OVERDUE FINES: 25¢ per cu per item RETUMUKS LIBRARY MATERIALS: Piece in book netum to nemve chime from circuhtion records SYNTHESIS OF POTENTIAL ORGANIC CONDUCTOHS By Shuh-Chung Chen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1981 ABSTRACT S fmT SIS OF POTELTIAL ORGA IC COZ ”DU TORS m c4 C ) :54 D Shuh-Chun: Recent interest in highly conducting charge-transfer ‘Lfi-‘H U...CL" 0): salts derived from the electron acceptor 7, 7, 8,8, cyanoquinodimethane (TCI‘) and the electron donor tetra- thiafulvalene TTF) has prompted the design of new organic heir electrical ct compounds wr ose structures would enhance properties, and consequently advance our understanding about the mechanism of con‘uctivity. He describe here our attempts to prepare the extended conjugated and heteroaton substituted analogs of these systems. The first area studied was that of the thiazoloC5,#-d]- thiazole ring system, a heteroatom substituted analog of ~T”AP, which contains two nitrogens and two sulfurs but possess lO-w electrons. The thiazolothiazole ring was constructed by condensation of dithio Kamide with aromatic aldehydes. The synthesis of 2, 5-dihydro meetm l and 2,5- chloromethyl thiazoloC5,4-d]thiazole was descrioed. Shun-Chung Chen The oxidation of substituted 4-hydroxyphenyl derivatives to the corresponding quinoids was also successful. Methylation of nitrogens on the thiazolo[5,4-d]thiazole heterocycles to give dicationic salts as radical cation precursor was highly successful. Synthesis of benzoEl,2-d:4,5-d"]diimidazole systems by making use of 1,2,4,S-tetraaminobenzene was also attempted. Two extended conjugated TTF analogs, the diphenyl and the tetramethyl tetrathionaphtho-Z,6-quinodimethane (TTKQ), were prepared. The diphenyl-TTNQ formed a non- stoichiometric charge—transfer complex with iodine. Synthetic efforts toward 7,8-dicyano-7,8-dinitro- quinodimethane and ll,lZ-dicyano-ll,lZ-dinitronaphtho- 2,6-quinodimethane were described including the prepara- tion of p-bis(cyanonitromethyl)benzene and 2,6—bis(cyano- nitromethyl)naphthalene. Finally, the attempt to introduce a second cyano group into a carbon center through enamines of active methylene compounds resulted in an unexpected ring closure to afford the S-aminoisoxazole derivatives. To my wife Jifang and my parents, without whose support this would not have been possible. ii AC}'£1;O)JLEDGLE1ITS The author gratefully expresses his sincere appreciation to Professor Eugene LeGoff for his guidance, assistance and patience throughout the course of this investigation. The author acknowledges the financial support from the Department of Chemistry at Michigan State University in the form of teaching assistantships. Ho Ho Ho \. TABLE OF CONTENTS Chapter Page LIST OF TABLES ............................... ..... ix LIST OF FIGURES .............................. ..... X INTRODUCTION ................................. ..... 1 RESULTS AND DISCUSSION I. Chemistry of thiazoloC5,h-d]thiazole A. 2,5-Disubstituted thiazolo[5,u-d]- thiaZOle nUCleuSC.......CCCOOCCCOOOO..... 22 B. Synthesis of thiazoloC5,#-d]thiazole dicationic salts l. Methylation of nitrogens .............. 34 2. Reaction of dithiooxamide with phosgene immonium chloride ............ 36 II. BenzoE1,2-d:u,5-d'Jdiimidazole ............. 38 III. Synthesis of 2,2'-(2.6-naphthalene- diylidene)biS(lg3-dithiOle) 00000000000000. 42 IV. Preparation of 7,8-dicyano-7,8-dinitro- quinodimethane and ll,lZ-dicyano-ll,12- dinitronaphtho-Z,6-quinodimethane .......... 52 V. Preparation of 4-phenyl-5-amino isoxazole ... 62 VI. Reduction of TCNQ with sodium azide ........ 67 .EXPERIMENTAL General procedures ........ ..... ................ 68 [Bis(methylthio)methylenejmalononitrile, fl. R=SCH3) .................O............... 68 iv Chapter Page Condensation of dithiooxamide with aromatic aldehYdeS O.......OOOOOOOOOOOOO0.0.0.000... 69 2,5-Bis(2-furyl)thiazolo[5,4-d]thiazole 1; ... 7o DiphenylthiazoloC5,u-d]thiazole .............. 7O 2,5-Bis(2-hydroxy-l-naphthyl)thiazoloC5,4-d]- thiaZOleg-Q..................OOCICOOOOOCOO 71 2,5-Bis(4-hydroxyphenyl)thiazoloC5,4-d]- thiaZOIel—é......C......0................. 71 2,5-Bis(3,5-di-tert-butyl-4-hydroxyphenyl)- thiazoloC5,4-d]thiazole I2 ................ 7l 2,5-ThiazoloE5,4-djthiazoledicarboxylic aCidg.........CCCCOCC......O....0....... 72 ThiazoloC5,4-d]thiazole ...................... 73 Dimethyl 2,S-thiazolot5,4—djthiazole- dicarboxylate 0000......0.0000000000000000. 71+ 2,5-Bis(hydroxymethyl)thiazoloE5,4-d]- thiaZOleJ-l0.00.000.0.0000000000000000000. 74 2;5-Bis(chloromethyl)thiazoloC5,4-d]- thiaZOIel-fl......0.00IOOOOOOOOOOIOOOOOOOOO 75 2,5-Bis(2,5-cyclohexadiene-3,E-di-t-butyl- l-diylidene-u-oxo)thiazolo 5,4-d3- thiazole gp, (R= t-butyl) ................. 76 Oxidation of 2,5-bis(2-hydroxy-l-naphthyl)- thiaZOlOE5,l+-d]thia201e £2 on... 00000000000 76 Oxidation of 2,5-bis(p-hydroxyphenyl)- thiazoloE5,4-d]thiazole ié ................ 77 N,N'-Dimethyl-2,5-diphenylthiazoloC5,4-dj- thiazole trifluoromethanesulfonate salt 2’48. on...coo-00000000000000.-uoooooooo 77 N,N'-Dimethylthiazolo£5,4-djthiazole trifluoro- methanesulfonate salt 25 ...... ..... ...... 78 Chapter Page N,N-Dimethylamino-triphenoxymethane.32 ...... 79 2, 6- Diphenyl-Tll(3)H. 5(7)H- benzoCl, 2- d: 4, 5- -d' J- diimidazole 15 ........................... 79 2,6-Bis(€, ,5- di- .tert- -butyl- -4-hydroxyphenyl)- benzo 1,2 4,5— -d']diimidazole.2_ ........ 8O Oxidation of 2, 6- bis(3, 78 -di- t- .buty l- 4- l, hydroxyphenyl)- -benzo W5 -d']- diimidazole 3_. ...... .................. 80 Reaction of 1,2,4,5-tetraaminobenzene with bisdithiomethyl dicyanoethylene .......... 81 2,6-Bis(bromomethyl)naphthalene ............. 81 2,6-Naphthalenedicarbodithioic acid, dipiperidinium salt 44b .................. 82 Dipfienacyl 2,6-naphthalenedicarbodithioate, ..................... ......... ......... 83 Bis(l-methyl-Z-oxopropyl) 2,6-naphthalenedi- carbodithioate, flé ........ ..... .......... 83 Trifluoromethanesulfonate salt of diphenyl-TTNQ 48b ........................ 84 Fluoroborate salt of diphenyl-TTRQ Egg ...... 84 Bisulfate salt of tetramethyl-TTUQ'EZQ ...... 85 Reaction of 48b with sodium methoxide inmethanOI ................ ....... ....... 85 Reaction of 42a with sodium methoxide in methane]. .........0....................... 86 Reaction of the dimethoxy derivative 59 WithHI ......I......................OO... 86 Reduction of 48b with tetrabutylammonium iOdide .................. ..... ............ 87 Complex of 48b with Li+Tcuo‘ ... ...... . ...... 88 vi Chapter Page APP Bis-1,2-dithioethane acetal of te re- phthaldehyde 52 ........ ................. .. 88 Dipotassium p-bis(cyanonitromethyl)benzene, ......... ..... . ..... . ...... .... ........ . 89 Disilver salt of p-bis(cyanonitromethyl)- benzeneéjooooo ..... ......O............... 90 p-Bis(chlorocyanonitromethyl)benzene 66 ...... 90 Dipotassium salt of 2, 6- -bis(cyanonitro- methyl)naphthalene:6J .....0.............. 9]- p-BiS(nitr0methy1)benzene _6_Z O o o o o o a o o o o o o o o o o 92 Reaction of p-nitrOphenylacetonitrile with bis-dimethylamino-t-butylmethane (Brederek's reagent) ........ ..... ......... 92 Reaction of phenylacetonitrile with Brederek'sreagentoooooooooooooooooooooooo 93 Preparation of 5- amino- -4- (p- nitrophenyl)- isoxazole 62a ..... .. ..... .............. 94 Preparation of 5-amino-4-phenyl isoxazole, 62 ...-......000 ..... ... ......... 00.00.... 94 5-Acetamido-4-(p-nitrOphenyl) isoxazole 20a .. 95 Preparation of TCNQ Na+ by TCNQ and sodium azide ........... ....... . . .. ......... .... 95 ENDIX PMR spectrum (top) and Ultraviolet spectrum (bottom) of gp .......... ..... ............. 97 Infrared specturm of gp ...................... 98 Mass spectrum of gp .......................... 99 PMR spectrum of $2 (top) and 24 (bottom) ..... lOO Infrared spectrum of 24 ...................... 101 130 mm (top) and P1213 (bottom) of 13 102 vii Chapter Page Mass spectra of 1} (tOp) and 14 (bottom) ...... 103 PMR spectra of‘flflp (tOp) and 41a (bottom).....- 104 PMR spectrum of 32 ...........................- 105 PMR spectra of 69 (tOp) and 63 (bottom) ....... 106 130 NMR spectra of 69 (top) and 6} (bottom).... 107 Infrared spectrum of 69 . ........ .............. 108 Infrared spectrum of 6} ....................... 109 PMR spectra of 68p (t0p) and 62; (bottom) ..... 110 PMR spectra of‘éga (top) and 62a (bottom) ..... 111 13C NMR spectrum (top) and Mass spectrum (bottom) 01.62 ........0......... ........ ......I.... 112 Mass spectrum (tOp) and PMR spectrum (bottom) szg......O...........O.‘......O........I.. 113 Infrared spectrum of 62b ....... .............. . 114 Infrared spectrum of 62 ...................... 115 Infrared spectrum of 29 ...... ............... .. 116 REFERENCES ....................................... 117 viii LIST OF TABLES Table 1. TTF analogs and the conductivity of their TCT‘IQ salts ..... ..... ............OO...... 2. TCNQ analogs and the conductivity of their TTF salts ............ ..... .............. 3. Melting points and yields of TTNQ derivatives ....0..........OOOOOOOOOI.... ix Page 13 51 LIST OF FIGURES Figure Page 1. Structure of TCHQ and TTF .................... 4 2. a) Scale of approximate conductivity b) Comparison of conductivity of TTF-TCNQ WithIVT‘P-TCI;Q coco-coococo-0.00.00.00.00... L" a) Schematic representation of TTF-TCRQ b) Crystal structure of TTF-TCNQ projected along the a-aXiS 00000000000000000000000000 5 4. a) TTF-TCNQ projected along conducting axis b) Molecular overlap in the columnar stacks inTTF-TCPJQ ...0..........O................ 6 5. Migration of aromaticity along TCNQ stack .... 10 . Condensation of dithiooxamide with aldehyde .. 23 7. Reaction of dithiooxamide with active dicyano methylene compounds .................. 24 8. syntheSis OfTCQQ ..........O..............O.. 25 9. Classical route to tetracyanoquinones ........ 26 10. Known stable oxygen radical anion ........... 29 11. Synthesis of 2,5-Bis(2,5-c clohexadiene- l-diylidene-4-oxo)thiazolo 5,4-d] thiaZOle-g-a;00.00.00....0000........00......0 30 12. Preparation of compound pp and 3 ............ 31 13. Radical anion 0f22....................OI... 32 14. Thiazolothiazole-TTF .. ..... . ...... .......... 33 '5 (\3 l\) Kl, “I v“. 1': JI . HA 4%. Chapter 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33- 34. 35. 36. 37. 38. 2,5—Bis(N,N'-dimethyl-4-pyridyl)thiazolo- £5,4'djth132018 o. ..... coco-0000000000. ..... Methylation of thiazolothiazole hetero- eyeles.......................O..........O Projected synthesis of compound 34 ......... Synthesis of TTNQ derivatives .............. Reaction of 48 with sodium methoxide ..... Reaction of 42 with sodium methoxide ..... Chemical shift change from dicationic salt to dimethoxy derivative ............ Iodine complex of diphenyl-TTNQ .......... Complex of 48b with LiTTCNQ7 ............... Iodine complex of tetramethyl-TTNQ ....... Syntheses of TTF analogs 55 and 56 ......... Proposed synthesis of TTQ .................. Classical route to TCNQ .................... Projected synthesis of TCNQ analog 2 ....... Nitration of active methylene compound ... Acidification of dipotassium salt 63 ....... ReactionOfsaJ-tfl00000000000000.0000... Hydrolysis of dipotassium salt 69 ....... Proposed syntheses of a,a-dinitrile ...... Synthesis of nitrile by using HOSA ......... Synthesis of a,a-dinitrile with HOSA ....... Cyclization reaction of 68a with HOSA ...... Cyclization of enamine 68b to isoxazole 62b. Mechanism of cyclization of 68 ........... xi 34 35 38 44 46 46 47 47 48 49 5o 50 53 54 55 57 58 6o 62 63 63 64 65 ' 66 SYNTHESIS OF POTENTIAL ORGANIC CONDUCTORS ulld uh s.~ INTRODUCTION The capacity for metallic behavior is most common among inorganic compounds: three-quarters of the total number of simple substance formed by elements, some of their compounds, and a large number of alloys are so constructed that the metallic state is the main stable state.1 By contrast, the majority of organic compounds are poor electrical conductors with room temperature 14 -l conductivities in the range of 10.9 to 10' ohm-1cm . Graphite is perhaps the only instance of metallic or semimetallic conductivity. In the 19th century, during the rapid development of organic synthesis, the idea of possible similarities between alkali metals and organic radicals was quite popular. Research in this direction revealed one of the main difficulties in the preparation of organic metals, namely, the pairing tendency of valence electrons. The synthesis of stable free radicals produced only para- magnetic insulators rather than metals. This is due to the localization of unpaired electrons on separate molecules. However, strong w-molecular donor (D) and acceptor (A) molecules often react to form organic charge-transfer salts.2 In contrast to the conventional molecular crystals, charge-transfer salts have unpaired electrons on the acceptor or both as a result of the simple electron transfer, D + A -—)l D? A7 .3 This strikingly simple result opens up a new area of electronic phenomena, for if the unpaired electrons delocalize over all molecular sites, the metallic state results. These materials form the basis of the field of "organic semiconductors", which has been actively studied since the mid-1950's.“’5 In the early 1960's, Melby and a duPont group discovered a new powerful w-molecular acceptor, tetracyano-p-quino- dimethane (TCNQ). As with other acceptors, the TCNQ radical anion forms organic semiconductors with a large number of cations.6"9 Only a few cases have been reported where the electrical conductivity is moderately large, with a negative temperature coefficient characteristic of metallic behavior. The primary example is the charge-transfer salt of TCNQ with the cation N-methylphenazinium (NMP)3, which was among the most electrically conducting organic solids known at that time. 10 In 1972, it was discovered that the chloride salt of a new organic donor, TTF (tetrathiafulvalene), has a high pellet conductivity (G'(300°K,r~0.2 ohm-lcm-l). The following year TCNQ and TTF were brought together to produce a new compound (TTF-TCNQ) and found to have .. .. v- N; conductivity which increases dramatically below room temperature, rising as high as 10L‘L ohm-lcm-l near 60°K (Figure 2b), high enough to be considered an "organic metal". A flurry of current research has been directed toward gaining an understanding of the electronic properties of this complex. Solid-state physicists were attracted by the high metal-like conductivity in such an exotic and unusual material and by speculations of possible high- temperature superconductivity.ll Substantial chemical studies have focused on the synthesis of a remarkable 12’13 Since the number of new derivatives of TTF and TCNQ. emergence of TTF-TCNQ several comprehensive reviews of this field from the perspective of solid state physics, physical chemistry, and synthetic chemistry have been published.2’12’M-18 Part of the driving force behind this intense interest is the hope that organic chemistry can be utilized to produce new organic solid for electronic applications. Spectrosc0py studies revealed that the TCEQ molecule serves as the electron acceptor while the TTF molecule is the donor. The crystal structure indicates (Figure 3,4) a rather closely spaced plane parallel arrangements of the TCNQ and the TTF molecules as segregated stacks. The interplanar stacking distance (3.17-3.30 R) and the NC __ CN [: >=<§ I} NC - CN TCNQ 77F Fig. 1: Structure of TCNQ and TTF. log a +6 9000 X z I r T 0'8 . Momls 1 311.3 + #33231: 4 +2 u _ 0 Molten salts 700° Nichromo -2 7.. E Silicon -4 AlkaHTCNQ salts 5. Saab-1 Cyancmhrono '5 3: E -8 g 2 -10 Ccrbazolo-Iodlno 3 3000-- Moat mozcculor comp“)! crystals ° '12 s , Fhflnaxm Suflur '14 . .45 Ian- Qucrtz Anthrcccno -18 l I l ' 0 so I00 :50 zoo 250300 °k Fig. 2: a) Scale of Approximate Conductivity b) Comparison of Conductivity of TTF-TCNQ with a€00] [001] TCNQ S TTF W lo‘ Fig. 3: a) Schematic Representation of TTF-TCNQ. b) Crystal Structure of TTF—TCNQ Projected along the a-axis. (taken from Ref. 14). P. Fig. 4: a) TTF-TCNQ Projected Along Conducting Axis. b) Molecular Overlap in the Columnar Stacks in TTF-TCNQ. (taken from Ref. 14). 4.- ...-< fiv- H‘ 'I 5n '1 resulting s-orbital overlap integral between TCNQ ions indicates that the band widths in these solids are rather narrow, probably in the range of 0.1-0.6 of an electron volt.lle The overlapping electron cloud can be considered as the conducting media. Because of the geometry of the ~- orbitals, the band-widths, and hence the electronic properties are inherently anisotropic. The direction of highest conductivity is in the stacking direction. This is similar to what is know also from graphite and other polymeric compounds.4 The reaction of TTP and TCRQ formally involves an electron transfer to give TTF radical cations and 1CRQ radical anions. For full electron transfer in the solid, the electrostatic repulsion, due to having like charges so close together in the molecular stacks, is expected to be considerable. However, the solid can gain significant stabilization by incomplete or partial charge transfer.20 In fact, several recent investigationszo-23 have concluded that the extent of charge transfer in TTF-TCRQ is only 0.5-0.6 of one electron. That is to say, that there is a complete and spontaneous electron transfer from a certain fraction of the donors to the acceptors, so that the solid consists of an assembly of now positively charged donor molecules, and negatively charged acceptors, plus a certain fraction of neutral donors and acceptors. Negative charges are mobile because electrons transfer from a negative acceptor to a neutral acceptor , much the same as in n or p type semiconductors. This situation can be promoted by regulation of the oxidation and reduction potentials of the interacting species. Hhelanqu prOposed that the difference between the ionic potential of the donor and the electron affinity of the acceptor would determine the relative conductivity of that complex, thereby suggesting why some donors only form conducting salts with particular acceptors. Most of the duPont TCNQ salts which have segregated stack structures also have periodically distorted linear chain structure and are insulator.5 Only the highest conductivity TCNQ salts have both a segregated stack and a non-distorted linear chain structure. The early notable member of this exclusive class were JMP-TCEQ and quino- 1inium (Q) TCNQ)21u. Another structural feature that makes TTF-TCUQ an attractive candidate for study is the polarizability of the cation portion of the complex. The higher the lattice polarizability, the higher is the local dielectric constant: so that the conduction electron-cation chain exciton coupling can facilitate conduction by reducing the effective Coulomb 25 96 J’~ repulsion (Ueff) between two electrons on the same site. '1' 1 0 18 e w n 0 Moreover, Ierlstein has deveIOped a Iormallsm for the synthesis of new highly conducting organic solids based ) on the "local aromaticity" concept. He contended that the most efficient conductors contain molecules a) whose radical ions form a new aromatic sextet upon one-electron oxidation or reduction and b) whose aromaticity can migrate ‘ oy mix-valence interaction (Figure 5). More information about t e prOperties of organic metals has been obtained by modifying either the acceptor or the donor in a systematic way. However, compilation of the conductivity data have shown that these factors do not 1 completely explain the phenomena. The problem is presumably due to the fact that more than one structural feature has always been altered. 1‘" o 1 ‘ ' Cowanl summar1zed some of tne or an1c metal des1gn c3 constraints as following: Stable Open-shell (free-radical) salts. Planar molecules with delocalized 7-molecu1ar orbitals. Inhomogeneous charge and spin distribution. Segregated stacks of radical ions. Uniform stacks — no periodic distortions. Both cation and anion have Open shell. Both cation and anion are nominally divalent. Fractional charge transfer. Polarizable cation. \0 CDV O\U\ 43kt) N H [.4 O ‘T 0 H "‘ 4...-.. o (.... o . . q (“J‘- c A \ no disorder (s3 metric anions and cationS). H H Cation and nion of similar size. 9) Fig. 5: Migration of Aromaticity along TCNQ stack. (taken from Ref. 18). u ... U-b - ul' 0 To a de: ”re unmatched among 0 the intrinsic onductors, the electronic properties of organic charge-t Hansf r salts are subject to chemical control. Rowever the design of new donor and acceptor molecules based on these constraints does not guarantee one a good new rganic metal. Among . . H l}: the derivatlves and analogs of TTF‘TCUQ I, for example, even minor differe ence s in molecular structure often manifest themselves as sharp contrasts in electrical behavior. Such compounis span the full range from insulators to the best organic conductor known. “rel a synthetic che.ist's point of View, this sensitivity to molecular detail presents an Lvrms1al challenge. Certainly it complicates the interpLetat1on of materials in tifi a class, out it also "ult1p11es the amount of information available and the possibilities for design. Since the discovery of *3 TF-TCNQ complex in 1972, the research in this field has been gu Md dby three major approaches, namely, side-chain modification, heavy atom substitution an d conjugated extension. These approaches as well as the synthesis of ‘11 analogous compounds of the well-known classes TC.1 and TT 3-35 are documented in the litera ture2 . Together with this, ther have also appeared communication 3 on highly conductive / o . compounds of new types. In 1977. Sandm nan3 synthes1zed a number of charge transfer complexes of TCRQ with L.’l4,' I A -bit‘1iopyran (iso-s-electronic to TTE“ with same 1f” conductivity as poly-crystalline anples of TTF-anj. Perlsteinlo re ported the hi h conductivity for the char e- transfer complex of the phenyl derivatives of A;’4 -bi- U} thiOpyran with T334. Electrochemical oxidation of tetramethyl tetraselenofulvalene has also been utilized to produce (TIT F)2X, where K is a symmetrical octahedra ' ~ - . - ». —1- ' 9 an10n37’383 PF/ , {s34 , sbr» or a tet ahedra anion3 , . \ is. \l "5 ClCu . These compounds exhibited superconductivity at moderate hydrostatic pressure Listed below in Table l are severalm odi ied donors 4- 4—11’7'1f'“-."f\ 18 and the room temperature conductivities of heir loud salts. In l-2, substitution of the TTF molecule with four methyl side chains causes a two-fold increase in the conductivity whereas the octanethyl derivative 1-4 has a much lone value. Besides sidechain modification, extension of the unsaturation was also carried out. The benzo analogs of ‘1 ET? and TTR (l—lj), l-5 and 1-14 respectively, have recently been prepared as has the phenylene analog of TTF. 1-12. These variations, in general, have not increased the conductivity to any great extent. The third major type of modification, heteroatomic exchange (heavy atom substi- tution), has resulted in significant improvement in conductivity. At room temperature, for example, the conduc- tivity of 1-10 is about two times higher than that of TTF. 13 Table 10 TTF analogs and the conductivity of their TCNQ salts° H [\J Kl) @- U1 0\ fl CHE 2::I: >==< :Ifi: CI:>=<:I> OE: Hi]. my: my - Ratio: TCNQ 1:2 1:1 1:1 1:1 1:1 1:1 Room Temperature Conductivity (Pellet) ohm"l cm- 500 103 500 5x10-5 100 2x103 ...— 14 Table 1. (cont'd.). Room Temperature Entry Ratio: TCNQ Conductivity (Pellet) ohm"1 cm-1 3 S 9 [w 1:1 550 10 W 1:1 800 11 S |’-fl 5 1:1 600 s (3*: _ 2H3CI ”451 1:2 10 3 H3c S 5 CH3 '8 13 (::)(::’ 1:1 40 3‘5 / \ 1:1 1 5—3 15 S- -S 1:1 30 , ‘Vl l\) Y- V. n.‘ "5 15 And the most striking example is l-8, whose TCHQ salt has the highest room temperature conductivity measured thus far. Regarding modification of the acceptor TCJQ, much less work has been done. The majority of the changes have involved sidechain variations in TCKQ itself. Many deriva- tives of TCKQ have been synthesized, incorporating F Cl, N CD" Br, I, SR, OR and alkyl substituents onto the ring. Some of these examples with higher conductivities are shown in Table 2, (2-l - 2-4). The extended conjugated systems 2-5 and 2-6 have been known for some time, with only 2-5 leading to a stable charge transfer complex. Heavy atom substitution in these acceptor molecules is less common, with only mono-sulfur and di-sulfur analogs of TCIQ, 2-10 and 2-ll, have been prepared. However, 2-lO has failed to undergo any charge transfer with its donor, whereas 2-ll forms only non-conducting comple: with TTF. In the case of 2-8, the dianion could not be oxidized to a stable radical. Instead, the corresponding trianion radical was isolated. Compound 2-9 is rather interesti-g. Since nitrogen and carbon have very similar covalent radii, it was expected to produce only a minor perturbation on the size . and shape of TCIQ. However, the radical 'anion" of 2-9 would be a formally neutral molecule and therefore reduce a) Kn (T\ '\1 16 Table 2. TCNQ analogs and the conductivity of their TTF salts. Ux Room Temperature Pellet Conductivity ohm.1 cm-1 500 40 21:10”1 (with TMTTF) Reference 28 28 29 30 31 32 33 17 Table 2. (cont'd.) Room Temperature Entry Pellet Conductivity Reference ohm'1 cm"1 N 11 Nc>==<2 10"6 34 NC 5 N 18 the intermolecular repulsion in stacking. It is generally believed that no one single factor could solely control the conductivity of charge-transfer complex, and that one of the important long-range goals is the development of predictive criteria fer the stability of organic conductors. In order to achieve this goal and to better understand the conduction mechanism, more deriva- tives need to be synthesized and studied. In light of the work done thus far, we decided that the most promising direction to follow was the modification of the most conductive charge transfer complex TCNQ and TTF molecules. Our first target molecule l demonstrated perfectly our idea of heavy atom substitution. By substituting nitrogen and sulfur for carbons, one increases the electron affinity of the molecule.35 Upon one-electron reduction, ; generates a 10 w-electron heterocycles thiazolo[5,4-d]thiazole, and therefore is the heterocyclic analog of TNAP (Table 2, Entry 5). We also would attempt to combine this 10 w-electron heterocycles with the well-known chloranil type electron acceptor. Molecules g and 3 are the result of this combi- nation. The synthesis of.& was also attempted, bearing in.mind that it would provide three new aromatic sextets upon receiving one electron. 19 N \ c 5 EN. :CN NC: ‘N’ 5 CN 20 The methylation of the nitrogens on thiazoloC5,4-d]- thiazole derivatives was successful to produce the di- cationic salts. The third part of this work was devoted to the synthesis of extended TTF systems, 5 and §. The relation between molecule 5, § an TTP should be similar to that between TRAP and TONE. The fourth part will describe the synthetic work of dicyanodinitroquinodimethane Z and dicyanodinitronaphtho- 2,6-quinodimethane §. we expected that the nitro groups would stabilize the radical anion as well as cyano groups and not to disturb significantly the molecular packing. In the last part of this work we described an £1 :3 (I) X '"d (D 0 Cf (D {L H Ho :3 o: O H O Q (I) O "b d" :34 (D (D :5 D Z? P. :5 (D O H) d D; (I) :3 << }—J m 0 (1) cf 0 * > ’40 d‘ H P 5...: (D agent. OzN No, NC CN / no, o,N / NC ...“— fl: RESULTS AND DISCUSSION I. CHEMISTRY OF THIAZOLOE5,4-dJTHIAZOLE A. 2,5-Disubstitutedgthiazolo£5,4-d]thiazole nucleus The first modification we made on the TTF-TCNQ system was the heteroatomic exchange. We tried to introduce nitrogen and sulfur into the system so that the molecules would have larger polarizability and better electron affinity as compared with simple carbon skeleton. In theory, the higher the polarizability, the smaller the coulomb re- pulsion energies of the odd electrons on the neighboring stacks. Simply speaking, molecules with heteroatoms would tolerate more drastic electronic change in the environment without disturbing their packing. The key molecule in this project 2,5-bis(propanedinitrile)thiazolo[5,4-d]thiazole 2 is of special interest to us. "° 51" °" —> "CHER“ \ Nc \Nl $>-LR —9 R=< _" H I >=< /0 , R1R2 =—OCH3 “oczfls- -SCH3 —OCHzCHzo —' Figure 7. Reaction of dithiooxamide with active dicyano methylene compounds. We proposed this synthetic possibility because dis- placement of leaving groups at spz-carbon of active methylene compounds by nitrogen, oxygen and sulfur nucleophiles have been well-documented. For example, reaction of [Bis(methyl- thio)methylenejmalononitrile (l9, Rf=R§= -SCH3) with 42 LP?- ethylenediamine , ethanolamineu3, O-phenylenediamine , O-aminophenoluu, and 2-aminothiophenoluzproduced the desired products in fairly good yields. Wudlushas utilized this approach in his synthesis of tetracyanoquinoquinazolino- quinazoline (TCQQ) (Figure 8). 25 Figure 8. Synthesis of TCQQ. However, when dithiooxamide was refluxed with lg (R1R2= -OCH3, -OCZH5, -SCH3 and -OCHZCH20-) in a series of different solvents such as ether, tetrahydrofuran, ethanol, butanol and pyridine, 60%-80% of the starting materials were recovered in most cases accompanied by 20%-#0% of decomposed and un- identified residues. More drastic reaction condition and longer reaction time only increased the amount of decompo- sition. Under any circumstances, no trace of desired product was detected. Failure to condense dithiooxamide with active dicyano methylene compounds prompted us to reconsider a classical but indirect route that had been utilized in the synthesis of TNAP (Figure 9). 26 HzN Is 0 \ ——> mIm 5/ NH: 0 Nis -/_/_ Hon,c—<\:I:\>—CH,OH é+— HOOc-<\:I:\>—COOH /3 l2 _ It. moo—(CI:\>—cooa é—J S ”'i NCl-l .c-(iEEHH zCN ——> (22:2<:IC>‘EHC3N) 00R 2%;E%: 4‘— (NCLMSIEHH (cm)2 Figure 9. Classical route to tetracyanoguinones. 27 2,5-Bis(2-furyl)thiazolothiazole ll was prepared by the condensation of dithiooxamide with furfuralul. Oxidation with potassium permanganate yield 2,5-thiazolof5,h-d]- thiazoledicarboxylic acid lg as white crystals. Reduction of the diacid lg should yield the diol.l3, which through halogenation and substitution by cyanide ion would yield the bis-acetonitrile‘lj. Carboalkoxylation of l5 followed by hydrolysis, decarboxylation and oxidation would possibly provide the tetracyano compound I. Some problem arose when lg was subjected to reduction with lithium aluminum hydride (LAH). When treated with LAH in THF, the solution of lg changed color rapidly, indicating a reaction had occured. However, the isolation of the product was unsuccessful. The rationale for this was that nitrogen and sulfur of thiazolothiazole have coordinated with aluminum so strongly that separation of the two was impossible. Lithium triethylborohydride (Super-Hydride) was introduced as an ideal reducing agent in this case since the boron atom was protected by ethyl groups and the possibility of coordination would be greatly reduced. Indeed, the dialkyl ester of lg, when treated with Super- Hydride in THF46 , was reduced cleanly to the diol I}. This diol was converted to the corresponding bis-chloro- methyl thiazolothiazole l& by simply warming with thionyl chloride. The substitution of chloride by cyanide ion was 28 carried out by the procedures of Stormu7(KCN, CHBOH) and Sandman.“8 (NaCN, alc. dioxane). Neither of them afforded the desired bis-acetonitrile l5. Only starting materials were recovered. Upon surveying the literature, this bis- benzylic type displacement was found to be of only poor yield. In the case of TNAP, the published result in di- methyl sulfoxide is 8%”9. Changing the solvent to methanol in that case gave 25% of the product. It is possible that the cyanide ion may be acting as a base to deprotonate the product or may just be less reactive under such polar condition. The use of a more soluble reagent tetraethyl- ammonium cyanide as previously reportedso. a rapid darking of the reaction mixture was observed. After usual work-up, a brown, insoluble and high melting solid was obtained, Which showed no signal corresponded to the desired bis- acetonitrile l5 in.mass spectra. This was presumed to be the result of polymerization. Similar difficulties were also encountered in the attempted introduction of cyano group into thienoE3,2-d]thiophene nucleusso. Therefore, this indirect synthesis of,l was also abandoned. At this point, we were led to believe that dithio- oxamide would react only with aromatic aldehydes.51 Nevertheless, this easily accessible route toward hetero- cycles as complex as thiazolothiazole encouraged our 29 continuing efforts toward the synthesis of molecules possess thiazolothiazole center , with benzoquino subs- tituents at 2,5 positions. For benzoquino type compounds such as chloranil are well-known electron acceptors. Furthermore, a recent article52dealing with the synthesis and properties of 2,5-dihydroxyl thienoE3,2-d]thiophenes showed that oxidation of the corresponding dianion led to solutions of stable oxygen radical anions (Figure 10). new“ :"2‘33' > «I» Figure 10. Known stable oxygen radical anion. With these in mind, we attempted the synthesis of quino 2a (Figure 11). Upon one electron reduction, 2; would convert itself to an oxygen radical l§ with concurrent generation of three aromatic systems, a necessary condition for a good electron acceptor according to Perlstein.18 30 WI ——> =C><§IZ>©=° _/§. __ i. was». a 1g Figure ll. Synthegis of 2,5:§is(2,5-cyclghexadiene-l- diylidene-h-oxo)thiazolo£5,4-d1thiazole fig _§ was easily prepared by the method of Johnsonb’O as a highly insoluble yellow solid. However, its solubility in basic solvent allowed us to perfbrm the oxidation with K3Fe(CN)6 in alkaline aqueous solution to afford a deep green micro- crystalline solid. Mass spectrum showed a parent peak at 324, an indication of the quinoid ga. Compound 2a was fOund to be in equilibrium with I6. In the absence of excess oxidizing agent, the peak at m/e 326 (I6) in mass spectrum increased its intensity with time until the equilibrium was reached. Heat favors the phenol form I6. The reaction mixture discolored rapidly when external heat was applied. We explained this discoloration as the result of re-aromatization and the consequent rotation 31 of the benzene rings with respect to the plane of thiazolo- thiazole. If I6 could be oxidized in the presence of a suitable electron donor, the direct complexation of 2a with the donor might be realized. If the rotation of the benzene ring was somewhat hindered, the quinoid form should have higher stability. To justify this hypothesis, 2,5- Bis(3,5-di-tert-butyl-4-hydroxylphenyl)thiazoloE5,h-d]- thiazole53,;2, and 2,5-Bis(2-hydorxynaphthyl)thiazolo- E5,4-d]thiazole, 29, were prepared?“ 5 , *NMW —_9 OER/13:1: s ' Q .212 » ‘51.: 20 3 ‘Figpre 12. Preparation of compound gp and 3. .32 Oxidation of IQ was carried out with lead peroxide in methylene chloride to give 2p in 60% yield. Quinoid gp exhibits a strong absorption in visible spectra ( Amax.‘ 600 nm, 6: 300,000), an indication of the long conjugation (Figure 12). o-Ioe@mod g4. .21. Figure 13. Radica; apion of 2b The oxidation of g9 was similarly done with lead peroxide in benzene to yield a green.compound 3 withflfihax. at 680 nm in visible spectrum. Solution of 3, in the absence of Pb02, lost its color presumably due to reductive aromati- zation. The interaction between the deorbitals of sulfur and the carbonyl group or C-8 proton of naphthalene might contribute to the distortion of the supposed planar molecule 3. (3 xx 4;:— ofto 3 fl 33 Some efforts have been given to construct a TTF analog with thiazolothiazole core (Figure 14). If the diol I} were to be oxidized to di-aldehyde, gg, condensation with 1,2- ethanedithio would give the corresponding bis-dithioacetal. Oxidation with DDQ would produce the TTF analog g}. The oxidation of I} was carried out in methylene chloride with MnOZ. Mass spectroscopic analysis indicated the major component in the reaction mixture to be mono-oxidation product. No di-aldehyde 3g was detected even with large excess of MnO . Home-(\iIZHHpH + OHC_<\:I:\>_ CHO 22 _/_\_3_ __ HOH ,c—<\:I:>—CH0 E: H:IE>_<:] one-{:15 CH0 [:X:I?:j 22 Figure 14. Thiazolothiazole-TTF. B. Synthesis of ThiazoloC5,h-d]thiazole dicationic salts. l. METHYLATION OF NITROGENS There are two main methods in the synthesis of radical cations. The majority of ion-radical salts was obtained by direct interaction of the neutral donor and acceptor in solution. Another general method for preparing highly conductive ion-radical salts involves the reduction of di- cationic salts of the donors. The synthesis of thiazolo- thiazole derivatives with double positive charges, therefore, would be interesting as a potential cation radical pre— cursor. In the literature, the only thiazolothiazole derivative with two positive charges was the 2,5-Bis(N,N'- dimethyl-fi-pyridyl)thiazolothiazole reported by Schimamura and co-workers55. (Figure 15). .S —— s —— 2 We case; Figure 15. 2,5-Bis(N,N'-dimethyl-4-pyridyl)thiazolo- £5,4-d]thiazol . We attempted to methylate the nitrogen on the parent rings of thiazolothiazole (Figure 16). 3# 35 MI a «I 24 X= a. CF3SOB— ’ fHa 7 Nil L J o-i— 9H3 Hooc-<\N || S\)—coom ——) (in —> §:IZ\> _/_2_ ‘51: 2x‘ 25 X: a. CFBSO3' — b. F303- _F_i_gure l6. Methylation of thiazolothiazole heterocycles 36 A mixture of 2,5-diphenylthiazolothiazole and methyl- fluorosulfonate or methyl trifluoromethanesulfonate in a sealed tube was warmed by a steam bath to give ;&g or 239 respectively. Compounds £53 and gjp were similarly prepared. g2; and 23p are stable for an indifinite period of time while 25; and gjp decomposed in contact with air. Treatment of an acetonitrile solution of 23g with a solution of LiTTCNQ' indicated no apparent formation of complex. Simple metal reduction of 23 with Mg or Zn resulted only in the recovery of the starting materials. No evidence was fbund to suggest the formation of radical mono-cation. 2. REACTION OF DITHIOOXAMIDE WITH PHOSGENE IMMONIUM CHLORIDE The chemistry of phosgene immonium chloride has been investigated extensively by Viehe56. Owing to its greater double bond polarization the immonium salt dichloromethylene- dimethylammonium chloride (phosgene immonium chloride) g6 should be more reactive than phosgen‘gz and methylchloro- fbrmimidoyl chloride g§ in nucleOphilic chloride substi- CH :=(3 ==C: 113C*——PQ==C<: r>\CJ \\C1 C1 26 27 _2_8_ 37 By reaction of phosgene immonium chloride gé with O-amino- phenol, O-aminothiOphenol and O-dihydroxylbenzene, compounds _2, 3_ and 3_ were obtained respectively. We. .1” >—~~. ;>=»?'Mez l 29 ‘19.; It was reported that when there were more than one acidic proton on either nucleophilic site, loss of the third molecule of'HCl was observed. In fact, in the cases of O-aminOphenol and O—aminothiOphenol, 22 and 39 were the only products isolated. Although the reactions were monitored carefully by NMR spectrosc0py until formation of the heterocycles were complete, no intermediate chloride salts has ever been detected. While phosgene immonium chloride was treated with phenol, the product we isolated was 3; according to the NMR analysis. ,JHQ ((3 ti CD) C%-N 6 5 13 \Wde 2.2. For our purpose of making doubly charged molecules, we needed to have the immonium chloride intact. However, when we went ahead and reacted phosgene immonium chloride with dithiooxamide we found only the recovered starting materials. II. BENZOEl,2-d:4,5—d']DIIMIDAZOLE As part of their continuing efforts in synthesizing TCNQ analogs, Wudl has reported35 the preparation of 2,7- bisdicyanoquinomethano-Z,7,H,H-quinazolinoE6,5,4-defj- ‘quinazoline (TCQQ) and its electrochemical behavior. We decided to carry out the same type of reaction by using 1,2,4,5-tetraaminobenzene (Figure 17), for the pro- jected synthesis of compound 33 appeared to be straight- forward. N CN + a >=< 1% H2" NHZ NC N N CN ,3 S 4Pfl3l ' I CH CH3 33 3 34 NC /N CN NCHNDIINHCN Figure 17. Projected synthesig of compgungmfifl 38 39 Treatment of 1,2,4,5—tetraaminobenzene tetrahydrochloride 33 with bisdithiomethyl dicyanoethylene in refluxing ethanol recovered only the starting materials. We contended that the protonated amines were not strong enough nucleo- phile to bring about the direct displacement. On the other hand, however, reaction of 33 with benzaldehyde or 3,5-di-t-butyl-h-hydroxylbenzaldehyde went smoothly to produce the desired products 33 and 3_ in fair yields. 11:: o :2: @4301: $0 4I+Cl i3. HOMEOH §__5_ 40 These twa reactions are in direct support of our earlier observations that the formation of aromatic- stabilized imines (Figure 6) is the decisive factor for the reaction to proceed. That even the less reactive hydro- chloride salt reacted quite nicely. Oxidation of 36 with lead peroxide in dimethyl sulfoxide gave a green solution ( Kmax. 750 nm), which seemed to contain quinoid fl». However, green color faded away rapidly without the presence of excess oxidizing agent. Further characterization were fruitless. A search in the literature revealed that the liber- ation of 1,2,4,5-tetraaminobenzene from tetrahydrochloride could be achieved by treatment with an ice-cold sodium hydroxide solution. When this procedure was followed and the resulting free amine protected under nitrogen, bis- dithiomethyl dicyanoethylene was added and refluxed in ethanol. An air sensitive product was isolated, which darkening slowly on the shelf but decomposed rapidly in solution. Spectroscopic analysis characterized this compound as 3§. CN H CN HZN N: , HZN u 38 41 Variation in reaction conditions has been attempted, but the tetra-functional benzene reacted only on one side of the molecule. Possibly due to the electron-withdrawing power of the cyano groups that deactivated the nucleOphilicity of the amino groups on the opposite side. III. SYNTHESIS OF 2,2'-(2,6-NAPHTHALENE- DIYLIDENE)BIS(1,3-DITHIOLE) The overlap of w-orbitals in a segregated stack of radical ions was believed to be the major factor governing the conducting behavior of the charge-transfer salt. Therefore, the unknown heterocycle 32 presents particularly interesting structural feature, since it combined in one molecule the p-quinodimethane of TCNQ with two 1,3-di- thiolidine moieties of TTF. and has more conjugated structure than TTF. By releasing one electron, TTQ molecule would restore two aromatic sextets. The incentive for TTQ to donate electron, in terms of gaining aromaticity, should we -> Ce TTQ (Tetrathioquinodimethane) Only one patent article in the literature described the structure of TTQSa. However, no identification, prepa- ration, as well as physical and chemical properties were given. Ueno59 and co-workers first reported the synthesis of the diphenyl derivative of TTQ. They found the compound was able to form complex with iodine, 3;. 42 43. “I: it 1:1 Fabre60 later revealed the structure of tetramethyl-TTQ:TCNQ complex, 3;. We decided to extend the conjugation of TTQ to that of naphthalene, £3. The relationship between 2,2'-(2,6-Naphthalenediylidene)bis(l,3-dithiole), CTetrathionaphtho-Z,6-quinodimethane (TTNQ)]. £3. the conjugated system of TTF and TTNQ should be expected to be similar to that between tetracyanoethylene (TONE) and TNAP, although the former two compounds are donors and the latter are acceptors. 2,6-Bis(bromomethyl)naphthalene was prepared according to the method of Diekmannug. Hitherto, the unknown tetra- thio-2,6-naphthalic acid 33 was successfully synthesized 59 and Beckeél, by the application of the methods of Ueno isolated as dipiperidinium salt 339 in 58% yield, mp 154-1560; IR: 1000 cm’1 (css'). The free tetrathio-2,6- naphthalic acid, from acidification of fifig, was unstable. It decomposed in the air to give a brown, unidentified solid. 44 H CH3———) er-c,c~/CHZBr Oi $.fiflzc”spcszcmc¢ -5 c.2gc:4 X= a. Na + b.< “H2 001 Figgre 18. Synthesis of;TTNQ derivatives 45 Diphenacyl ester 33 and dibutanone ester £6 were easily prepared in 87.2% and 30% yields by the reaction of salt £39 with phenacyl bromide and chlorobutanone respectively. The cyclization reactions were simply carried out by dissolving the ester 35 or gg in cold concentrated sulfuric acid, fluorosulfonic acid, trifluoromethane sulfonic acid or chlorosulfonic acid. The corresponding salts were isolated as orange or yellowish orange solid by diluting the reaction mixture with ethyl acetate. The structure of £2 and 38 were substantiated by spectral data. The absence of C20 and C=S bands in the IR spectra indicates the dehydrative cyclization is complete at both B-ketodithioester functions. The NMR spectrum of 38 has, in addition to the central aromatic protons (6H) at 6 8.2-8.u ppm, a singlet of 1,3-dithiolium ring protons (2H) at 6 8.8 ppm. And the spectrum of £2 has 12 allylic protons at 6 2.8 ppm plus 6 aromatic protons at around 7.6 ppm. Appearance of the above signals at considerable down-field indicates the highly electron deficient cationic structure of 32 and £8. The fluoroborate salts were pre- pared by refluxing the corresponding dithioester with phosporus pentasulfide, fluoroboric acid in acetic acid62 Elemental analysis indicates that 38b existed as a mono- hydrate whereas EZg complexed with one molecule of sulfuric acid. 46 Treatment of 43 with sodium methoxide in methanol yielded 42 in 72% yield (Figure 19). The 1,3-dithiol ring protons of 42 appeared at 6 6.4 ppm, a 2.4 ppm up- field shift from 48. Under the same reaction condition, +?X¢ ¢ . ©© S s s .Xé. 418 3.x. @639” Figure 19. Reaction of 48 with sodium methoxide 41 gave 39 in almost quantitative yield (Figure 20). \CH3 \CH3 flow 3043 0 CH3 H3 4 ._7 ‘g’socm Figure 20. Reaction of 47 with sodium methoxide The drastic up-field shift observed in going from 48 to 42 has a precedent in the chemical shift change recorded in diphenyl-TTQ system (Figure 21). 47 HEW. I: O 3;» GA OCH, oCH, PPM Figure 21. Chemical shift change Moationic 83116. to dimethoxy derivative. The reduction of dication 48 with lithium iodide or tetrabutylammonium iodide proceeded cleanly to give the black microcrystals of iodine complex 3; in 80% yield (Figure 22). A bathochromic shift of 210 nm was observed in the visible spectra of 5;. This shows clearly the highly conjugated structure of 5; instead of the simple dicationic structure like 5_. The fbrmation of 5; might be best 2|- Figure 22. Iodine complex of Diphepyl-TTHQ 48 63 explained by the electron transfer from iodide anion to the dication and further aggregation of iodine moiety. The ratio of the diphenyl derivative of TTNQ and iodine moiety in 5; was estimated to be 25:49 (n=l.96) by the 64 The starting material 48 was recovered elemental analysis. from complex 5; by treating with excess concentrated acid. To form a complex of diphenyl-TTNQ with TCNQ, 48b 60 was treated with Li+TCNQ' in acetonitrile , to give, after stirring overnight, a dark green solution which displayed strong TCNQ° plus TCNQ- absorption in UV spectra (Figure 23). It was known that when TTF formed 13¢ Q 20:350. ‘0 5., 2'rc~o= :81» 29.. Figgge 23. Complex of 48b with IIITCNQ: simple charge-transfer complex with TCNQ, no absorption due to neutral TCNQ was observed.9 Therefore, the presence of the neutral TCNQ in 53 represents the possibility of the formation of a mix-valence complex (TCNQ)+(TCNQ)°(TCNQ)': since example of oxidation-reduction between donor and 49 acceptor with concomitant formation of the complex salt 6 has been reported. 0r possibly it was simply the result of the incomplete redox reaction between 4§p and TCNQ'. More information is needed before a conclusion can be reached. An ideal situation would be to crystallize the complex 53. However, a suitable solvent has yet to be found. Attempts have been made to prepare the iodine complex of 41 by treatment of 41 in acetonitrile with tetrabutyl- ammonium iodide. The reaction mixture turned slightly darker, but no formation of any complex was realized. Therefore, an indirect route of complex formation was approached. A solution of dimethoxy derivative 59 was treated with aqueous HI solution followed by dilution with ethyl acetate to give a black-brown solid, presumably 54 (Figure 24). The salt 42g could be re-generated by treating 54 with concentrated sulfuric acid. (Further characterization f . $13— .I’ Si H+ 50 —> — ‘6 e 47a Sv d' "“_“ \ s 'n 54 Figuge 24; Iodine compIex o;;Tetramethyl-TTNQ 50 were, however, unsuccessful. Ueno and coworkers66 have reported the preparation of cyclohexa-Z,5-diene-l,4-diylidene-bis-l,3-benzodithiole (CBDT) 55 and tetraethylthioquinodimethane 56 through the oxidation of corresponding bis-dithioacetals (Figure 25). o . .0 e0 we ‘ C- S» 15 .S .S 55 as 0 SH 3’ as SEt Egs Sit SEt 56 Figure 252_Synthe§§s offiTTF analogs ii.§2§.i§- If this reaction were applicable to bis-1,2-dithioethane acetal of terephthaldehyde 52, the successful synthesis of 39 (TTQ) would be one step closer (Figure 26). CW: +> CW: Ii>=©=<§I 57 39 ~ — Figgre 26. PrOposedpsynthesgs of TTQ Unfortunately, 51 reatment of 52 with DDQ under several different reaction conditions resulted in only extensive darking with no trace of the product was detected. The dark material obtained might be the complex of 32 with DDQ. counter ion 48 41 mp yield mp yield H801: 246-249 (d) 85% > 3 5o 72. 5% CFBSOB- 227-230 (d) 95% 313-314 (d) 61.5% F803“ 256-257 (d) 90% --- --- BFu- 249-252 (d) 86.2% --- --— 01803' 268-270 (d) 73% >320 37. 5% TabIe 3. Melting points and yields of TTNQ deriyatives IV. PREPARATION OF 7,8-DICYANO-7,8-DINITROQUINODIMETHAN AND 11,12-DICYANO-11,12-DINITRONAPHTHO-2,6-QUINODIMETHAN At room temperature, TCNQ and TNAP form the two best conducting complexes with TTF. Both theory and experiment suggest that in TCNQ and TNAP the odd electron resides to a large extent on the terminal portions of the molecule containing the four electron attracting nitrile groups. The replacement of nitrile group by nitro group was of special interest to us since nitro groups are stronger electron-withdrawing groups and therefore might better stabilize the resulting radical anion, provided the nitro groups stay in the plane of the aromatic ring. On the other hand, the planarity requirement is not very stringent. The intermolecular interaction of compounds in the conducting stack of charge-transfer complex is realized through inter- action of molecular w-orbitals formed by wave function of unpaired electrons. Despite the directional character of the w-orbitals, they are delocalized and belong, to a certain extent, to the entire (20-30 A) organic molecule. This is one of the reason why deviations from planarity are permissible. Thus, fulvalene rings in the TTF molecule are slightly distorted, deviating from planarity by 2° 67'68: the cyano groups in TCNQ are also arranged slightly out of the ring plane.69 Small substituents in TTF and TCNQ do not destroy the metallic state in the corresponding 52 53 complexes.69 For example, at room temperature, HMTTFxTCNQ (Table 1, Entry 3) has the same conductivity as TTFzTCNQ whereas the conductivity of’HMTSeF:TCNQ (Table 1, Entry 8) is three times higher than that of its parent compound's (Table 1, Entry 10). 3 The first synthetic approach to 5§ made use of the classical route utilized in the synthesis of'TCNQ8 (Figure 27). Surprisingly, the attempted condensation of 1,4- cyclohexanedione with nitroacetonitrile has failed, due Cbl COI H a — _ (N==<::::>==C) +'CI;JC$02 “'5 (”q (‘ii') ("V Figure 27. Cla§sical route to TCNQ mostly to the instability of nitroacetonitrile, which was generated in situ by the dehydration of methazoic acid.7o’71 A search in the literature revealed that the condensation of nitroacetonitrile with some similar systems e.g., 2,6-dimethyl-y-pyrone and 2,6-dimethyl-y-thiapyrone have been reported7l, but only in very low yields (8.5% and 4% respectively). The failure of this bifunctional condensation prompted us to consider a stepwise introduction of cyano and nitro groups (Figure 28). The reaction of an alkyl nitrate with an active methylene compound in the 54 ch‘.-CH3 —9 ‘BrH2C-.—CHZBI' -—) NCH2C-.—¢H2CN 3 . Oz _N02 ,) ozu>=©=[RC(CN)= N02} '+ R’OH Fi re 2 . Nitration_gf active methylene compound 56 carbanion of the active methylene compound (step 1). 0n the other hand, it has been pointed out that alkyl nitrates undergo several reactions with strong alkaline reagents.77’78 This becomes especially important in di- nitration reactions. If it is assumed that the nitration of compounds possessing two active methylene groups occurs stepwise. The presence of excess base required for the formation of the carbanion of the second methylene group would enhance the decomposition of the alkyl nitrate. Fortunately, for our purposes, both potassium t-butoxide and potassium ethoxide worked equally well. Amyl nitrate was found to be a satisfactory nitrating agent and was used throughout. Upon consideration of step 1, it is suspected that the presence of hydroxylic solvents would not favor the formation of the carbanion of active methylene compounds. However, in nitration of p-bis(cyanomethyl)benzene with potassium ethoxide, we have used as much as ten times excess of ethyl alcohol but did not suffer any loss in yield. Under identical reaction condition, the dipotassium salt of 2,6-bis(cyanonitromethyl)naphthalene Q} was syn- thesized in almost quantitative yield as a yellow plates with metallic shining. é} exhibits the same characteristic N02“ absorption in IR spectra. 57 CN 11 NC no, a: 5.4. Figure 30. Acidification of di otassium salt 6 Conversion of the salts go and g; to the corresponding neutral cyanonitro compounds was not satisfactory, since only intractable solid were obtained after acidification. This was in accord with the observations made by Feuer and Savides76 that most free a,a'-dinitrodinitriles were unstable at room temperature. We found that the protons on C-7 and C-8 of Qi are more acidic than acetic acid. These highly acidic protons might protonate the nitrile groups which were then attacked by the carbanionic nucleophiles. This explained the thermal instability of Qi and fig. To further characterize the structure of fig and to attempt the preparation of quinoid z, the following reactions were carried out (Figure 31). Treatment of dipotassium salt fig with bromine in C01” resulted only in extensive decomposition. Acidification with concentrated HCl followed immediately by bromine water produced a yellow, water in- soluble solid whose structure was presumed to be 1. However, it was observed that the CEEN stretching in .. 11°;— A ...?f’fim 2K’ 2“9’ 6O -— Figggejl. Reactions of salt 60 l was very weak and could not be the region of 2200 cm- easily detected. A similar observation has been made by Kitson and Griffith.79 They found that the intensity of this band decreased considerable when oxygen-containing groups were attached to the carbon atom alpha to the 59 nitrile group. The CIN band was completely absent in the spectrum of the supposed a,a'-diacetoxy-a,d'-dimethyl- succinonitrile. The lack of nitrile absorption in IR certainly did not help us in determining the structure of our product. Furthermore, this yellow solid was unstable, decomposing with the evolution of oxides of nitrogen. Therefore, further characterization were unsuccessful and inconclusive. If one considered the negative charge as mainly distributed on the oxygen of the nitro group, the difficulty in oxidation was not unexpected. The disilver salt éj was prepared by treating an aqueous solution of fig with AgN03.78 Although éj is someWhat light sensitive, as most silver salts are, a satisfactory elemental analysis was obtained. The dichloro derivative, éé, was prepared by the direct chlorination of.§9 with chlorine gas.78 This compound melted above 300° with a color change from yellow to white at 110°. It exhibits the characteristic nitrile l l , nitro absorption at 1580-1620 cm- and chloride absorption at 720 cm-1. absorption at 2200 cm- In direct support of its structure, g9 was hydrolyzed 60 in boiling alkaline aqueous solution, acidification led to spontaneous decarboxylation and yielded p-bis(nitromethyl)- benzene80(Figure 32). 02 N0; NO; NO- “I u u n r 2 OZNHZC‘.“CH2N02 {‘——’ H02N= HC‘.’ CH =N02H 67 Figure_32.qfiydroiysis of dipotassium salt ég Same approaches were applied to the dipotassium salt é}. But the difficulties we encountered with.§9 also occured to ‘é3, namely, lack of strong GEN absorption to prove the existence of nitrile: extremely low solubility limits the use of 13C NMR: no parent peak in mass spectrometry due to thermal decomposition. Treatment of Q} with silver nitrate in aqueous solution did result in the precipitation of a light green disilver salt. But it was less stable than éj and decomposed rapidly when exposed to the air. 61 Nevertheless, in each case, a characteristic nitro absorption appeared in the 1560-1670 cm"1 region and the disappearance of the original N02- absorption suggested that the carbanion has been oxidized into a neutral species. Adequate condition for the oxidation of QQ and Q3 to the quinoids Z and 8 has yet to be found. V. PREPARATION OF 4-PHENYL-5-AMINO ISOXAZOLE To deal with the difficulties we had encountered in introducing a second nitrile into the same carbon atom50, we searched for an alternative to replace the traditional SN-2 displacement. Two approaches we attempted but proved fruitless were a) The replacement of diazo-group by CN' b) Reaction of enamine with sulfimide (Figure 33). CN R-CHZCN -7$/—> R-E-CN fl R-CH/ N2 \CN-l» N21 CN R—CH/ ‘NMez \CN C3-C7 t Figure 33. Proposed syntheses of d,a-dinitrile Hydrogen halides are known to replace the diazo- group in a-diazo ketones. But when we treated phenylaceto- nitrile with tosylazide, no expected diazo—nitrile compound was obtained. It was also well documented that reaction of free 81 We used sulfimide with aldehyde provided nitriles. enamines as hidden aldehydes, but detected no desired products. We then turned our attention to the use of HOSA82 (Hydroxyamine-O-sulfonic acid), since an additional 62 63 nitrile synthesis using HOSA has recently been reported83 (Figure 34). The method would be extremely useful for us COOEt COOEt l C==H -NMe2 -———->. O‘CH “ON OZN. 0 == 0 -—NM422 ———>02N@-CH2—‘ 0“ Figure 34. Synthesis of nitrile by using HQ§A since enamines could be prepared from readily available mono-cyano compounds and bis-dimethylamino-t-butoxymathane84 Brederek's reagent)(Figure 35). i:——hflfle2 X: a) p-l 02 b) H 68W XML: Figure 35. Synthesis of‘d,g- -dinitrile with HOSA 64 Surprisingly, when this method was used on enamine §§, an unexpected reaction was observed. Treatment of p-nitro- phenylacetonitrile with Brederek's reagent affored the enamine §§g in 83% yield as a bright yellow solid, mp 184- l85°. Enamine égg when refluxed with HOSA in ethanol or isopropanol yielded an orange solid whose structure was assigned as fig; based on the spectroscopic data. When heated with acetic anhydride fig; yielded acetamide 22g in 55% yield (Figure 36). (:-huwez ...li-.~ 680 Figure_36.prclization reaction of 68a with HOSA Under similar reaction condition 68b afforded 62b as a white solid in 32% yield. Literature search revealed that a-cyanoketo compound cyclized to give amino-isoxazole when reacted with hydroxylamineas. (Figure 37). lat-L“ cm: Figure 37. Cyciization oienamine 68b to igoxazole 69b For comparison, enamine éég was refluxed with hydroxyl- amine and égg was isolated. The behavior of enamine as a carbonyl function was what we have expected. However, the detailed reaction mechanism was not clear. The reaction was run in either aqueous alcohol or in anhydrous alcohol. Both gave the same product in comparable yields. Water was involved in both cases during work-up to remove excess HOSA. One would expect HOSA to be hydrolyzed before reacted with the enamines. However, in aqueous alcohol the re- action of fig with HOSA or NHZOHOHCl required one hour at refluxing temperature to go to completion. While in anhydrous alcohol treatment of the mixture with water at room temperature after reflux generated the product 66 instantaneously. Based on the difference in reaction rate, 0- N(M-e)2 HOSA I}! —>AR- 0 CN __9 U AR- 0- CN 3 5} M92; ('3— NH- OSO3H AR- C-CN 0H _§ AR-g-CN ——> AR {5 If —} AR \N C-NH-OH 0 \ 0' (NH Figure 38. Mechanism gi'cyclization of 68 we proposed a reaction mechanism (Figure 38) in which HOSA attacks the enamine in the rate determining step followed by hydrolysis and cyclization to 5-amino-isoxazole. Moreover, in a control experiment, when water was used as the reaction medium87, there was recovered only the starting enamines. This appeared to be the result of low solubility of enamine in water. II- ;_pUC_iON 0F TCNQ WITH sopIUM AZIDE Treatment of TCNQ with sodium azide in refluxing acetonitrile afforded a deep blue solution, from which purple crystals precipitated upon cooling. This purple product was identified as TCNQ’Na? by IR and UV-Visible spectra analysis.88 The deep blue acetonitrile filtrate was taken into near dryness and was purified by preparative TLC to give a deep blue solid which displayed a strong absorption at 7\max.585 nm in visible spectra with no trace of either TCNQ or TCNQ‘. This reaction demonstrated NC CN NC ‘N NC .. CN NN3 \ + + . ’ .7_0_ NC CN N3 the first example of the use of sodium azide as a reducing agent. We assumed the structure of the blue residue as the azide addition product 29. Although the elemental analysis of ZQ was non-informative, addition of one equivalent of trifluoroacetic acid yielded 55% of TCNQ’NaT, but the formation of the blue solution was indeed completely suppressed. Instead, the filtrate was identified as un- reacted TCNQ. 67 "‘7 "-’\T‘.'."fi"'"1‘-r H V. . pl \ _;* {a} L .t...‘ DU; id inn GENERAL PROCEDURES Melting points were determined on a Thomas Hoover Uni- melt melting point apparatus and are uncorrected. 1H HER spectra were taken on a Varian T-60 or Eruker JM-250 spectrometer and are reported in parts per million from 1 a. 3C th spectra internal tetramethylsilane on the 6 scale. were determined on a Varian CFT-20 spectrometer. Infrared spectra were taken with a Perkin-Elmer 237B instrument. Ultraviolet spectra were obtained on a Unicam SP-BOO spectrometer. Mass spectra were recorded on a Hitachi Perkin-Elmer RMU-6 or a Finnagan #000 instrument. Elemental analyses were performed by Spang Microanalytical Laboratory, Eagle Harbor, Michigan and Galbraith Laboratories, Inc., Knoxville, Tennessee. C313(methylthig)methylenejmalononitrile (i9, R=SCH3). A number of similar compounds have been prepared by 57 Gompper and Topfe whose directions were followed with some modifications. A solution of 11.2 g potassium hydroxide (0.2 moles) in 60 ml of methanol was cooled to 0°C with ice-salt bath. / '9 To this stirred solution was added portionwise 6.6 g of malononitrile (0.1 moles) so that the temperature would not go over 0°. Six milliliter of carbon disulfide was then added in one portion. When the yellow precipitate of potassium dithiolate started forming, 13 ml of methyl iodide (0.2 moles) was added drOpwise over a period of 30 minutes. After the addition was complete, the reaction mixture was allowed to warm to room temperature with stirring and diluted with 350 ml of cold water. The suspension was allowed to stand for a few hours to complete the pre- cipitation. The yellow product was collected on a Buchner funnel and recrystallized from methanol to give 12.7 g (75%) of [bis(methy1thio)methylenejmalononitrile as color- less needle-like crystals: mp 79.5-80°C; PER (acetone-d6): 5 ppm 2.8 (s); IR (nujol): 2190, 1320, 1210, 960, 925, 870 cm'l; mass spectrum m/e: 170 (parent), 155, 130, 123, 109, 98, 86, 79, 47. Condensation of dithiooxamide with aromatic aldehydes- Dithiooxamide (rubeanic acid) obtained from commercial sources was purified by dissolving in a large volume of hot ethanol, filtering to remove the dark, insoluble impurities. Concentration of the filtrate gave orange crystals of dithiooxamide. The aldehydes were recrystallized or redistilled before use. The condensation reaction does not appear to be particularly sensitive to small amount of impurity in either reactant. 70 The mixed reactants were placed in a round bottom flask fitted with a Dean—Stark apparatus to trap the small amount of water formed. 2,5-Bi§(2-fUryl)thiazolofl5,4-d]thiazole, ii. Dithiooxamide (20 g, 0.167 moles), 60 g of phenol, and 200 g (2.08 moles) of freshly distilled furfural were heated in an oil bath at l80-200° for 45 min. After standing overnight the dark crystalline product was collected with suction and washed with several portions of ethanol, ether, and hexane. The crude product was heated under reflux with l l. of chloroform and the filtered solution was taken to dryness on rotory evaporator to yield 16 g (#O%) of greenish yellow needles; mp 238-239° (lit. mp 238- 200°)41. DiphenylthiazoloC5,4-d3thigzole A mixture of 2 g of dithiooxamide (0.017 moles) and 17.7 g of benzaldehyde (0.17 moles) was heated in an oil bath to ca. 200° for 45 min. The reaction mixture was cooled to room temperature and diluted with acetone. The crude product was collected and recrystallized from benzene to give 3.3 g (77%) of faintly yellow crystals: mp 208-209°. (lit. mp 209-210°)5#. 71 2,5—Bis(2-hydroxy-l-naphthyl)thiazolol§,h-dlthiazole,.29. This compound was prepared in the same manner as diphenylthiazoloE5,h-d]thiazole but using DMF (dimethyl- formamide) as solvent. Recrystallization from pyridine gave 65% yield of light yellow crystals: mp 335-337° (lit. mp 339-3#2°). 2,5-Bis(#-hvdroxyphepyl)thiazoloE5,4-thhiazole,‘ié. A mixture of l g of dithiooxamide (8.3 mmoles) and 2.h8 g of p-hydroxybenzaldehyde (20.3 mmoles) in 10 ml DMF (dimethylfbrmamide) was refluxed fer 15 hours then allowed to cool to room temperature. The resulting dark solution was diluted with water and the precipitate collected on Buchner funnel. The dark crystalline product was washed extensively with ethanol, recrystallized from cyclohexanone and washed with ethanol again to give 0.8 g (31%) of the desired product as yellow solid: mp 365-370° (lit. mp 369-373°)“°. 2, 5-Bis( 3, 5-di-tept-bmty1-u-hwenyi) thiazolotifll- thiazole,‘i2. Dithiooxamide (1.2 g) and 3,5-di-tert-butyl-h-hydroxy- benzaldehyde (4.7 g) were refluxed in 50 m1 of ethylene glycol (oil-bath temperature 200°) for 15 min. After cooling, the reaction mixture was poured into 300 m1 of 72 water. The precipitates which formed were collected on a Buchner funnel and washed with methanol and hexane. One recrystallization from chlorofbrm/petroleum ether gave 3.1 g (56%) of light yellow crystals: mp 313-314° (lit. mp 314--315°).53 This compound oxidezed to a pale-green solid when exposed to the air: PMR (CD013): 6 ppm 1.5 (s, 18 H), 5.4 (s, l H). 7.6 (s, 2 H). ‘gijzihiazoloflj,4-d]thiazoledigapboxylic agig,.ig. A suspension of 17 g of 2,5-bis(2-fury1)thiazolo- thiazole (0.062 moles) and 475 ml of pyridine were heated on a steam bath with rapid stirring until the solid had dissolved. After cooling to 70°, 75 ml of water was added to produce an uniform suspension of fine crystals. The mixture was cooled to 15-20° and held at this temperature While 104 g (0.66 moles) of’powdered potassium permanganate and 70 ml of water were added, with continued rapid stirring. After the permanganate had been added, the temperature of the mixture was allowed to rise slowly to 40° and mains tained there for 12 hours. Caution.has to be taken to avoid overheating immediately after the removal of the cooling bath. The mixture was then cooled and l g of sodium bisulfite and 100 ml of water were introduced. The brown precipitate was collected with suction on a large filter, washed sparingly with water, and squeezed dry 73 with a filter dam. The filtrate might be discarded with little loss of product since the sparingly soluble potassium salt of the acid was effectively salted out by the high concentration of potassium ions in the system. The filter cake was boiled with l 1. of water and the extract filtered hot with suction. Two further extractions of the residual manganese dioxide were_made with smaller volumes of hot water. The colorless filtrates were acidified with concentrated HCl through addition funnel and the precipitated acid was collected. The air-dried material weighted 7.6 g (50%): mp 212-213° (dec) (lit. mp 214° dec): IR (nhjol): 3450, 3350, 1690, 1410, 1320, 1310, 1280, 1112, 750 em‘lg Mass spectrum.m/e: 230 (parent), 186, 142, 115, 88, 70, 44. This compound was Shown to exist as a dihydrateul. ThiazoloE5,4—d]thiazole 2,5-Thiazolothiazoledicarboxylic acid dihydrate (600 mg, 2.3 mmoles) was heated under reflux with 100 m1 of absolute ethanol fOr 14 hours. The suspension turned clear after 30 minutes of heating and the product reprecipitated gradually. The solvent was removed by rotory evaporator. There remained 298 mg (91%) of crystals of thiazolothiazole. The analytical sample was purified by recrystallization from ethanol: mp 150-l5l° (lit. mp 150°)41; Mass spectrum m/e: 74 142 (parent), 115, 100, 88, 70, 44. Dimethyl 2,5-thiazolol5,4—dIthiazoledicarboxvlate This dimethyl ester was prepared by the method of ulin 87.6% yield: mp 248.5—249.5° (lit. mp 249- 250°): Mass spectrum m/e: 258 (parent), 227, 200, 114, 88, 70, Johnson's 59. The diethyl ester was prepared in the same manner but in lower yield (42%, mp l39-l40°: lit. mp 140-141°) due to the facile decarboxylation of 2,5-dicarboxylic acid into its parent compound, thiazolothiazole: 130 NMR (CD013): 0 ppm 162.05, 159.36, 154.22, 63.22, 14.00. 2 -Bis h dro eth 1 thiazolo 4-d thiazole, i3. Into a three-neck round bottom flask equipped with condenser, addition funnel and thermometer was charged 0.572 g (2 mmoles) of diethyl 2,5-thiazolothiazoledicarboxy- late in 25 ml dry tetrahydrofurane and cooled in ice-bath until it reached 10°. To this stirred solution, 8 m1 of 1 M Super-Hydride (lithium triethylborohydride, 8 mmoles) was added dropwise so that the temperature was kept below 15°. After the addition was complete, the brown solution was further stirred at 10° f0r one hour. Five milliliters of water and 15 ml of 3 N HCl were added slowly to quench the exm uncle and dar] wit] laye morc em; 13 (bn 2‘03 75 excess hydride. The resulting brown solution was distilled under vacuum (bath temperature 35°) to remove triethylborane and tetrahydrofurane. After the volatiles were removed, the dark brown residue was diluted with ethyl ether and treated with 25% potassium carbonate aqueous solution. The organic layer was separated and the aqueous layer was extracted with more ether. The combined ether extracts was dried (MgSOu) and evaporated on rotory evaporator to give 0.278 g of the diol i3 (69%) as a yellow solid; mp l67-l69°3 IR (nujol): 3200 (broad), 1350, 1160, 1060, 720; PMR (DMSO-dé): 5 ppm 4.8 (d, 4 a), 6.2 (t, 2 a). When 0 0 was added: 4.8 (s): 13c 2 NMR (DMSO-dé): 6 ppm 175.52, 148.39, 61.33. Mass spectrum m/e: 202 (parent), 173, 143, 114, 102, 88, 84, 70, 57, 45, 31. This compound could also be prepared from the dimethyl ester but in lower yield (47%). 2,5—Bis(chloromethyl)thiazolo[5,4-djthiazole,,_4. Diol i3 (0.202 g, l mmole) was refluxed with 5 ml of thionyl chloride for two hours. After cooling, the reaction mixture was poured carefully into 25 m1 of cold water with stirring. The precipitates fermed were collected on Buchner funnel and washed with water to give 0.23 g of the bis- (chloromethyl) derivative (96%) as brown plates; mp 123-125°3 PMR (CD013): 6 ppm 4.8 (s): Mass spectrum m/e: 238 (parent), 203, 168, 88, 84. Anal. Calcd. for C6H4N232012‘C’ 30.13%: H. p) {=- .L bu 10 la (b In l..L 76 H, 1.69%. Found: 0, 30.27%: H, 1.72%. 2,5-Bi§(2,5-cxclohexadiene-3,5-di-t-butvi-1-diyliden - 4-oxo)thiazolo|5,4-dlthiazole,lip (R = t-butyl). To a stirred solution of 0.5 g of 2,5-bis(3,5-di-t- butyl-4-hydroxyphenyl)thiazolothiazole,ig (0.9 mmoles) in 100 ml of methylene chloride was added 2.2 g (9 mmoles) of lead peroxide. The mixture was refluxed for 5 hours affbrding a deep blue solution. The solvent was removed in vacuo and the residue treated with 50 ml of’chlorofbrm and filtered. The solvent of the filtrate was removed to give 0.3 g of the deep blue quinoid compound 2p (60%): mp 289.5-290.5°: PMR (CDClB): 6 ppm 1.33 (s, 18 H), 7.8 (d, 2 H); IR (nujol): 1630, 1600, 1375, 1250, 1150, 1080, 1030, 910, 830 em‘l; Mass spectrum m/e: 548 (parent), 533, 493, 275, 260, 246, 216, 105; UV-Visible (CH2012):7\hax 555nm (e: 9.1x10“), 602 nm (e: 3.0x105). Oxidation of 2 -Bis 2-h droxy-l-naphthyl)thiazoloE5,4-dJ- thiazole 20. This compound was oxidized in the same manner as above using benzene as the solvent to give the quinoid 3 as a green solid: UV-Visible (CHCI3)’7‘max.675 nm. Compound 3 became discolored without the presence of excess lead peroxide. Therefore, further characterization was not possible. 77 Oxidation of 2,5-Bis(p-hydroxyphenyl)thiazolo[5,4-d]thig§ole i6. This compound was not soluble in.most organic solvents and consequently the oxidation with lead peroxide in organic solvent failed. Instead, 0.25 g of i6 (0.77 mmoles) was dis- solved in 30 ml of diluted NaOH aqueous solution and treated with 1.0 g of'KBFe(CN)6 (3.08 mmoles). After 2 hours of stirring at room temperature, this suspension of fine green solid was centrifuged and the precipitate collected on a Buchner funnel to give 0.17 g (69%) of the corresponding quinoid compound 2g. Very insoluble for spectroscopic determinations: Mass spectrum m/e: 324 (parent). Without the presence of excess K3Fe(CN)6 this compound was in equi- librium with the "phenol" form i6, as shown by the increasing intensity at m/e 326. N,N'-Dimethyl-2,5-diphenylthiazolofi5,4-d]thiazole tri- iiuoromethanesulfonatgpsalt, 24a. 2,5-Diphenylthiazolothiazole (l g, 3.4 mmoles) and 5 ml of methyl trifluoromethanesulfonate were placed in a sealed tube and heated on a steam bath overnight, until the dissolution was complete. The contents in the tube were cooled to room temperature and diluted with ethyl ether to precipitate a white solid. The solid product was collected *9 \a) (1") In 78 and washed with ethyl ether and alcohol. Recrystallization from trifluoroacetic and ethyl acetate afforded 1.96 g of the product as white shining crystals (92.5%): mp 260° (dec): PMR (CFBCOOH): 6 ppm 4.53 (s, 6 H), 7.8 (s, 10 H): IR (nujol): 3050, 1600, 1520, 1485, 1275, 1225, 1150, 1030, 1000, 945, 880, 855, 760, 700. Anal. Calcd. for CZOH16NZS4F6O6: C, 38.58%: H, 2.59%. Found: C, 38.62%; H, 2.62%. The fluorosulfonate salt 249 was similarily prepared by treatment with methyl fluorosulfonate in 90% yield. WJN'-Dimethylthiazoi9[5,4-d]thiazole trifluorqmethane- guifonate salt 25a ThiazoloC5,4-d]thiazole (200 mg, 1.4 mmoles) and 3 ml methyl trifluoromethanesulfonate were sealed in a glass tube and heated on steam bath for 1 hour. Starting material dissolved and a white plate-like crystal formed. After cooling, the product was collected on Buchner funnel and washed with dry ether. Caution should be taken not to draw too much air through the funnel for the product 25g appeared to be air sensitive. The product was dried under vacuum to give 300 mg (45.3%) of gig; mp 258-260°(dec); PMR (CF3C00H): 6 ppm 4.2 (s, 6 H), 10 (s, 2 H). The corresponding fluorosulfonate salt 259 was prepared in the same manner but was extremely air sensitive and decomposed rapidly. Mm 2._ ac 79 . . . 56b NiN-Dimethylamino-tr1phenoxymethane,.32 A mixture of 1.625 g (0.01 moles) of phosgene immonium chloride and 1.88 g of phenol (0.02 moles) in 10 ml of methylene chloride was refluxed for 1 hour over the steam bath until the evolution of hydrogen chloride gas had ceased. The solvent was removed under vacuum to afford an oil which precipitated upon addition of ethyl ether: PMR (acetone-d6): 6 ppm 2.95 (s. 6 H), 6.6-7.3 (m, 15 H). mpj>300°(dec). 2,6-Qiphenyl—Nl(3)HL_5(7)H-benzo[1323d:4,5-d']-diimidazole,.35 In a 50 ml round bottom flask heated with oil bath (bath temperature 190-200°) was refluxed 1.42 g of l,2,4,5- tetraaminobenzeneo4HC1 (5 mmoles) in 10.6 g of benzaldehyde (0.1 moles) for 3 hours. After cooling to room temperature the reaction.mixture was diluted with large quantity of ethanol and a yellow solid precipitated. The solid was collected on Buchner funnel and washed with water, ethanol and ether to give 0.3 g (20%) of the title compound 35; mp)350°: IR (nujol): 3675, 1610, 1570, 1500, 1250, 1000, 855 cm-l: Mass spectrum m/e: 310 (parent), 155, 122, 105, 91, 77. 80 2,6-Bis(3,5-di-tert-buty1-4-hydroxyphenyll-benzofii32-d:4L5- d ' Idiimidazolei 3p. 1,2,4,5-Tetraaminobenzeneo4H01 (1.42 g, 5 mmoles) and 2.34 g of 3,5-di-t-butyl-4-hydroxybenzaldehyde (10 mmoles) were mixed in 25 m1 of ethylene glycol and heated under reflux for 30 minutes. The cooled reaction mixture was poured into 200 m1 of water and the precipitate formed was collected and washed with water to give 1.1 g of off-white solid 36 (36%): mp)>360°. IR (nujol): 3400, 1610, 1420, 1245, 1115, 890 cm-1: Mass spectrum m/e: 566 (parent), 555, 499, 382, 338, 298, 218, 103, 161, 82. Oxidation of 2,6-Bis(335-di;t-buty1-4-hydroxypheny1)- benzoEl,2-d:4,5-d']diimidazole, 36 A solution of compound 36 in dimethyl sulfoxide was treated with excess lead peroxide. A deep green solution of quinoid 4 was obtained. Excess lead peroxide was filtered and the UV-Visible spectrum of the green filtrate was taken;7\ max.(DMSO)‘ 750 nm. However, the solution lost its color gradually. Addition of’PbO2 regenerated the color. P11 Ifi 81 Reaction of 1,234,5-tetraaminobenzene with bigdithiomethyi- dicyanoethyleng. Under a blanket of nitrogen, 2.84 g (10 mmoles) of tetraaminobenzene-4H01 in cool, boiled oxygen-free water was treated with ice-cold 15% sodium hydroxide solution. The precipitate was filtered with suction under nitrogen, washed with ice-cold ethanol and suspended in 30 m1 of dry n-butanol. To this stirred suspension was added 3.4 g of bis(dithiomethy1)- dicyanoethylene (20 mmoles) and heated under reflux for 60 minutes. The hot reaction mixture was filtered to collect 1.9 g of slightly yellow solid (89.6%), mpf>300°3 IR (nujol): 3405, 2210, 2170, 1635 cm-l: Mass spectrum m/e: 212 (parent), 170, 94, 79, 66. _216-Bis(bromomethyl)naphthaiene A mixture of 50 g (0.32 moles) of 2,6-dimethylnaphalene, 120 g (0.675 moles) of recrystallized N-bromosuccinimide, 800 m1 of carbon tetrachloride, and 200 mg of benzoyl . peroxide was stirred at reflux under nitrogen for 4 hours. Then 500 mg of azobisisobutyronitrile was added, and the mixture was further stirred and reflux for 24 hours. The resulting thick mixture was cooled and filtered. The filter cake was washed well with water. One recrystallization from dioxane gave 59 g of white crystals (58%): 82 O ' o “'9 89 mp 181-183 (lit. mp 182.5-184 ) . Brown suggested the use of a sun lamp as the source of irradiation and heat. His method was feund to give more by-products as brown residues and was satisfactory. 2L6-Uaphthalenedicarbodithioic acid, dipiperidiniumigalt 44b To a solution of sodium methoxide (0.08 moles) in 60 ml dry methanol was added 2.56 g (0.08 moles) of elemental sulfur and refluxed for two hours. 5.8 g of 2,6-bis(bromo- methyl)naphtha1ene (0.0185 moles) was then added to the above refluxing mixture during a period of one hour through a solid addition funnel. The resulting mixture was further refluxed for 7 hours. The reaction mixture, which contained the disodium salt of the tetrathio acid 44g was cooled and the solvent was removed by rotory evaporator. The residue was dissolved in water and filtered to remove any undesirable impurities. The deep violet filtrate was acidified with 10% HCl. The solid which precipitated upon acidification was dissolved immediately in 1.5 1. of chloroform and filtered. The resulting red solution was treated with 3.15 g of piperidine (0.037 moles) and taken to near dryness in vacuo to give, after suction filtration, 4.8 g of olive-colored dipiperidinium salt 54p (yield 57.8%): mp 154-156°; IR (nujol): 1000 cm-l(CSS-). 83. Diphenacyl 2,6-naphthalenedicarbodithioate.45 To a stirred suspension of 0.45 g of dipiperidinium salt of 2,6-naphtha1enedicarbodithioic acid flflp (1 mmole) in 100 m1 of methylene chloride was added dropwise a solution of 0.398 g of phenacyl bromide (2 mmoles) in 25 m1 of methylene chloride at room temperature. After 1 hour of stirring the suspension turned to a clear red solution and the product began forming thereafter. The solvent was removed in vacuo after a total of 3 hours reaction time. Recrystallization from chloroform-hexanes yielded 0.45 g of 45 as a red crystals (87.2%): mp l98-100°; IR (nujol): 1690 (0:0), 1205 (C=S), 1050, 880. Anal. Calcd. for C28H200284: C, 65.08%; H, 3.90%. Found: C, 64.93%; H, 3.86%. Bis(1-methy1-2-oxopropyl) 2L6-naphthalenedicarbodithioate‘46 To a stirred suspension of 0.45 g (l mmole) of the dipiperidinium salt of 2,6-naphthalenedicarbodithioic acid, 44p in 100 m1 of dry ethanol was added 0.21 g (2 mmoles) of 3-chloro-2-butanone in 10 ml of ethanol. The brown suspension dissolved with spontaneous precipitation of orange crystals. After 1 hour of stirring at room temperature, the solvent was removed by rotory evaporator and the residue was recrystallized from chloroform to give 0.13 g of red crystals 46 (30%): mp 180-181°: IR (nujol): 84 1700 (C=O), 1325, 1220 (C=S), 1175. 1150, 1130, 1045: 925, 880, 815, 800 cm-l: UV (CHClB):7\max. 3’45. 505 nm: PMR (00013): 6 ppm 1.65 (d, 6 H), 2.4 (s, 6 H), 4.9 (q. 2 H), 7.6-8.4 (m, 6 H). Anal. Calcd. for C20H203402‘ C. 57.11%; H, 4.79%. Found: C, 57.14%: H, 4.93%. Bistrifluoromethanegulfonate salt of diphenyi3TTWQ, 48b Into 10 m1 cold concentrated trifluoromethanesulfonic acid carefully dissolved 0.1 g of phenacyl 2,6-naphthalene- dicarbodithioate 45. The cold solution was poured into 50 ml of ethyl acetate to give, after washing with ethyl acetate, 0.11 g (84%) of the bis-trifluoromethanesulfonate salt 483. mp 246-249°(dec). IR spectra indicated the disappearance of'C-O and CIS bands: PMR (CF3COOH): 6 ppm 8.8 (s, 2 H), 8.1-8.8 (m, 6 H), 7.6-8.0 (m, 10 H); UV (CHBCN):7\maX. 465, 340, 250, 225 nm. Anal. Calcd. for C3OH18F6S6O6'H203 C, 45.10%: H, 2.52%. Found: C, 45.08%: H, 2.62%. The bisfluorosulfbnate salt 489, bisbisulfate salt .48g.and bischlorosulfoante salt 48g were similarly prepared by dissolving the ester in the corresponding cold acids. Bisfluoroborgte salt of diphepyl-TTNQ, 48d Phenacyl ester 45 (1.2 g, 2.3 mmoles), 0.95 ml of .fluoroboric acid (48% aqueous solution) and 0.6 g of 85 P285 (2.? mmoles) were refluxed in 14 m1 of glacial acetic acid for 20 hours. An orange suspension was obtained at the end of the heating period. The solvent was removed in vacuo and the residue was diluted with 50 m1 of ethanol to yield 1.3 g (86.2%) of 48d as orange solid; mp 250°(dec). Bisbisulfate salt of tetramethvi:TTNQ, 42a. Carefully dissolved 0.1 g of the di-butanone ester of the 2,6-naphthalenedicarbodithioic acid 46 in the cold sulfuric acid. The solution formed was then poured into a beaker of 50 m1 ethyl acetate with stirring to precipitate the salt 41a in 72.5% yield. mp>350°3 PMR (CFBCOOH): 6 ppm 2.82 (s, 12 H), 8.0-8.6 (m, 6 H): UV (CHBCN):7\max : 442 nm. Anal. Calcd. for 9.2 C20H205608'H2504 : C, 35.39%: H, 3.27%. Found: C, 35.32%, 35.19%: H, 3.51%, 3.64%. -The bistrifluoromethanesulfonate salt 422 and bischloro- sulfonate salt 42c were similarly prepared (see Table 3). Reactipn of 48b with sodium methoxide in methgnol To a solution of 0.1 g of trifluoromethanesulfonate salt 48b (0.13 mmoles) in sufficient dry methanol was treated with a solution of 0.26 mmoles sodium methoxide in methanol. 86 The orange color of the original salt faded away immediately. After 2 hours stirring at room temperature, the solvent was removed by rotory evaporator. The white solid obtained was collected, recrystallized from methanol and dried to give 0.05 g (72%) of the dimethoxy derivative 42: mp 175-180°: IR (nujol): 1550, 1165, 1130, 1080, 730 om‘1; PMR (0001 ): 3 6 ppm 3.5 (s, 6 H), 6.4 (s, 2 H), 7.1-7.5 (m, 10 H), 7.6- 8.3 (m, 6 H). Reaction of 47awith sodium methoxide in methanol To a solution of 50 mg bisulfate salt 41; (0.086 mmoles) in 50 ml of dry methanol was added 3 equivalents of sodium methoxide in 20 m1 methanol. The yellow color of the salt solution disappeared completely in 10 minutes at room temperature with the formation of a white solid. The reaction mixture was taken to near dryness in vacuo and filtered to give a quantitative yield of the dimethoxy derivative 50: mp >350°: PMR (CD013): 6 ppm 1.95 (s, 12 H), 3.47 (s, 6 H), 7.67-8.17 (m, 6 H). Reaction oiithe dimethoxy derivative 59 with HI To a suspension of 0.45 g (1 mmole) of 59 in 10 m1 of water was added 15 ml of 48% HI aqueous solution. The 87 resulting brown suspension was further stirred for an hour at room temperature and the precipitate was filtered on a Buchner funnel. There was obtained a dark brown solid, 54. The corresponding bisulfate salt-41g could be re- generated by simply dissolving 54 in concentrated sulfuric acid. Reduction of 48b with tetrabutylammonium iodidg To a solution of 50 mg (0.064 mmoles) of trifluoro- methanesulfonate salt 483 in 100 ml of dry acetonitrile was added dropwise with stirring a solution of 2 equivalents of tetrabutylammonium iodide in acetonitrile at room temperature. The color of the solution changed from orange to green and the precipitated black microcrystalline solids. The black solid was collected with suction filtration, washed with acetonitrile and dried under vacuum to give 37 mg (80%) of the iodine complex 5i: mp 301-302°: UV (CHBCN):7\max 675 nm. Anal. Calcd. for 028Hl8S4Il.96‘ C. 45.98%: H, 2.48%: I, 34.01. Found: 0, 46.19%, H, 2.57%; I, 34.25%. Reduction with lithium was similarily carried out. The salts 48 could be recovered from the complex by dissolving 5i in the concentrated, cold acids. 1.0-L. n- v: 0035 U) .1. (I) 88 Complex of 48b with Li+TCH ' To a solution of 0.1 g (0.128 mmoles) of trifluoro- methane sulfonate salt 482 in large amount of acetonitrile (ca. 250 ml) was added 0.055 g (0.256 mmoles) of LiTTCNQ? . After stirring overnight, a brownish green solution was obtained. UV spectra indicated a strong absorption of neutral TCNQ plus TCNQ'. Analysis of the UV spectra suggested the formation of (TCNQ)?TCNQ°(TCNQ)7 mix-valence complex. Bis-1,2-dithioethane acetal of terephthaldehydg, 52 To a three-neck round bottom flask equipped with magnetic stirrer, Dean-Stark trap, reflux condenser and addition funnel was charged with 5.66 g (0.06 moles) of l,2-ethanedithiol, 60 ml benzene and a catalytic amount of p-toluenesulfonic acid. This solution was brought to refluxing and a solution of 4.02 g (0.03 moles) of tere- phthaldehyde in 15 m1 of benzene was added dr0pwise over a period of one hour. The reaction was nearly 90% complete in 30 minutes. After 1.5 hours of reflux the reaction mixture was cooled to room temperature and the solid that has deposited was collected on a Buchner funnel and washed first with water and then sparingly with ethanol. Re- crystallization from chloroform yielded 8.2 g of 57 (95.6%) as colorless crystals: mp 207-208°; IR (nujol); 89 1425, 1280, 870, 820, 730 cm-1; Mass spectrum m/e: 286 (parent), 258, 225, 197. 181, 166, 153, 121, 105, 89, 69, 60, 45. Dipotassium p-bis(cyanonitromethyl)benzene‘69 Potassium t-butoxide or potassium ethoxide (0.034 moles) was formed in situ from 1.34 g (0.034 g atom) of potassium and 10 m1 of t-butanol or ethanol. Dry THF (tetrahydrofurane) was added during this stage to help solubilize the alkoxides. To this solution at -78° was added a solution of 1.56 g (0.01 mole) of p-bis(cyanomethyl)benzene in 15 ml of THF, followed by the addition of 2.93 g (0.022 moles) of amyl nitrate. The dry-ice bath was then removed and the reaction mixture was allowed to reach room temperature and stirred overnight. The resulting brown slurry was filtered with suction filtration and was washed, successively, with 20 ml portions of THF, absolute ethanol and 30 m1 of anhydrous diethyl ether, and dried in vacuo to yield 3.0 g (93%) of the dipotassium salt 66: mp 359-361 (dec); IR (nujol): 2210, 1500, 1390, 1350, 1275, 1250, 1150, 1020, 987, 820 cm’l: PMR (020): 6 ppm 7.6 (s): 13C NMR (020): 6 ppm max.375’ 225 nm. Anal. Calcd. for ClOHuNuouKz : c, 37.25%: H, 1.25%. Found: 128.73. 125.76, 118.04, 102.30; UV (020):7\ c, 36.85%: H, 1.41%. 90 Qisilver salt of p-bis(cyanopitromethyl)benzene,.65 The disilver salt was precipitated from solution when silver nitrate was added to an aqueous solution of the dipotassium salt. The light green disilver salt was light sensitive. The yield was quantitative: mp 200- 202°: IR (nujol): 2200, 1230, 1150, 975 cm-l. Anal. Calcd. for C10H4N4O4Ag2 : C, 26.12%, H, 0.88%. Found: C, 25.02%: H, 1.01%. p:Bis(chlorogyanonitromethyl)bengene.66 Dipotassium p-bis(cyanonitromethyl)benzene (l g) was suspended in 50 ml of anhydrous ether and cooled in ice-bath to 0°. Chlorine was bubbled in slowly until the suspension dissolved and the solution turned yellow. The reaction mixture was filtered to remove small amount of brown solid and the filtrate was concentrated by means of rotory evaporator. The yellow solid which was collected by suction filtration was washed with water and dried under vacuum. The yield was 0.59 g (61%): mp)>300°: IR (nujol): 2200, 1610, 1420, 1330, 1060, 920, 825, 730 em'l; PMR (CDC13): 6 ppm 7.9 (S). 91 Dipotassium salt of 2,6-bis(cyanonitromethyl)naphthalene, 63 Potassium t-butoxide (8 mmoles) was formed in situ by refluxing 0.31 g of freshly cut potassium and 0.60 g of t-butanol in 15 m1 of dry THF until all of the metal was dissolved. To this solution at -78° was added dr0pwise a solution of 0.412 g (2 mmoles) of 2,6-bis(cyanomethy1)- naphthaleneBlin THF, followed immediately by 0.59 g (4.4 mmoles) of amyl nitrate. The reaction mixture was allowed to warm to room temperature and stirred at that temperature overnight. The precipitate which formed were then suction filtered and washed extensively with THF, absolute ethanol and anhydrous ether. A light brown water soluble solid was obtained in quantitative yield: mp)>320°. An analytically pure sample was prepared by recrystalliza- tion from hot water to give yellow shining crystals: IR (nujol): 2200, 1380, 1320, 1270, 1250, 1160, 1130, 1010, 890, 875, 800 cm"1 : PMR (DMSO-d6): 6 ppm 8.328 (s, 2 H), 7.762, 7.728, 7.672 and 7.637 (AB quartet, 4 H), determined by Bruker WM-250 NMR. 130 NMR (DMSO-dé): 6 ppm 130.29, 129.65, 126.76, 122.96, 120.13, 119.61, 97.1: UV (Dmso): ’%~max 380, 295, 285, 227 nm. Anal. Calcd. for CleéNuOuKz , C, 45.15%: H, 1.63%. Found: C, 44.86%: H, 2.04%. 92 pp:Bis(nitromethylibenzene‘62 To a boiling solution of 3 g of potassium hydroxide in 15 ml water was added, in small portions, 2 g of dipotsssium salt 69. Caution has to be taken to avoid excess foaming. Boiling was continued for at least 24 hours until the evolu- tion of ammonia ceased (negative to wet litmus paper). The hot alkaline solution was cooled and stirred with 20 g of ice. With the help of the external ice-salt bath, the solution was cooled down to -5°. Acidified with concentrated HCl gave a yellow solid precipitate which was collected and triturated with 50 m1 of water to yield 0.8 g (65.7%) of compound 62: mp 123-125°: IR (nujol): 1550, 1425, 1320, 895, 890, 680 cm-l: PMR (DMSO-d6): 6 ppm 5.67 (s, 1 H), 7.4 (s, l H): Mass spectrum m/e: 150 (parent - N02), 104, 78. Reaction of p-nitrophenyiacetonitrile with bis-dimethy - amino-t-butylmethane (Brederek's reagent) To a stirred solution of 6.5 g (0.04 moles) of p- nitrophenylacetonitrile in 30 m1 of DMF (dimethyl formamide) was added dropwise 7.0 g of bis(dimethylamino)-t-butoxymethane in 15 ml of DMF. A deep green solution was obtained immediately. After stirring overnight at room temperature, the solvent was removed under vacuum. The brown residue was collected 93 with a suction funnel and washed with ether to give 7.3 g of the corresponding enamine 66g (83%). An analytically pure sample was obtained by column chromatography (Silica gel, CH2012)' The bright yellow solid melted at 184-185°: Mass spectrum m/e: 217 (parent), 187, 171, 156, 130: IR (nujol): 2200, 1625, 1585, 1120, 855, 845, 750 em’l; PMR (DMSO-dé): 6 ppm 3.3 (s, 6 H), 7.6 (s, l H), 7.2, 7.4: 7.8, 8.0 (AB quartet, 4 H): 130 NMR (00013): 6 ppm 151.02, 144.51, 143.76, 124.18, 122.89, 119.50, 76.14, 42.91. Reaction of phepylacetonitrile with Brederek's reagent A 100 m1 round bottom flask was charged with 4.7 g of phenylacetonitrile (0.04 moles) and 7.0 g of bis(dimethyl- amino)-t-butoxymethane (0.04 moles). After stirring at room temperature for 12 hours a white solid precipitated. This was collected on a Buchner funnel and dried under vacuum to yield 6.5 g of the enamine 666 (94.5%): mp 77-78.5° (lit. mp 79-80°)86; IR (nujol): 2180, 1625, 1140, 990, 770 om'l; PMR (0001 6 ppm 3.0 (s, 6 H), 6.7 (s, 1 H), ): 3 7.2 (broad s, 5 H): Mass spectrum m/e: 172 (parent), 157, 144, 130, 117, 103. 91, 86, 77, 42: 130 NMR (00013). 6 ppm 149.85, 136.64, 128.67, 124.99. 123.88, 121.09, 76.55. 42.42. 94 Preparation of 5-aminq-4-(p-nitrophenyliisoxazole 69a In a 100 ml round bottom flask, 1 g (4.6 mmoles) of 666 and 1.3 g of‘HOSA (Hydroxyamine-O-sulfonic acid, 11.5 mmoles) were refluxed overnight with 50 m1 of iso- propanol under nitrogen. The reaction mixture was cooled and diluted with water. Rotory evaporation to partially remove the iso-propanol and to precipitate an orange solid. After washing with water, the product was dried under vacuum to give 0.9 g (95%) of 62a: mp 188-189°: IR (nujol): 3325, 1 1640, 1600, 1500, 1330, 1187, 1110, 850 em‘ : PMR (CD 00003). 3 6 ppm 6.55 (broad, 2 H), 7.5-8.1 (AB quartet, 4 H), 8.4 (s, 1 H): 130 NMR (DMSO-d6): 166.25, 150.83, 143.64, 138.72, 124.76, 124.01, 91.08: Mass spectrum m/e: 205 (parent), 189, 162, 149, 132, 116, 104, 89, 77, 63, 51. Ppeparation of_5-amino-4-phenylisoxazole 69b Enamine 666 (200 mg, 1.16 mmoles) and 330 mg of‘HOSA (2.9 mmoles) were refluxed in 50 m1 of iso-propanol over- night. The reaction mixture was taken to near dryness on a rotory evaporator. To the resulting white, thick suspen- sion was added 40 ml of water to obtain a clear solution. After standing for 10 minutes, the white crystals which precipitated were collected on Buchner funnel and dried 95 to give 60 mg of 5-amino-4-phenyl isoxazole 626 (32%): mp 144.5-146.5°: IR (nujol): 3440, 2250, 1675, 1275, 1210, 800, 760, 725, 685: PMR (CD3COCD3): 0 ppm 4.93 (S, 1 H), 6.2 (broad, 2 H), 7.3 (broad s, 5 H): Mass spectrum m/e: 161 (parent+l), 117, 90. 5-Acetamido-4—(p-nitrophenyllisoxazole 70a A solution of 0.5 g (2.4 mmoles) of 5-amino-4- (p-nitrophenyl)isoxazole 626 in 15 ml of acetic anhydride was refluxed for 24 hours. The hot solution was poured into 100 ml of water and cooled in the refrigerator for several days. The precipitate was collected on Buchner funnel and washed with water and dried under vacuum to give 0.33 g (55%) of 166: mp 213-215°(dec): PMR (DMSO-dé): 6 ppm 2.1 (s, 3 H), 7.6-8.2 (AB quartet, 4 H), 8.6 (s, 1 H): IR (nujol): 3250, 1690, 1630, 1600, 1505, 1320, 1225, 1170, 1105, 1020, 990, 920, 860, 850, 750 cm-1: Mass spectrum m/e: 247 (parent). Prgparation of TCNQ’Na? by TCNQ and sodium azigg A 50 m1 round bottom flask was charged with 600 mg (3 mmoles) of TCNQ, 195 mg of NaN3 (3 mmoles) and 15 ml of acetonitrile. This mixture was refluxed for 6 hours 96 then cooled to room temperature. The precipitate which formed was collected on Buchner funnel, washed sparingly- with acetonitrile and dried to give 0.5 g (75%) of TCNQ’Na? . IR and UV spectra of this compound were essentially identical to that of the authentic sample of TCNQ’Na?. The solvent of the blue filtrate was removed in vacuo and the residue purified on preparative TLC (Silica Gel: eluted by a 92.5% : 7.5% mixture of ethyl acetate : acetonitrile). The resulting blue solid melted o _ . . . . above 300 . UV V131ble (0H30h).?\ max.585 nm. APPENDIX absorbancc rrr v1tr TFYfi' tvrv tifv vvvv vav vvvv rrvv w 121.. ! -41.- -J a J. 1 - a L 1-! .L 1 .L .L .L .m N. ‘ ._____. ———““—-f~rw-.v—-—’q .-.H.—.—.‘_— - - .v.—--._ a..- .... ...-.- .a—hon ...—.0 .- ...... - ...-.- --— 4‘50 500 55f) wavelength minimicrons l. PMR spectrum (top) and Ultraviolet spectrum (bottom) of g6. 98 no no no no mw Re .6 4 20 . m n~ no 2 n~ o O m m S ”N nu R mm M 2 o. I 8 . no M” 7” “w .M 6 mm 51) 0 2000 61) 4.0 M'U‘UW 5.0 a‘ 115 31) 25 0'06... ....O-oc- x6605taro l . . o 2: h. 1500 2500 0 3 3500 4000 2. Infrared spectrum of 2b. 54.8 99 all! [a ‘90 4:20 [... l l 3. Mass spectrum of 2__b 100 -..... __ _ .... u 70 0.0 5,0 “LEW 4.0 10 20 :o 0 7‘? ‘7 T V V V V l V7 V V V I V V V V fiv '7 V I r V ‘r V j T V V V‘ l V V V L V V V V l V T ' 1 ' I f ' 1 ' ' f ' T ' I ' I I ' ‘ l ' ' T I l :00 «A II no mo 0». :4 "" 3 N. @631: ‘ . s I 6:» 2X? 3 2 I i I Ml I ( A AA A AV A L .— T f 1 A A A i A A A l A A A L A A l . ALA a J A A A A l L A A L ‘ A L A A A r A L A I A L A A 1 l A,‘.L A k L A L 1 AA A AA 1 A A L A I A A A A I J to ,, 70 00 so p... W 40 W; in 20 to 4. PMR spectrum of i2 (top) and 24 (bottom). 101 P no 0. o. _ n. _ . M50 800 1000 1200 8.0 M'CRONS 10.0 11012.0 71) 1400 1600 61) 1800 5.0 80 0 0 20 f . 23100 6 4 Adxozozfzzmzée 100 80 60 40 20 n~ .U n. . .0 no 5 n? .6 o. m s m c. N. my D“ K MM 0. m. 4 3 . Aw 1. 1. n~ a. 5 2. IL KN .. 2 mm .2 nu no MW mu nuA 0 0 .oxozuzftiwzga 5. Infrared spectrum of 24. 102 DMSO 5 HO 14,04N ll :>-CH&OH W r V' r I ff 7* r r I v v v Y 1' f r f V ‘ 5 I W a Honfimzy-cupu l AAAAL A A I A Ll A A l A A A A A l— L A A A l A A A A T A A A A 1 A AAAAA r A L A A I 1 A A A I A l A A IA A AA A I L A A A T A A A A L . no 70 00 so you 4 00 )0 :0 ‘ ‘ 6. 13C NMR (top) and PMR (bottom) of i} If“ 103 HOI-gC-girSSW 102.64 203 1 S N 6’ (1.0-4 IWHC/ N 5 a 88 . saw 84 168 4 2 58 ’0 ,0 :29 51 75 m m , L... .. m 13 film, 193 219 ,1) fir ''''' 'l V r ' ' r Y r V—r 'f‘ r V . "'5 50 109 158 209 ‘4 7. Mass spectra of i3 (top) and _1_4 (bottom) -—'---u«——~.—-—r.—. --.. ‘11 A A L A A A A A l A L _l L A 4 I A A A A r A A A A 1 AA A_ L A I l A A A r—AA l A_ l 1 L L A A I A A A A r A A L AT no )0 00 so m N do ' Jo f‘ld 20 no 0 l L I V V W _1' v *f 17 v I v v v ‘V 1 v v v T T v 1* f‘v I fir 77 v 1 v v v v I v v v v I r f V I f v I v f l v I v I *r :00 a no no ND 0': -o 8. PMR spectra of 48b (top) and 42a (bottom) 105 r .' V I V l v v l v v v r I V v v v r v v r v I ' V v f I I I v 7 f I h , ' ' ‘ I fi ! l . I Mn a m m kl‘ I "I. f I l ' i I l L A A A L A l A A A A A A l A A A A L l A A A L A l A A L L I A A A l L A A A r l A A A I A A 4 fL A r A I r ‘0 70 to ’0 ”A l. 40 ‘ ’0.41) DMSO N 1.0 0:11 NH; 13.3-6 14‘) 44 0¢N NH, 63 194 ‘7 51 $130.0 1 "at? 6? a 16 13C NMR spectrum (top) and Mass spectrum (bottom) of 113 A...__.__._.._____ ... .A -. “ --.... _ 4 ,1 4 ___..A_.__..—.A_.-v¢.-- C - “mm--1..— u 1;; . ; i at? 1 I Q} - 111 ' ' ‘ I - 0‘. "i". h ‘: '- I ‘ so I ! 1‘ 1,: ,1. '1 1" 1‘ y. x , ELFJJM. 1?"! r. . ' ' (3‘ It" ' ' I r ' r ' I ' I ' I ' I r r I ' ' I I r 1’ l l I * l 900 ‘00 300 no 100 .4. \ NO, \ d , COCH, ! I t‘ 2"." 17. Mass spectrum (top) and PMR spectrum (bottom) of ]_O 111+ 5.0 6.0 7.0 .- 8.0 M'CRONS 10.0 11.0 12.0 . - 16.0 ‘00 . . - . . . . . . . . . . 100 80 80 3': “J U 260 60 < . ’— '2 2 '2 (40 40 a: p— 20 20 0 t :. : ' , :' z‘ ; r .‘ : "5, 3 i . ' : O 2000 1800 . 1600 1400 1200 1000 800 2 5 -' ' 3.0 3.5 ~ 4.0 M'CRONS 5.0 6.0 ~ 8.0 ' 1 1 1L1111111-131L111111111...1....L. 100 100 80 80 SE r__o/FJ’_/r) H 360* 60 z < E 1 240 40 < . K '— 1 . 20 201 \ 1 ' L \ I 0.. 1 ”Hz 0 0 4000 4 3500 3000 2500 2000 1500 18. Infrared spectrum of 62b 115 n . v O O O n- n 6 A 7. 16.0 _ .. ...-o.... ...... M'CRONS 10.0 11012.0 10 12 .tfl I400 15 I 6 4 Aeozuzztingp 100 80 1' MICRONS 5.0 3.0 25 12500 2000 1500 0 35 19. Infrared spectrum of 62a 116 5.0 ~ " 6.0 7.0 ~ ,8.0 M'CRONS 10.0 11.0 12.0 ~ 16.0 100 ' " ‘ ' ‘ ' ‘ ; ' ' 100 80 so i”; . ... - U 260 60 ' < p— : 2 ‘2 (40 40 M .— I. n O‘N 40 U .. ' - f'.. :- "1;: : t 1 .T O 2000 1300 1600 1400 1200 1000 800 2.5 i ‘3.0 3.5 1 a 4.0 M'CRONS 5.0 6.0 ~ 3.0 . J l l l llLJlll 111111l1.llll..11111111 100 ' 100 1 80 80 g b 560 6° 5 w E 240 H 40 < . a: p... . g A} 20 ‘ 20 0 4000 3500 3000 ' 2500 2000 1500 20. Infrared spectrum of 29 CT: 1317137717.”? T1577 .1. LAJLAJ-...-¢J.L.J~ [:4 H m H 1. Polling, 1., Cbshchaya_§himiya (General Chemistry), Mir, Moscow, 1964. _ l\) Garito, A. F. and Heeger, A. J., Acc. Chem. Res., 1224, Z, 232 and the references cited therein. 3. Melby, L. R., Can. J. Chem., $965,743, 1448. 4. Gutmann, F. and Lyons, L. B., "Organic Semiconductors", Niley, New York, 1967. 5. Herbstein, F. H., in "Perspectives in Structure Chemistry", Vol.IV, J. D. Dunitz and J. A. Ibers, Eds., Niley, New York, 1971, p 166. 6. Kelby, L. R., Harder, R. J., Hertler, N. R., Mahler, 3., Benson, R. E. and Mochel, N. B., J. Am. Chem. Soc., 1263. fig. 337“- 7. Acker, D. S. and Blomstrom, D. C., U.S. Patent No. 3,162,641, 1962. 8. Acker, D. S. and Hertler, N. B., J. Am. Chem. 800., 1,292,. .6251. 3370. 9. Ferraris, J., Cowan, D. 0., Nalatka, V. Jr., and Perlstein, J. H., J. Am. Chem. Soc., $973,195, 948. 10. Wudl, F., fiobschall, D. and Hufnagel, E. J., J. Am. Chem. Soc., $2Z§1.2E1 670. ll. Coleman, L. B., Cohen, M. J., Sandmann, D. J., Yamagishi, F. C., Garito, A. F., and Heeger, A. J., Solid State Commun., 191}, 1g, 1125. 120 angler. E. N60, Chemtegh. 2.12%, _6_, 271+. 13. Narita, M. and Pittman, C. U., Synthesis, 1276, 489. l4. Keller, H. J., Ed., "Chemistry_and Physics of One- Dimensional Metal", Plenum Press, New York, 1977. 15. Torrance, J. B., Acc. Chem. Res., lgzg, 1;, 79. 16. Whangbo, M. -H., Hoffmann, R. and Woodward, R. B., Pro. R. Soc. Lond. A., 1272, 366, 23-46. 17. Khidekel, M. L. and Zhilyaeva, E. 1., Synthetic metals, L2§Ao £2 1‘3“- ll? 18. 19. 20. 21. 22. 23. 24. 25. 26. 270 28. 29. 30. 31. 32. 34. 35- 118 Perlstein, J. R., Angew. Chem. Int. Ed., ggzz. IQ. 519. Bloch, A. N., Ferraris, J. P., Cowan, D. O. and Poehler, T. 0., Solid State_Commun-, 1233, 13, 753; Schafer, D. B., Wudl, P., Thomas, G. A., Ferraris, J. P. and Cowan, D. 0., ibid, 1224,‘;4, 347. Torrance, J. B. and Silverman, B. D., Bull. Am. Phys. Soc., II, ng5, 29, 498. Denoyer, P., Comes, R., Garito, A. F. and Heeger, A. J., Phys. Rev. Lett., 1225, 35, 445. Grobman, w. D., Pollak, R. A., Eastman, D. B., Mass, E. T. Jr., and Scott, B. A., Phys. Rev. Lett., ngfi, .22. 534. LaPlaca, S. J., Corfield, P. N. R., Thomas, R. and Scott, B. A., Solid State Commun., Igzg. 12. 635. Wheland, R. C., J. Am. Chem. Soc., 1276, 98, 3926. Chaikin, P. M., Garito, A. F. and Heeger, A. J., J. Chem. Phys., 1272, 58, 2336. LeBlanc, o. R., J. Chem. Phys., I2§§..&2. 4307. Data taken from reference 18, p. 524. Wheland, R. C. and Martin, E. L., J. Org. Chem., $223..&9. 3101. Saito, G. and Ferraris, J. P., J. C. S. Chem. Comm., ggzg, 1027. Anderson, J. R. and Jorgensed, 0., J. C. S. Perkin I, $222: 3095. Sandman, D. J. and Garito, A. F., J. Org. Chem., AZZR: 22. 1165- Addison, A. R., Dalal, N. S., Hoyano, Y., Ruizinga, S., and Weiler, L., Can. J. Chem., 12Z7,.55, 4191. Maxfield, M., Cowan, D. 0., Bloch, A., Poehler, T.O., Nouveau J. Chem., 1222,.3, 647. Haley, N. B., J. C. S. Chem. Comm., lgzg, 1030; ibid, $222: 207- Wudl, F., Kaplan, M. L., Teo B. K. and Marshall J., J. Org. Chem., 12ZZ,.42, 1666. 37. 38. 39. 40. 41. 42. [4‘3- an. 45. 46. 47. 48. “'90 50. 119 . Sandman, D. J., Epstein, A. J., Holmes, T. J. and Fisher, A. P. J. C. S. Chem. Commun.,,1927, 177. Jerome, D., Mazand, A., Ribault, M. and Bechgaard. K., J. Phys. Lett. (Orsay, Fr.),‘$2§9, El. L-95: ibidd Aggg, £2, L—397. Ribault, M., Pouget, J. P., Jerome, D. and Bechgaard, K. Jgngys. Lett., and Parkin, S., Ribault, M., Jerome, D. and Bechgaard, K. J. Phys. C., submitted for publica- tion. Bechgaard, R., Carneiro, K., Rasmussen, F. B., Olsen, M., Rindorf, C., Jacobson, C. S., Pedersen, H44J. and Scott, J. C. J. Am. Chem. Soc., 128%, 103, 2 0. Johnson, J. R. and Ketcham, R. J. Am. Chem. 809., igég. Ba. 2719. Johnson, J. R., Rotenberg, D. H. and Ketcham, R. J. Am. Chem. Soc., $2Z2, 92, 4046. Gompper, R. and Topel, W. Chem. Ber., 1962, 25, 2871. Kuwayama, Y. and Kataoka, S. Yakugaku Zasshi, 1285, 85, 387-90. C.A. 63:8342. Bocion, P., Wijitha, D. S. and Winternitz, P. Ger. Offen. 2,512,564. Wudl, F., Kaplan, M. L., Teo, B. K. and Marshall, J. J. Org. Chem., ngZ, 42, 1666. Lane, C. F. Aldrichimica Acta, 1274, z, 32. Storms, P. N. and Taussig, P. R. J. Chem.mand Eng. Data, .1299 a. 272- Sandman, D. J. and Garito, A. F. J. Orgg Chem., 1274, ‘39, 1166. Dieckmann, J., Hertler, J. R. and Benson, R. E. J. Org. Chem., 1963, 28, 2719. 'Klein, L. Ph.D Thesis, Michigan State University, 1980. 51. 57. 58. 59. 61. 62. 63. 120 This behavior suggests the formation of aromatic imminium intermediate has a determinate effect for the cyclization of thiazolothiazole. Johnson has isolated the uncyclized intermediate bis-salicylidenedithiooxamide .;;%‘ "’S’ Fa £3 15;;]::;qé?CV:jgiii»’ RC) Testafenni, L., Tiecco, M., Zanirato, P. and Martelli, G. J. Org, Chem., A2Z8, 4 , 2197. 40 Spivack, J. D. , Steinberg, D. H. and Dexter, M. Belg. Patent 632, 033. Thomas, D. A. J. Hgtero. Chem., 1270, Z, 457. Shimamura, I. Shishido, T., Ohi, R. and Iwano, H. Ger. Offen. 2,350,209. C.A. 81:8367t. a) Bohme, H. and Viehe, H. Advances in Organic Chemistry: Methods and Results: "Imminium Salts in Organic Chemistry". b) Viehe, H. and Janousek, Z. Angew. Chem. Int. Ed., 32%: £9 [+051 8063 ibid- 3233-: _1-99 573- Gompper, R. and Topfl, 0. Chem. Ber.,,lgég, 25, 2861. Miles, M. C., Wilson, J. D. and Cohen, M. H. U. S. 22123:. 3.779.814 (1973). Ueno, Y., Bahry, M. and Okawara, M. Tetrahefron Lett., 1977, 4607. Fabre, J.-M., Torreilles, E. and Giral, L. Tetrahedron Lett.,,ggzg, 3703. Becke, P., Hagen, H. Cerman.Patent 1,274,121 (1968). Takamizawa, A. and Hirai, K. Qhem. Pharm. Bull.,1269, 17, 1924. The charge- transfer complex of the onium iodide was first observed spectrophotometrically by Kosower in the case of II-alkyl pyridium iodide. Kosower, E. R., Skorcz, J. A. 82_Am. Chem. Soc., lgwg, 82, 2195. 64. 65. 66. 69. 70. 71. 72. 73- 74. 75. 76. 77. 78. 79. 121 In: “onstoichiometric ratio" of the TTF-iodine complex has been reported, e.g., (TTF)2616,: Nudl, F. J. Am. Chem. Soc., A915, 92, 1961. J LeBlanc, O. R. Jr., in."Physics and Chemistry of the Organic Solid State", Vol. 3, M. M. Labes, D. Fox and Heissburger, Ed., Interscience, flew York, 1967. p.133. Ueno, Y., Nakayama, A. and Okawara, M. J. C. 3. Chem Commun., 2918, 74. Belsky, V. K. and Voet, D. Acta. Crystallogg. Sect. B, 12.2832: 272- Cooper, W. F., Kenny, H. C., Edwards, J. 8., Nadel, A., Nudl, F. and Coppens, P. J. C. S. Chem. Commun., 1221» 889- Coleman, L. B. Dissertation in.Physics, University of Pennsylvania, Pennsylvania, 1975. Ried, Von W. and Kohler, E. Liebigs Ann. Chem., 1293. 298. 145. Belsky, I., Dodiuk, H. and Shuo, Y. J. Org. Chem., ABZBw 22: 989- Nislicenus, 8., Endres, A. Chem. Be;., 3932. 35. 1755- Nislicenus, W. and Waldmueller, M. Chem. Ber., $928, 21. 3336. Feuer, R., Shepherd, J. N. and Savides, C. J. Am. Chem. Soc., 2956, 8, 4364. Nieland, H., Garbsch, P. and Chavin, J. J. Ann., 3.2%. 461, 2950 Feuer, H. and Savides, C. J. Am. Chem. 888., 1939, .81. 5826. N ” Lucas, C. R. and Hammett, L. P. J. Am. Chem. Soc., A 4 , 89, 1928 and Eastey, D. M. J. Chem. Soc., , 1193 and 1208. Nitucki, B. P., Maya, N. and Frankel, M. Org. Prep. Proced. Int., 2989, 22, 197. Kitson, R. E. and Griffith, N. 3. Anal. Chem., 1952, 21. 334. "N 81. 82. .83- 84. 85. 86. 87. 88. 89. 122 Black, A. P. and Babers, F. H. Org. Syn. Coll. Vol.11, p. 512-515 0 Furukawa, R., Fukumura, M., Akasaka, T., Yoshimura, T. and an, S. Tetrahedron Leit., A989, 761. Wallace, R. G. Aldrichimica Acta, £982.,29, 3. Biere, H. and Russe, R. Tetrahedron Lett., $979, 1361. Bredereck, H.& Simchen, G. and Griebenow, H. Chem. Be;., 5 1273. M. 15 Hagar, 7., Eiden, F. and Seemann, N. ,Arch. Pharm., B. S $ZNZ' 300, 615. Hovelli, A., DeVarela, A. P. G. and Bonafede, J. D., Tetrahedron, $988, 29, 2481. Note that most reactions involve HOSA are run in water. See reference 82. For electronic spectra of TCRQ anion radical salts see Iida, Y. Bull. Chem. Socg, Japan, A989, 82, 71. Brown, R., Ph.D Thesis, Michigan State University, 1979-