobutadiene dication (IV). This system possess the necessary complement of 2n electrons for a closed shell configuration and would therefore be expected to be aromatic.. H H (3: E.) H H IV Cyclobutadiene dication itself has not yet yielded to synthesis. However, the following derivatives, IV and VI have been reported by Freedman C6H5 C6H5 C6H5 OH 3' \ I ‘ I+9 {+9 ‘\\ ‘\\ C6H5 C6H5 HO C6H5 V VI and Farnum respectively.“"5 The report on the synthesis of V,“ has been shown to be in error. X—Ray studies show that rather than the hexachlorostannate salt of V, chlorotetraphenylcyclobutadienyl pentachlorostannate (VII) was produced.6 C H c H 6 5 sn-Cls /’ C6H5 C6H5 VII As aromatic systems 30, benzene is a pillar, being the most well known in the 6H electron class. As cyclobutadiene is to cyclobutadiene dication in structure, so benzene is to benzene dication VIII. VIII Benzene dication itself has not been made. This is understandable in view of the fact that there is a net loss of electronic stability in going from the aromatic benzene system, to the antiaromatic dication.7 It is noteworthy and interesting in relation to benzene dication, that orthgfbenzoquinone (IX) and parafbenzoquinone (X) do exist and are stable, though reactive. One can write resonance contributors XI and XII for 0 IX 1: x 0’ 0' + + + 0_ _ XII XI structures IX and X, which involve the benzene dication. In view of the predicted instability of benzene dication, contributions from ionic resonance forms XI and XII to their respective quinones, should be small at best. However, resonance contributors XIII and SIV might be expected to be more important, since a full complement of six n electrons is maintained. Thus 0. 0. XIII XIV these quinones exhibit some of the stability associated with 6n electron systems, rather than the instability expected for An electron systems. The great facility with which these systems undergo free radical hydrogen atom addition indicates their tendency to go to an authentic 6n electron system. Compared to orthgf and pagafbenzoquinones, the still unprepared isomer mgtafbenzoquinone is an anomaly. On attempting to write down its structure one becomes immediately aware of the uniqueness of this species in comparison to both its 2££h9_and para isomers. Once can write either a dipolar or triplet diradical species such as XV or XVI as possible structures 0 0 XV XVI for this system. Consequently one can imagine that if formed, a meta—quinone might be reactive. Cyclobutadiene like benzene, has its series of quinone isomers, fittingly entitled cyclobutadienoquinones. Two systems are possible, namely, cyclobutadiene ggghgrquinone and cyclobutadiene mgtgfquinones. Cyclobutadiene orthgrquinone (XVII), the parent member in the series of l,2—cyclobutadienoquinones has not as yet been prepared.1 H\ //0 §§O XVII However several derivatives have been reported. The first member in the series to be reported was phenylcyclobutadienoquinone (XVIII),8 while later members included the pOpular 3,4-dihydroxy—l,2—cyclobutadienoquinone or H i0 H /O % C6H5 O HO O XVIII XIX "squaric acid" (XIX).9 With benzoquinone it was thought that resonance forms XI and XII contributed negligibly to the stabilization of these systems since the n electron density Operating over the ring would be reduced. The system would thus be destabilized relative to benzene. This same effect in cyclobutadiene Egghgfquinones however, would only contribute to the system's stabilization by reducing the n electron density so that the system resembles the cyclobutadiene dication. Resonance structures such as XX, XX which involve a dication, would then be expected to be much more important in cyclobutadiene orthgfquinones than in benzene ggghgf and pagarquinones. Cyclobutadiene has a 1,3-quinone system, cyclobutadiene meta: quinone, which is analogous to benzene metafquinone. This system must have either a singlet ground state as implied by dipolar resonance contributors such as XXI, or a triplet diradical ground state such as XXII. As in the case of mgtafbenzoquinone, it is not possible in terms of simple resonance theory 0 O ./ {/ 04 ' + \\ XXI XXII to predict which structure the molecule would actually have. The first cyclobutadiene mgggfquinones were reported in 1965 by several German groups. Treibs and Jacob reported that condensations of "squaric acid" with activated pyrroles gave rise to systems having the meta-quinone skeleton.10 This paper was soon followed by several others, HO //O | + / \ ________;> meta-quinone N HO/ %0 1'1 11-14 which added to a growing list of known meta-quinones. No mention was made however, by any of the authors that these were indeed the first cyclobutadienefmgtarquinones. Furthermore, no rigorous structure determination to verify that the carbon skeleton was that of a mgtaf quinone was presented, so that several structures have been assigned to these systems. For example, Treibs and Jacob refer to the condensation products of pyrroles and squaric acid as having either structure XXIII or a tautomer XXIV.10 Sprenger and Ziegenbein favor resonance contributor XXV.11 XXIII XXIV XXV The object Of this research was to prepare several cyclobutadiene meta-quinones and to establish their structure. Results and Discussion The goal of this project was the synthesis of the cyclobutadiene meta-quinones illustrated in Scheme 1. Each of these systems will be considered separately in Sections A, B, C and D. Scheme 1 / \ | + _ / \\ XXVI ”’ ‘1\R = C6H5— CH3 N/ CH3r' ”’ R 0 / _ 3 _ . l + <: R-(CH3)2N—C5HA» 3> I + _0 CH3 -0 R _p ,,’ XXIX \\ / ,—CH3 XXVII / OCH3 / \ / 0 // | . 0 ,,’ \ / O XXVIII l C Section A Approaches to the Synthesis of 2,4—dipheny1cyclobutadiene meta-quinone (XXVI) The route considered for the synthesis of meta-quinone XXVI was that shown in Path A. Condensation of phenylacetyl chloride with triethylamine Path A C H C H O 6 5 \ 6 5 / Br2 C6H5CH2COC1 + (CH3CH2)3N ___> ' E CCl ) H 4 / Br H HO 6 5 C6HS C6H5CH2—C=O XXX XXXI (CH3CH2)3N 0 C H 0 C6H5 0 C6H // 6 5 / C6H5\ 0 + 1’ warm | 1 I l / ‘vacuum J36H5 \J C6H5 Br 0 J _0 6H5 H0 /+ —CH2CH3 _ j -CH2CH3 - 6H5 CH3CH2 I CH3-CH2 H CH CH CH C 2 3 2 3 (CH3CH2)3+NH Br“ XXXIV XXXIII XXXII produced phenylketene trimer XXX.15 This compound was then brominated with a solution of Br2 in CC14, hydroxycyclobutenone (XXXI), a white substance when pure, which turns red after to give the reactive 2-bromo-2,4-diphenyl-3— a short time in air.5 It was hOped that from bromoketone XXXI and triethylamine, -10- dipole XXXIII might be produced (salts XXXII and SSSIV also seemed to be likely products). From XXXIII gentle warming might lead to meta-quinone XXVI by loss Of (C2H5)3N. On addition of (CH3CH2)3N to an ethereal solution of bromoketone XXXI a precipitate forms with the simultaneous appearance of a transient pink color. A comparison of the ir spectrum Of the crude reaction product with those of the alcohol XXXV5 and triethylammonium bromide showed that the C6H5 490 OH HO C6H5 XXXV product was composed primarily of these two substances. Work up of the reaction mixture with non-nucleophilic solvents produced a 50% yield of alcohol and in general a greater than 50% yield of (CH3CH2)3NHBr. NO other products were isolated. It was not possible in this attempt to isolate any of the salts XXXII, XXXIII and XXXIV. Only alcohol XXXV and triethylammonium bromide could be isolated. -11_ Section B Synthesis of the "Blue Product" yggéfquinone XXVII with N,N-dimethyl substituents on the ring might be more stable than its counterpart XXVI, because of the possibility for delocalization of the positive charge to the nitrogens as shown in XXXVI and XXXVII. There are, of course, numerous other contributing forms. Furthermore, the report by Sprenger and Ziegenbein13 that N,N-dimethylaniline CH3 +N/ \/ XXXVI and "squaric acid" give rise to a blue substance having structure XXVII, added more incentive to try the synthesis. An alternate route to the blue substance would Offer more credence to the prOposed structure. The synthesis of the "Blue Product" involved the reaction sequence shown in Path B. Reaction of pfnitrophenylacetic acid with formaldehyde, under reducing conditions produced the amino acid XXXVIII.16 Carbonyl absorption at 5.90 p verified the presence of a carboxyl group in the molecule. The nmr spectrum exhibited singlets at T 7.14 (area = 5.7) and 6.53 (area = 2), -12.. <3 i H H CH CH CH3\|/CH3 3\\$,,, 3 5% Pd/C A PC15 > + H2C0___H2 /, // CH CO H CH2C02_ CHZCOCl 2 2 XXXIX XXXVIII (CH3CH2)3N “' CH / 3 V CH3-” / ‘\ ” 0 47 Br2 ' H "Blue Product" .(E CHC13 CH3\‘ I | \\ DL_<: :>_C=O // / CH3 /N \ CH3 CH3 XL which can be assigned to the N—methyls and benzylic protons, respectively. .An A2B2 pattern centered at T 3.18 (Av = E; CH2C12 | ~€> H // Br Produc I HO C ZCOCl .// O=C l ' \ CH2 L_- __ OCH3 OCH3 XLVIII OCH3 XLVII with (CH3CH2)3N produces prmethoxyphenylketene trimer (XLVII). This trimer unlike its analogue in the phenyl series XXX, upon treatment with Br2 in CHZCl2 produces instead Of the expected bromoketone XLVIII, a purple substance which will be referred to as the "Purple Product".19 Crystallization from acetonitrile-benzonitrile produces beautiful delicate purple needles. Analysis of the material was consistent with the formula C18Hl404 and its spectral properties were very similar to those of the "Blue Product". Electronic spectra are indicative of a species having a highly conjugated _17_ H-electron system undergoing a facile H-H* transition as shown by the magnitude of the extinction coefficients. xCH2c12 ACH2C12/CF3COZH max (mu) 6 max _51_ 536 1.41 x 105 529 1.58 x 105 500 4.15 x 10L+ 348 9.73 x 103 The rather simple ir spectrum indicates that the substance might be highly symmetrical. The 6.1 u and 6.3 u bands with no carbonyl absorption less than 6 u, are characteristic of a highly conjugated ketone. The substance gives a beautiful nmr spectrum in deuterochloroform- trifluoroacetic acid, quite simple in appearance. A sharp singlet at T 5.97 (area = 3) is characteristic of a methoxy grouping in the molecule. An A2B2 pattern centered at T 2.16 (Av = 1.35 ppm, J = 9 Cps, area = 4.1) is characteristic of a pfdisubstituted benzene, in which a deshielding influence must be causing the low field resonance of the aromatic protons. Several structures can be written for the "Purple Product", which are analogous to those written for the "Blue Product", that might arise from the bromination of trimer XLVII, and are compatible with the formula 018H1804° These include XXVIII and XLIX — LII. Again structures such as LI and LII OCH3 OCH3 l f N XLIX L _1g_ OCH 0 LI LII can be ruled out on the basis of nmr evidence. The simple spectrum Observed would not be predicted for these structures. The bicyclobutanedione L, is ruled out by both the electronic and ir spectra. The molecule does not possess enough conjugation to expect a band at 536 mu in the electronic spectrum. Furthermore 6.1 and 6.3 u are too long wavelength to be a cyclo- prOpanone. Only two structures need to be considered further namely XXVIII and XLIX. On the basis of an esr spectrum determined by Dr. D. H. Geske at Cornell University the diradical XLIX can be eliminated since the "Purple Product” did not give rise to an esr signal. Although at this point all the physical data available pointed to XXVIII as the structure for the "Purple Product", it was considered desirable to obtain some chemical evidence for the presence of the four-membered ring. On this basis, reduction of the "Purple Product" with tropylidene might give rise to dienone LIII.15 This would clearly demonstrate that the carbon // V -19_ skeleton had remained intact in the conversion of trimer XLVII to the "Purple Product". Treatment of the "Purple Product" with tOpylidene in acetic acid, acetic anhydride—fluoroboric acid mixture, after two hours, produced a white solid on recrystallization from ethyl acetate. Comparison of the data for the white solid and the dione clearly reveals that they are different Species. A possible explanation for the formation of the white solid involves a "dimerization" to give a product having structure LIV. This structure would account for the observed spectra. 0 CH o-c H 3 6 4 6? C6H4OCH3 1 OH CH3OC6H4 . I 67 C6H4-OCH3 LIV The nmr spectrum in deuterochloroform and trifluoroacetic acid exhibited two distinct sharp signals at T 6.20 and 6.17 (total area = 6) characteristic of two different methoxy groups and two distinct A patterns in the aromatic 232 region at T 2.99 (Av = 60 ppm, J = 9 Cps) and 2.80 (Av = .63 ppm, J = 9 cps), characteristic of two different pfdisubstituted benzenes. The ir spectrum with peaks at 2.90 — 5.0 u (br, w), 5.75 p (s) and 6.20 p (s), is compatible with the hydroxycyclobutenone system. Furthermore the suggestion of dimer formation seems reasonable on the basis of the evidence for a similar dimer XLV, in the phenyl series, whose ir spectrum gave similar absorption at 5.80 p (s) and 6.18 H (s). It is also interesting that the "Purple Product" is reformed in 40% conversion from the white solid after ten hours in chloroform-trifluoroacetic acid, as evidenced by the A B2 pattern at 2 T 2.24 (Av = 1.38 ppm, J = 9 Cps) in the nmr spectrum. The dimer formation can be envisioned in the following way. H CH 0C H CH OC H o 3 6 49 3 6 4 ./ + + ‘ + /’ H BF4 o 6H40CH3 Ho C6H40CH3 H o 0 0 CH3OC6H 0 C 3 C6H4 ‘45 g? f? I C6H4OCH3.0H + I —-> H + H C H OCH 0 \t 6 4 3 _ H OCH C6H40CH3 6 4 3 a; H OCH 6 4 3 If a dimer is formed by the route described above, then a reaction between the dione LIII and "Purple Product" in the absence of tOpylidene should also give rise to the dimer. And in fact when the dione and Purple Product are mixed together in acetic acid, acetic anhydride and fluoroboric acid, a solid is formed, which on recrystallization from ethyl acetate gives a white Solid whose ir and nmr Spectra are identical to those of the white solid arising from the tropylidene eXperiment. Neither the dione nor the "Purple Product” alone gave the dimer. On this basis then it is concluded that a dimer has formed in these reactions and has the structure LIV. Furthermore since the "Purple Product" by itself, gives rise to the same substance that arises from the "Purple Product" and another substance having the cyclobuta— dienone skeleton, then the ”Purple Product" must also have the cyclobutadienone CH H0 -21_ skeleton. Therefore, on the basis of a chemical structure proof and on the basis of the physical data presented, it is concluded that the "Purple Product" is 2,4—dijpfmethoxypheny1-cyclobutadiene-meta-quinone. Section D Approaches to the Synthesis of 2,4-dimethy1cyclobutadiene.maLarQUinone (XXIX). An attempt to synthesize meta—quinone XXIX is discussed in this section. The synthetic route followed was similar to that employed in Section A. And is illustrated in Path D. Path D CH 0 3 // CH3CH2COC1 + (CH3CH2)3N -————e> H \. CH3 /C-O CH3CH2 LV LVI Br2/CC14 \/ CH 0 CH 0 CH 0 0 3 a i // (CH ) N // 1’ 3 3 +0113 + l + V ., I u—CHB CH3 0 Br Br CH3 CH3 _0 N_CH - CH3 H0 “H3 - | + Br CH3 CH3 (CH3)3NH LI LVIII LVII 0 CH // + _22- Reaction of prOpionyl chloride with (CH CH2)3N in anhydrous ether produced 3 an orange oil after workup from (CH CH NHCl and the solvent. The mixture 3 2)3 which was composed primarily of methylketene trimer LV and B—lactone methyl— ketene dimer15 LVI, was not further purified. Instead the crude reaction product was diluted with carbon tetrachloride and treated drOpwise with a solution of bromine in the same solvent, according to the method developed by Farnum and Webster.19 A white precipitate formed which was recrystallized from a mixture of ethyl acetate in benzene. The white solid 2-bromo—2,4- dimethylcyclobutenone was identical by nmr and ir to the material reported.20 The transformation involving the conversion of trimer LV to bromoketone LVII is analogous to that conversion in the phenyl series. It- can be envisioned mechanistically as shown in Scheme 2. Scheme 2 Br-Br Br— CH '1 / 0 Br /0 Br 0 CH //0 / / / a ———> ———> _9 G H // H {/ H m 0 CH 0 CH CH Ho /C-CH2CH3 //C-CH2CH3 CH3CH2COBr 0 / 0 5‘ It was hOped as shown in Path D, that treatment of LVII with (CH3)3N would produce the dipole LIX, (though LVIII and LX represent possible products from this reaction). Isolation of LIX might lead to the desired meta-quinone, XXIX, by loss of (CH3)3N with gentle warming in a vacuum. _23- In a typical reaction a sample of bromoketone LVII was placed in a solvent trap cooled to -78°C. Trimethylamine was condensed in the trap until a estimated 10—15 equivalents was present. The suspension was then stirred using a magnetic stirring bar or vibro—mixer. The cooling bath was removed and the sample was stirred until all the (CH3)3N had evaporated and only a powder remained. The nmr of the crude reaction mixture in d -dimethy1 6 sulfoxide produced the spectrum shown in Figure 1. Figure l 7.19 (J = 9 c.p.s.) 6.82 0 8.74 intensity 8'54 + '1' (ppm) —> The signals at T 8.74 and 8.54 are assigned to the alcohol LXI20 produced on solvolysis of bromoketone LVII. Addition of an authentic sample of alcohol to the reaction product served to increase these signals. Addition of bromide to the reaction product, produced two new signals at T 8.48 and CH O 3 6/ 40H 110 CH 3 LXI _24_ 8.33. With time these signals decreased and disappeared while those assigned to the alcohol increased. Thus the bromoketone LVII is readily hydrolyzed under the nmr conditions to the alcohol LXI. Addition of a sample of (CH3)3N+H (from (CH NHCl) to the 3)3 reaction product increased the intensity of the doublet centered at T 7.19, J = 9 CpS. A coublet for (CH N+H is consistent with the spectrum reported 3)3 for this ion at low pH values.21 Yet still remaining to be explained is the product (5) giving rise to absorption at T 8.43, 8.30 ppm (total area = 6) and 6.82 (area = 7.5). Possible structures which come to mind for the species giving rise to these nmr signals include LIX and the desired zwitterion LX. There are inconsistencies with either Of these products, since one might imagine that these species - would also hydrolyze since bromoketone LVII was so readily hydrolyzed. In fact addition of a drOp Of H O to the nmr sample does not noticeably alter 2 the character of the unidentified signals. Nothing further can be said about the identity of the species giving rise to the nmr signals. Isolation of the products has not as yet been feasible. _25- SUMMARY The ultimate goal of this research - namely, the synthesis and structure proof of mggafquinones XXVI, XXVII, XXVIII and XXIX - was achieved only in part. This was in the case of 2,4-difpfmethoxyphenylcyclobutadiene megafquinone XXVIII and perhaps in the case of the dimethylamine analogue. 2,4—Dimethylcyclobutadiene mgtafquinone was not synthesized. The presence of an unidentified material arising in the synthetic route was demonstrated. The attempted synthesis of XXVI from bromoketone XXXI was not successful, but led to hydrolysis to form the alcohol XXXV. A transient pink color was observed in the reaction between bromoketone XXXI and triethylamine, though the species giving rise to this color was not identified. In the near future, an X-ray study to establish the bond lengths and bond angles and perhaps even the electron density map in 2,4-difipfmethoxy- phenylcyclobutadiene meta-quinone will be undertaken. ~26- EXPERIMENTAL Preparation of 2,4—diphenyl-3—phenylacetoxycyclobutenone XXX15 In a 3-1, three necked round bottomed flask equipped with a Friederichs condenser, a mechanically driven stirrer with a teflon paddle and a 500 ml. separatory funnel (equipped with a drying tube) was added 500 ml. of anhydrous ether and 53.3 g (.345 moles) of phenylacetyl chloride. To the refluxing solution was added a 500 ml. solution Of 34.5 g (.342 moles) of triethylamine in anhydrous ether, over a period of 5 hours. The mixture was stirred at reflux for an additional 2 hours. The reaction mixture was filtered warm to remove the precipitated triethylammonium chloride. Immediately on cooling, the light greenish . filtrate began to precipitate tiny platelets of the desired product. The precipitated product was filtered and the filtrate was concentrated to about half by evaporation at room temperature with a roto-evaporator. On cooling precipitation produced more Of the desired product. The concentration precipitation process was repeated again to give a third crop of crystals. The total yield of crude product was 85%. The product was not further purified at this stage but carried on to the next stage. If desired the product can be recrystallized from boiling ether. Yields in the reaction are variable. ADUJOI 5.6 6(5), 5.72 u(s), 6.1 u(s), 6.25. max nmr (deuterochloroform) T 6.17 (singlet, area = 2), 4.70 (singlet, area=1) and 2.03 - 2.78 (complex multiplet, area = 15). The spectrOSCOpic data correspond with those reported by Farnum, et. a1.15 _27_ Conversion of XXX to 2—bromo-2 4-diphenyl-3—hydroxvcyclobutenone (XXX1).19 ’OJ’L’VL’L’D‘D’VL’VL’L’VL’VL’VD’M’VVLW’V'be’l/C’b’b’b’b ’Vb ’Vb’Vl/b WWW/UL“: WWWVVVVL'UVLW’VL A stirred solution of 13.0 g (.038 moles) (XXX) in carbon tetrachloride was treated with a solution of 6.1 g (.038 moles) of bromine in the same solvent. A white precipitate formed which was filtered with a suction funnel and washed with benzene. Yield of crude bromide was 9.7 g (83.7%). ir 1::iOI 2.9—4.5 H (br), 5.75 H (m), 6.1-6.5 H (br,m); (lit.)5 INUjOl 5.71 H. max €88H868888H8€HXXHHHHemzHézééHXHHHHXaéaézééehsexésxséeHHEHHHHHHHXXXXI-5 A sample of bromoketone XXXI was added to a 5% sodium bicarbonate solution. Upon acidification with 1N hydrochloric acid a white solid formed which was filtered and dried in a vacuum desiccator. ir 122i01 2 9-5.0 H (br, w), 5 75 H (s), 6.3 H (w), 8.70 H (m); (lit.)5 Nujol Amax 3.0 H, 5.82 H. BECCE$QCNQ£N¥§¥$%¥$£hm££$§£h¥$§$i§€m Triethylamine was distilled from and stored over potassium hydroxide pellets for use in the reaction. Diethyl ether of reagent grade quality was distilled over lithium aluminum hydride, directly into the reaction vessel. All glassware was baked dry at 150°C. for several hours before use. The reaction mixture was kept under an atmosphere of helium. -28- In one such reaction 40 m1. of ether were distilled into a 100 ml. 3-necked flask previously swept with helium gas. To this was added .669 g. (2.02 mmoles) Of recrystallized bromide XXXI. To the magnetically stirred ether solution was added drOpwise .280 ml. (2.02 mmoles) of triethylamine with a syringe. The solution was bright orange after addition of the first drop of amine, but gradually turned yellow. Workup consisted of filtration of the solid from mother liquor. Evaporation of ether produced a small amount of gummy oil, which had a broad ir band at 5.75u, but no other distinguishing features. Further workup did not achieve isolation of any pure compounds. The solid filter cake isolated had an ir spectrum which exhibited all the peaks of a composite of the ir spectra of alcohol XXXV and triethylammonium bromide and no others. Washing Of the solid with methylene chloride removed the triethylammonium hydrobromide. Evaporation of the solvent produced .225 g. The remaining filter cake (.230 g) had an ir spectrum identical to that of alcohol XXXV. £€$£€£€£$88H86H8$§H§ES%H€£R 8886888HHHX$888£$8.8&%%.£§%%¥$%%4o16 In a 500 m1- Parr pressure bottle was placed 18.1 g (.1 mole) of pfnitrophenylacetic acid. TO this was added 150 ml. of glacial acetic acid, 25 ml. Of formalin and .20 g Of 5% Pd/C. The mixture was hydrogenated in a Parr hydrogenation apparatus for about 3 hours, or until the theoretical uptake of hydrogen had occurred. The hydrogenated product was filtered and the filtrate evaporated until a white solid remained. The crude product has the odor of formaldehyde and is probably some polymeric form of the material. The product forms in almost quantitative yield, but recrystallization from 95% ethanol cuts the yield sharply to 37.8% (6.8 g). _29- Nujol max ir A 3.0—5.0H (br,w), 5.9H (s), 6.18p (m), 6.56m (m), 6.82m (s). nmr (deuterochloroform) T 7.14 (area = 5.7), 6.53 (area = 2) 3.18 (A2B2 pattern, Av = .45 ppm, area = 4) and -l.18 (area = 1). 2222222222.22.2222222. 2.22222 222222 222222222222222222222222222 hxdxgch13£1dgflXXX1Xf7 In a 500 m1. bottle was placed 7.2 g (.04 moles) of XXXVIII and 200 m1. of carbon tetrachloride. To this was added 16.7 g (.08 mm) of phosphorus pentachloride. The bottle was stoppered tightly and. placed in a shaker for 24 hours. The crude acid chloride hydrochloride separated as a tan crystalline material which floated on the top of the carbon tetrachloride and caked the walls of the bottle. The product was scraped from the walls and filtered with a suction funnel accompanied by many washings with carbon tetrachloride to insure complete removal of excess phosphorus pentachloride. The product was not further purified but carried on to the next step. ir 2E§i°l 5.62 (s), 6.26H (w), 6.62H (s) and 6.86H (s). nmr (deuterochloroform) T 6.70, 5.72 and 2.12 (A2B2 pattern, Av = .43 ppm, J = 8 cps). 222222222222222M222222222222222222222222222222- To 8 . 5 8- <-04 mo 18 s) of XXXIX suspended in 500 ml. of refluxing anhydrous ether was added 8.0 g. (0.08 moles) of triethylamine in 250 m1. anhydrous ether. The addition was carried out over a period of about 2 hours under an atmosphere of nitrogen. -30- Following the addition the reaction mixture was allowed to stir at reflux for about another hour. The ether was then evaporated to leave a brown solid. The solid was dissolved in chloroform and triethylamine hydrochloride was precipitated with hexane. The filtrate was treated with bromine solution to give a small amount of a black solid which was filtered then treated with a small amount of dimethyl sulfoxide. This caused solution of the impurities, leaving behind about 1 mg. of tiny blue crystals, mp 290—305° (dec). Hv 2CHCl3 623, 416, 392, 368, 314 mH. max 222222222222222222222222.222222222222222222222222222222-13 11-4 g (.1 moles) of squaric acid and 24.2 g (.2 moles) of N,N-dimethylaniline was added to a. solution of 150 ml of l-butanol and 60 m1. of benzene under reflux. The mixture was refluxed until a total of 3.2 ml. Of H20 was collected by azeotropic distillation - about 4 hours. The reaction mixture was deep blue and on cooling deposited a total of .33 g Of a blue powder. The filtrate which was green brown was heated on a steam bath and more benzene was added. The boiling was continued until the solution appeared green. The solution on cooling deposited more blue material. This process was continued until a total of .5 g was collected. The reported yield for the reaction is 9.2 g.13 The .5 g material was recrystallized from one gallon of boiling acetic acid to give 249 mg. of "blue product", m.p. >300° (lit. = 276°)13 Anal. Calcd. for CZOHZOOZNZ: C, 73.6; H, 6.58; N, 9.19. Found: C, 74.3; H, 6.27; N, 8.72. -31- uv 232313 623, 416, 392, 368, 314 mH; (112.13 628, 414, 389, 366, 306, 263 H). ir AEEiOl 632p (s) (shoulder at 6.2H) (no absorption in carbonyl region below 6H). nmr T 6.37 (area = 6), 1.98 (A2B2 pattern, Av = .83 ppm, J = 10 cps, area = 3.4). IRHHHHEHHéKR£¥RHHER2EEHEHCRIEQQHSSn'19 TO a mechanically stirred refluxing solution of 36.9 g (.214 moles) of pfmethoxyphenylacetyl chloride in about 300 m1 of anhydrous ether, was added 2.16 g (.214 moles) of triethylamine in about 200 m1 of the same solvent. The reaction mixture was stirred for two hours following the addition. The reaction mixture was then filtered free of precipitated triethylammonium chloride. The ether solution was evaporated to half its volume and stored in a refrigerator overnight. The precipitated material was combined with another crop Of crystals obtained on further concentration of the ether solution. Treatment of a methylene chloride solution of this product with a solution of bromine in the same solvent gave 3.5 g of a purple powder. Recrystallization from a boiling acetonitrilghenzonitrile solution produced 1 gm of beautiful and delicate purple needles mp 212-214° (dec). Anal. Calcd. for C20H14O4: C, 73.46; H, 4.83. Found: c, 73.08; H, 4.90. _32- ir 2:2: ; 6.1H (s), 6.3 (s) and no absorption less than 6H. Hv - visible 2322012 536 (a = 1.41 x 105), 500 (2 = 4.15 x 10“) and 348 HH (2 = 9.73 x 103). nmr (deuterochloroform - trifluoroacetic acid) T 5.97 (singlet, area = 3) and 2.16 (Av = 1.35 ppm, J = 9 cps, area = 4.1). It is also possible to prepare the "Purple Product" directly from the hydrolysis product Of trimer XLVIIISalg. 22222222222222222222222222222222222.22222222222- 300 mg- <1 04 mmoles) of "Purple Product" was added to a solution of 25 ml of 48% fluoroboric acid and 25 m1 of acetic anhydride in a teflon bottle. TO the mixture was added approximately 180 mg (2 mmoles) of tropilidene. This mixture was allowed to stir at room temperature for 2 hours. During this time a small amount of solid formed. Filtration of the reaction mixture produced a solid which was recrystallized twice from ethyl acetate to give 7.2 mg of material melting at 167°. The solid began to turn pink almost immediately. ir 223101 2.9-5.0H (br,w), 5.75H (s) and 6.2H (s). nmr (Deuterochloroform—trifluroacetic acid) T 6.20,6.l7 (total area = 6), and 2.99 (AZBZ pattern, Av = .63 ppm, J = 9 cps). No white solid formed in this reaction under identical conditions in the absence of tropilidene. -33- I H _ _m _ 222.2222.22.222.‘.222 2222222222.2222.2h 2.22211 “12222222222 12222222022 €19. butenone (LIIIA.15 50 mg. of "Purple Product" and 50 mg Of dione were added MWWJ’L’L’DW’VV’VW’D to a solution of 5 m1 of 48% HBF4 and 6 ml Of (CH3CO)20 in a teflon bottle. The mixture was allowed to stir for 3 hours at room temperature. A solid precipitate formed which after crystallization from ethyl acetate produced enough material for an ir spectrum in nujol. ir ANujOl 2. 9- 5. OH (br, w), 5.75H (s) and 6.2H (s). max 22222222222222.22222m0-2 dimet22222222& 222222222222n°222222222 To 2.16 moles Of propionyl chloride and 1120 ml of methylene chloride in a 3-1 three necked flask, equipped with a mechanical stirrer and condenser,, was added drOpwise, over a period Of two hours, 2.16 moles of triethylamine in 250 of methylene chloride 213.3 500 ml addition funnel. The solution was allowed to reflux for an additional two hours followed by stirring at room temperature for a period of ten to twelve hours. The solid triethylamine hydrochloride was filtered to give an orange solution. The filtrate was then concentrated to about half and again filtered free of precipitated triethylamine hydrochloride. This process was repeated until an orange Oil remained. The orange Oil, a mixture of both methylketene trimer LV and B-lactone methylketene dimer LVI was not further purified. The orange oil was diluted to about twice its volume with carbon tetrachloride and to this stirred solution was added a solution of bromine 19 in the same solvent. The addition of bromine was accompanied by the -34- evolution of a vapor with the odor of propionyl bromide and the precipitation of a white solid. The addition of bromine was continued until it appeared that no further solid formation was occurring. The reaction mixture was then filtered with a suction funnel. Washing the filter cake with benzene gave 21.5 g of a white powder which on recrystallization from an ethyl acetate-benzene mixture produced a first crop of crystals, mp 155-156° (lit.20 158-159°), weighing 11.6 g (11.92 based on propionyl chloride). Overall yields in this reaction were variable. ir 2Nuj01 2.9—4.5H (br), 5.75H (s). max nmr (deuteroacetone) (T) 8.14 (singlet, area = l) and 8.46 p.p.m. (singlet, area = 1). 222222282222222222222222222222282222- In a typical experiment .88 g of LVII was placed in a clean solvent trap equipped with a magnetic stirring bar and cooled to —78°C. In the trap was condensed about 10 m1 of trimethylamine. The Dry Ice-acetone trap was then removed and the reaction mixture was allowed to warm to room temperature with vigorous stirring. The crude reaction product was a cream colored powder from which no products other than alcohol LXI and trimethylammonium bromide could be isolated. nmr (crude product in deuterodimethyl sulfoxide) (T) 8.74, 8.56, 8.43, 8.30, 7.19 (doublet, J = 9 cps), and 6.82. Upon addition af alcohol LVI to the nmr sample of the reaction mixture the absorption at T 8.74 and 8.56 ppm increased in intensity. Addition of trimethylammonium ion (from (CH3)3NHC1) increased the intensity of doublet at T 7.19. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. -35_ Bibliography Cyclobutadiene and Related Compounds by M. P. Cava and M. J. Mitchell, Academic Press, New York (1967). L. Watts, J. D. Fitzpatrick and R. Pettit, J. Am. Chem. Soc., 81, 3253 (1965). V. H. . Sprenger and W. Ziegenbein, ibid., R. Sandel, H. H. Freedman, ibid., 2059 (1968). H. Freedman, A. M. Frantz, Jr., 121$}, 84, 4165 (1962). G. Farnum, B. Webster, 123$}, 85, 3502 (1963). F. Bryan, ibid,, 86, 733 (1964). Breslow, Chemical and Engineering News, p. 90, June 28, 1965. J. Smutny and J. D. Roberts, J. Am. Chem. Soc., 11, 3420 (1955). Cohen, J. Lacher and J. Park, ibid., 81, 3480 (1959). Treibs and K. Jacob, Angew. Chemie I. E., 4, 694 (1965). 553 (1967). Q, Ziegenbein and H. Sprenger, ibid., 5, 893 (1966). 2, 894 (1966). Sprenger and W. Ziegenbein, ibid., . Treibs and K. Jacob, Liebigs Ann. Chem., 222, 153 (1966). G. Farnum, J. R. Johnson, R. E. Hess, T. B. Marshall, and B. Webster, . Am. Chem. Soc., 87, 5191 (1965). E. Bowman and H. H. Stroud, J. Chem. Soc., 1342 (1950). . Levine, ibid., 19’ 1382 (1954). . G. Farnum and R. Hess, unpublished work. G. Farnum and B. Webster, unpublished work. G. Farnum, M. A. Tyrell Heybey, and B. Webster, 86, 673 (1964). Lowenstein and S. Meiboom, J. Chem. Phys. 27, 1067 (1967). ”722172111211111211112111:21'“