A Total Synthesis of Spiro[Bis(2,3-Diazabicyclo[2.2.1]-Hept-2-Ene-7,7'] Citation Anderson, Julie Ann (1983) A Total Synthesis of Spiro[Bis(2,3-Diazabicyclo[2.2.1]-Hept-2-Ene-7,7']. Master's thesis, California Institute of Technology. doi:10.7907/cqp3-1k14. https://resolver.caltech.edu/CaltechTHESIS:04022010-142818016 Abstract The total synthesis of spiro[bis(2,3-diazabicyclo[2.2.1]-hept-2-ene)-7,7’] from trans-1,4-dibromo-2-butene was accomplished. One advantage of this synthetic route is that it can be carried out on a relatively large scale. Another advantage is that the synthesis is flexible. With a few minor modifications, one can generate various C₉H₁₂ spiro ring system derivatives. Item Type: Thesis (Master's thesis) Subject Keywords: (Chemistry) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Dougherty, Dennis A. Thesis Committee: Unknown, Unknown Defense Date: 27 May 1983 Record Number: CaltechTHESIS:04022010-142818016 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:04022010-142818016 DOI: 10.7907/cqp3-1k14 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 5684 Collection: CaltechTHESIS Deposited By: Tony Diaz Deposited On: 19 Apr 2010 16:27 Last Modified: 22 May 2025 23:17 Thesis Files Preview PDF - Final Version See Usage Policy. 1MB Repository Staff Only: item control page

A TOTAL SYNTHESIS OF SPIRO[BIS(2,3-DIAZABICYCLO[2.2.1] HEPT-2-ENE-7,7'] Thesis by Julie Ann Anderson In Partial Fulfillment of the Requirements for the Degree of Master of Science California Institute of Technology Pasadena, California 1983 (submitted May 27, 1983) ii To My Mother and Father iii ACKNOWLEDGEMENTS I would like to thank the entire Dougherty group, especially Professor Dennish Dougherty and Lisa McElwe-White. Without their help, I could have never managed to accomplish this work. iv ABSTRACT The total synthesis of spirolbis(2,3-diazabicyclof2.2.1}- hept-2-ene)-7,7'] from trans-1,4-dibromo-2-butene was accomplished. One advantage of this synthetic route is that it can be carried out on a relatively large scale. Another advantage is that the synthesis of flexible. With a few minor modifications, one can generate various C,H 9°12 Spiro ring system derivatives. TABLE OF CONTENTS Introduction. .cceccecsveee seme eee ee tere ner e ese cesecesed Introduction to the SyntheSisS....eereees see e cece nee wee eee ed Results and DiscuSsions..ccccccsccnecerereseccens see neee .18 CONCLUSION. scacceccceserescvcsvevseseces cent naw ween aee 00 35 ReferenCeSersesesscrccsees sees cece een eae cece cee weno 38 Experimental Sectionssssecsesveceeereees seen e eee n cee eeee 39 INTRODUCTION Recently, there has been interest in the possibility of generating and characterizing 1,4,6,9-spiro [4 Ay nona- tetraryl i, an organic tetraradical. The singly occupied p orbitals of i interact via spiroconjugation?. This allows 1 to have the electronic structure of a biradical (two electrons in a degenerate pair of nonbonding MO's), although 1 is, in a structural sense, a tetraradical (two broken bonds). Scheig's emperical formula for the estimation of the spiroconjugative split in spiro 4.4] nonane derivatives predicts a 1.9 eV gap between the by and the ao molecular orbitals of 1, which has Dog symmetry .* Ab initio calcula- 3 tions are fully consistent with this result. AS 4) we e 2 oR _-- 8 Oh _ 3 The initial studies attempting to generate the tetraradical 1 were carried out by Lisa McElwee-White and Dennis A. Dougherty on the azoalkane 2, which was synthesized 4 as outlined in Scheme I. Scheme I Ne he A . ’ + A} 4X >425nm oh . ane n UNH, ae 2) KOH, 2) KON,'Pron 3) 0, The results of the sensitized photolysis of the azo- alkane 2 indicate that the best model of the reaction is as shown in Scheme II. This model of the reaction requires one of two unprecedented results: frontside fYadical attack on a C-C bond (kp) or the formation of an organic tetraradical 1 (k, or ki). These results indicate that further studies of this type of system are necessary. x x —_——»> (TET a | \ rng 4 One possible way for further investigating the results of the sensitized photolysis of the azoalkane 2 would be the radical trapping studies of the mono-radical 3. This could provide further information of the viability of the frontside radical attack on the C-C bond. $f 0 Another way to further investigate the results of the sensitized photolysis of the azoalkane 2 would be to study the decomposition of spirolbis(2,3-diazabicyclo \2.2.il\hept- 2-ene)-7,7'] ‘(bisdiazene 4 or diazene 4). Assuming 2 photons are required to generate the tetraradical 1 from the diazene 4, several processes are possible. The first photon would cause the expulsion of a No molecule to give the biradical 5. If C-C bond formation is slow, absorption of a second photon would lead to the tetraradical 1. However, if bond formation occurs, the diazene 2 would be formed, and then it would enter into Scheme II. A) hy xo N a N re The diazene 4 was prepared by Lisa McElwee-White, as \- a N \> am (or shown in Scheme TII.° Unfortunately, the low yields of the reactions in this scheme preclude a large scale synthesis of 4. Scheme IIL Ne + 1S 425am 2.N2H2 Fh ~~ —N I.KOHyiProH 40 NEN 2.CuClo t : 4 ‘ 4 N ,, 4 Preliminary results of the studies of the diazene 4 indicate that direct photolysis of 4 leads primarily to 5,5'-spirobis(bicyclo [2.1.q pentane )6,which in itself could be useful for studies of spiroconjugation. It follows from the orbital diagram of the tetraradical 1 that a dication obtained by chemical or electrochemical oxidation of 6 could be spiroaromatic. The charge would be distributed over a large region of space and over five carbons in addition to the spiro-conjugative stabilization; therefore, the dication 7could be quite a stable dication. Electro- chemical, 2-electron oxidation of picyclo [2.1.0] pentane has recently been accomplished.°© A similar experiment on 6 could lead to 7, which is expected to show interesting properties. N=N 1° direct _ fo] photolysis % ’ _ 4" 6 7 Preliminary results of the sensitized photolysis of the diazene 4 indicate that interesting reactions are occurring. A new product, different from those resulting from the sensitized photolysis of 2 has been observed.” This new product has not yet been identified. Therefore, there is a need for a synthesis of the diazene 4 that would be flexible enough that only minor modifications would be needed to generate precursors to the mono-radical 3. The research described in this thesis, therefore, was directed toward the development of a large scale synthesis of the diazene 4 which could be easily modified to produce precursors to the mono-radical 3. THE SYNTHESIS OF SPIRO [BIS(2, 3-DIAZABICYCLO (2.2.1\ HEPT-2-ENE)-7,7 5] INTRODUCTION The synthesis of spiro 4.4 nonatetraene 8 was accomplished by M.F. Semmelhack, J.S. Foos, and S. Katz. While 8 is too thermally labile to be a precursor to the bisdiazene 4, the synthetic methods employed in that study are analogous to those that were proposed to be employed in the synthesis of the diazene 4. The spiro ring system construction that was proposed in the synthesis of 4 was the same construction employed by Semmelhack et. al. in their studies. wwe spiro ring system construction Br Gr Y Br =0 CH So) aS ) a | MPBAL Do Sag Te On Om CHz r. Br Ez 10 To eliminate some of the problems that Semmelhack had with spirol4.4]nonatetraene, and to add the necessary N-N bond, it was proposed that ethyl azodicarboxylate be added in a Diels-Alder type of reaction to the diene 9 to give 10. Then one could hydrogenate 10 to give 11. \y x * x g R-N=N-R {HJ ad — . N—=r = fA cats i eas 2 10 §~COpcoH, UM COaCols R=CO2CoHs, One can envision many ways to go from the compound 11 to either the diazene 4, or a precursor to the mono-radical 3. Scheme IV shows a possible routeto precursors of 3. ° | oe NS COacaHs trans-cyclo-octane-1,2-diol.® 11 a The exchange reaction of an aldehyde for a ketone, which is driven by distilling off the more volitile one, in this case acetone, is a well known reaction. The elimination reaction is analogous to the production of trans-cyclo-octene from the benzilidene derivative of 12 From the mono-olefin 13, one can expect access to mono-radical 3 precursors via reactions at the allylic carbons. Scheme V shows a possible route to the bisdiazene 4. Scheme WZ. TosO QTos + H 7120 Test base, J “O2CaHs J (ees deals I C0acoHs ~ COacaHs 15 © ~C02CoHs R R \, . NEN _L.KOHsiPrOH L.RN=NR 2.CuClo - 2. £H7 voice N N—CO: 5 / COC aHs, id r \6 "~CO2C oH. ~co2c2H5 4 N fb a oN. R=COsCoH, or N-N= NN 22 25 13 Hydrolysis of a ketal toa diol,’ and the substitution 7 of the p-toluene sulfonate for the diol hydrogen’ are well precedented reactions. Base promoted elimination of the tosylate anion to give an olefin is also a precedented event.” The addition of ethyl azodicarboxylate or PTAD, and the subsequent hydrogenation of the product is the same type of reaction as seen in the last two steps of the over- all synthesis that produced compound 11. The hydrolysis- oxidation of that product, 22 or 25, was accomplished on the analogous di-PTAD adduct. Schemes VI-1IX show variations of the basic route of Scheme V. Scheme YC Ho H3CSCo oli sca HI tC —SeacaHs N—CO2CoHs 4 NCO. | ig "$02, oH iG N~CO2C oH 14 Regarding Scheme VI, the synthesis of the dixanthate by first making the dianion of the diol, then adding carbon disulfide, and then reacting that product with methyl iodide is a common method of synthesis of xanthates.’ The cis elimination of a xanthate to an olefin, by pyrolysis, is also a common reaction.!° The steps that would take the diene formed, 16, to the bisdiazene 4 are the same steps as outlined in Scheme V. Scheme VIC Tos Tos Br Br Libr bdse , N—~Co Nw vets feats fe Cees ec, oo Alternatively, the bis-elimination of a 1,2-dibromide has precedence in literature, and could be applied to the present system as shown in Scheme VII. The steps that would take the diene formed, 16, to the bisdiazene 4 are the same steps as outlined in Scheme V. bdse, VY) Br 15 Scheme WIT Tos Tos PhSe $SePh PhseNa MCPBA N—~Co C ri os hi ~C02CaHs 15 COaCaH, "~ COpcoH. 0 0 Hon PhSe SePh A ———n N— M1 CO Cos }~C0eC os Ono. 16 NS COzc oH. Precedence for the syn-elimination of selenoxides (Scheme VIII) can be found in the work of Hans J. Reich, /¢ K.B. Sharpless, 2? and others. The steps that would take the diene formed by such an elimination of the di-selenoxide to the bisdiazene 4 are, again, the same steps as outlined in Scheme V. “A cHs y A + OH) = re€i2 se —{,,, a! ee Mex CArSe Mes 16 Scheme IX TsO OTs I r Nat _| NBS BS N N—~Co, Ma CCH. eas a coscat IS NS COz¢He ~ S26 2Hs | 2C2Hs Br =h0k thr | C2Cots fe Cas 20 "~COscoH. IS ~ C02 aH The reaction of Scheme IX, in which a di-tosylate is converted to an olefin via a di-iodide has a great deal of precedence. From the mono-olefin 13 one could go two ways. 13 is an entry into Scheme IV, which is a route to the precursors of the mono-radical 3. One could also allylically brominate 13, and then do the base promoted elimination reaction to the diene 16, which can be taken to the bisdiazene 4 as in Scheme V. 17 Semmelhack et. al. did a very similar series of reactions 7 with encouraging results,‘ although his final product, a non- hydrogenated analogue of the diene 16, decomposed over a period of a few hours at room temperature. Semmelback : Toso 0S Nal | , R—-N=N—-R: NBS em 1 —|" N—R N—R R =CO2CHz —Br to ether N—R. N— i tor /. na " N~p N~p 18 RESULTS AND DISCUSSION The starting point of the synthesis of the bisdiazene 4 is the readily available spiro-dioxolane 9, first synthesized by Semmelhack et. al. (Equation 1). Equation | or CH ( C) 3 ee 0 “snci4 fe oe Gt pe Ox Diethyl azodicarboxylate was added in a Diels-Alder reaction to the spiro-dioxolane 9 by reaction at 55° for 47 hours in carbon tetrachloride in 82% (Equation 2). The product, diene 10, was characterized by 136 NMR and PMR. The PMR showed a multiplet at 6.50 corresponding to the two vinyl protons. Initially, catalytic hydrogenation over Pd/C with 1 atm. of Hy was attempted. This gave only starting material. This may be due to the steric hindrance resulting from the interaction of the C=C bond to be hydrogenated and the CH, hydrogens on the other ring. Then the hydrogenation of the of the olefin 10 was accomplished with a diimide reduction 19 using NjH)*H,0 in ethanol (Equation 2). The mixture was heated at 52° for 24 hours in 83% yield. The product 11 was characterized by 13¢ NMR and PMR. The disappearance of vinylic protons in the PMR was noted as evidence that the reaction was successful. Equation 2 0 R—-N=N—R N2H2 -82.% 83% a | N= _ i] N C02C oH. N. COColts 9 lo "~COscaH, MNS COecoH, R=CO2CoHs, Initially, the synthesis of the mono-olefin 13 was attempted through the reaction of n-butyl lithium with the acetal 12, which was produced from 11 by an exchange reaction with benzaldehyde(Equation 3). The exchange reaction took place by heating 11 with benzaldehyde in carbon tetra- chloride for 5 hours at 62° in a yield of 86%. When the acetal 12 was heated for 5 hours at 51° with one equivalent of n-butyl lithium, only starting material was present. When a ten fold excess of n-butyl lithium was added to the 20 solution, the desired olefin was still not produced. Instead of removing the hydrogen on the benzalhyde acetal, the n-butyl lithium apparenly attacked the esters. The resulting product may have been the hydrazine, which would be produced upon work up. The evidence for this can be seen in the PMR. The protons on the esters are missing, the protons on the benzaldehyde acetal are still present, and there is no sign of vinylic protons. Equation3 21 The next methods for the attempted preparation of the bisdiazene 4 involved elimination reactions of the spiro- di-tosylate 15, which was prepared from the diol 14. The diol 14 was prepared by the hydrolysis of the ketal 11 by heating the ketal at 92° for 45 minutes in 50% water/50% methanol with a catalytic amount of HCl (Equation 4). The yield of this reaction was 82%. The diol 14 was characterized by PMR. The spiro-di-tosylete 15 was prepared by allowing the crude diol 14, and p-toluene sulfonyl chloride to stand in pyridine at 4° for 3 days (Equation 4). The yield was 75%. The product was characterized by PMR. Equation 4+ 7 H Tos Tos HE/H20, TosCl_ pyr eter: COsc N—CO2CoHs NCOs aH. . 60 2Hs5 Ll N~o 2COHs i+ N~ CO2CaHs IS. 2C2Hs, Initially the elimination of the spiro-di-tosylate 15 to give the diene 16 was attempted by reacting 15 with potassium tert-butoxide in dimethyl sulfoxide at 25° for 2 hours (Equation 5). This base was chosen because of its 22 steric bulk. No reaction was observed. Upon work up with water,the diol 14 was regenerated. One possible explaina- tion of the lack of elimination is the steric hindrance of the exo-hydrogens and the bulky tosylates of the spiro- di-tosylate 15. Equation 5 Toso Tos He. +H —hoK N—~Co, N J. CoHs }' C026 os 5 2CaH. 16 ®~COsC aH. In view of this conclusion, the elimination reaction of the spiro-di-tosylate 15 was attempted using sodium ethoxide (Equation 6). This base is smaller, so that some of the steric problems may be overcome. The spiro-di-tosylate was heated for 23 hours at 73° in ethanol with sodium ethoxide. The product of this reaction resulted from the elimination of one side of the di-tosylate 15, and substitu- tion of an ethoxide group on the other side, 17. The product, 17, was identified by PMR and mass spectrometry. A distinction between 17a and 17b was not possible, but 23 Since the exo hydrogens hinder the production of 17b, it seems likely that 17a is the predominant product. Equation 6 Tos QTes CoHs 2s C2Hs5ONa and/ or N N 5 icogeat a Nome & Ncrcse Attempts were made to carry out the elimination reaction of the spiro-di-tosylate 15 with more reactive, less nucleophilic, albeit, sterically bulky amide bases. The reaction of 15 with lithium diisopropyl amide (LDA) in tetrahydrofuran at room temperature for 18 hours gave only starting material back. LDA is too bulky a base for any elimination to occur. The elimination was also attempted with 1,5-diazabicyclo 5.4.0 -undec-5-ene (DBU), which was heated with 15 at 75° in benzene for 18 hours. This gave some of what appears to be the product of the elimination of the hydrogen on a carbon attached to the tosylate (Equation 7). The product was identified by PMR with olefin peaks at 4.60-5.45. The partial loss of tosylate protons was also noted. 24 Equation 7 Tos’ Tos ° Toso Tos BU and/or N~CO0C N N Nn "2Cels 1 © 2Cotts p “2CaHs Since the elimination reactions of the spiro-di-tosylate 15 with various bases did not give the desired product, the diene 16, another approach had to be tried. One approach that was tried was the pyrolysis of the dizanthate 18. The dixanthate was prepared from the diol 14 by initially making the dianion of the diol by the reaction with NaH, then allowing that to react with CS, for 2 hours at room tempera- ture, and then allowing the product of that reaction to react with CH51 for $ hour at room temperature (Equation 8). The dixanthate was characterized by PMR. The yield was 75%. The dixanthate pyrolysis was initially tried by heating the dixanthate to 130° for 4 hours under argon, and then increasing the temperature to 190° for 2 hours. The product of this reaction appeared to be some of the product of the elimination of just one of the xanthates and some starting material (Equation 8). Then the reaction was tried by 25 heating the dixanthate to 200° ina Kugelrohr under vacuum. What appeared to be the product of the elimination of one of the xanthates distilled over (Equation 8). This product was then heated under argon at 190° for one hour, but no further reaction occurred. The product was characterized by PMR. There was an incomplete loss of the xanthate protons, and the appearance of a multiplet due to vinylic protons at 5.35-6.20. Equation8 S S ; : Ys CSCO Ho H3CSCo — OCSCH3 H3CSC| 1. NaH , A 2.CS2_, N—COsCoH N—CO2CaHs hae 25 N~ S 12 N— CO2CoHe N—~ CO2CoHs 14 CO2CaHs i} — Gnd the other isomer The next reaction attempted was the formation of the dibromide 19, and its subsequent elimination with base. The spiro-di-tosylate 15 was heated with LiBr in acetone for 48 hours in an attempt to make the dibromide. Apparently, once the bromide 19 was formed, it eliminated bromine to give the mono-olefin 13.(Equation 9). This reaction was characterized by PMR. Therefore, another method of generating the diene 16 was needed. 26 Equation 9 TosO Tos ey rs AiBr_ —+ Keoscaig | LON Peecgig| LAC BPeecat 15 “Cosco, | "Cost. | |B “PRC The next reaction scheme that was conceived was the elimination of a diselenoxide formed from the spiro-di- tosylate 15. 15 was allowed to react with sodium phenyl- selenide in ethanol at room temperature for 2 hours. The product was characerized by PMR. There was a new peak at 2.22-2.42 perhaps due to CH-SePh. The yield was very low, and the diselenide was never obtained pure. At that time, a literature search revealed that allylic selenoxides normally undergo a 2,3 sigmatropic rearrangement very rapidly at low temperature to yield an allylic alcohol after solvolysis of the intermediate selenate ester (Equation 10); 14 therefore, since initiallly an allylic selenoxide would be formed in the desired transformation, this route was abandonned, without attempting any oxidations of the selenide. 27 Equation 10 Toso Tes Page $ePh Phoe SePh PhSeH MCPBA Aa N N q PhSe OH > N—C. _— CO2C He . ~COecoH. The next procedure that was attempted, and that ultimately succeeded, was to make the mono-olefin 13 from the spiro-di-tosylate 15 (Equation 11). A solution of the spiro-di-tosylate and an excess of sodium iodide in acetone was heated for 4 days at reflux. Initially the di-iodide was formed, which then eliminated I, to give the mono-olefin. 13 in 82% yield. This product was characterized by 136 NMR , PMR , and mass spectrometry. There were 10 peaks in the 13¢ NMR plane of symmetry of the molecule. A broad singlet was evident in the PMR at 5.60 due to the vinylic protons. 28 Equation It TsO Ofas | NaI _, NBS Ceca, (pccals N CoO (5 CO eC oH. 13. 2C2Hs B _ _ and/or Br =O x COzC aH | C02CoH. de OCs 20a i, oH ~20b “~COz¢>4. 6 Cosa, The mono-olefin 13 was then heated to reflux in carbon tetrachloride with N-bromosuccinimide and dibenzoyl peroxide (as initiator) for 2 hours to give the allylic bromide in 87% yield. This was characerized by PMR. A distinction between isomers of the allylic bromide was not possible. 20b is probably the isomer that is predominatly fromed, due to the steric hindrance of the exo hydrogens during the formation of 20a. The allylic bromide 20 is somewhat sensitive, and it turns brown upon sitting at room temperature in oxygen; therefore, the next reaction must be carried out as soon as possible. 29 The allylic bromide 20 is then transformed into the diene 16 by reacting with 100% excess of potassium tert- butoxide in ether for 4 hours at room temperature in 69% yield (Equation 11). The diene was characerized by PMR. It had the characeristic conjugated diene pattern from 5.95-6.50 (a pattern very similar to that of the diene 9). The diene 16 decomposes at room temperature over a period of about a day, so that any reactions with it should be carried out as soon as possible. A control experiment was run with the spiro-di-tosylate 15 under conditions similar to those that made the transformation of the allylic bromide 20 to the diene 16 work. The spiro-di-tosylate was stirred at room temperature with an excess of potassium tert-butoxide in ether for 8 hours. The product of this reaction was not identified, but it was clearly not the diene 16 due to lack of vinylic protons in the PMR. Ethyl azodicarboxylate was added in a Diels-Alder reaction to the spiro-diene 16 by heating a solution of the two reactants in carbon tetrachloride at 65° for 64 hours to produce the adduct 21 in 60% yield (Equation 12A). It seens likely that the temperature of this reaction must be greater than 64° for the reaction to proceed at a rate 30 Equation 12. iL Lt RN=NR N2H2 N~COpC oH, ' —C026dHs5, } —CO262H5 IG "~COsC=H. N~cozc2Hs5 ~c0202H5 A.- R=CO2CaHs 2 22 No B.N-N= N— 24. a5 Significantly faster than the rate at which the diene decomposes. This assumption is based on PMRs taken during the course of the reaction. The olefin peaks became broader when the temperature was below 65°, When the temperature was 65° or higher, the vinylic peak characteristic of 21 grew in. the product was characterized by PMR, with two vinylic protons at 6.57 with a similar appearance to that of compound i0. The adduct 21 was hydrogenated by the diimide method in 85% yield (Equation 12A). Hydrazine hydrate and 21 were heated to 61° for 19 hours in ethanol with On bubbled into the solution. The product was characterized by 136 NMR , PMR, and mass spectrometry. The 136 NMR had only a few peaks due to the Cy axis of the molecule. The disappearance of 31 of vinylic protons in the PMR was noted as evidence that the reaction had proceeded to completion. A variation on the last two reactions was also tried. Instead of adding ethy? azodicarboxylate to the diene 16, PTAD was added. The addition was done at room temperature over a period of a few minutes.(Equation 12B). The yield of this reaction was 96% versus 60% for the azodicarboxylate addition. The product, 24, was characerized by PMR. The PMR showed 2 vinylic protons at 6.54. The PTAD adduct 24 was then hydrogenated by the diimide reduction method.(E (Equation 12B). Hydrazine hydrate and 24 were heated to 65° for 28 hours with 0, bubbled into the solution. The 2 yield of the product 25 was 88%. The product was charaterized by PMR and 136 NMR. The PMR showed the loss of the vinylic protons. The final reactions to make the bisdiazene 4 from the hydrogenated adducts 22 and 25 were carried out in three steps (Equation 13). Initially the esters and/or the urazol were removed by reaction with KOH in refluxing isopropanol for 3 hours to produce the hydrazine 23. Then the compound 23 was oxidized by adding CuCl, to make a copper adduct, followed by the reaction of that adduct with NH), 0H to generate the bisdiazene 4. The final product was sublimed to give white crystals. The yield of the reaction 32 22 to 4 was 47%. The yield of the reaction 25 to 4 was 18%. The reason for the low yield of 25 to 4 was probably that the hydrazine 23 was oxidized to the diazene 4 sometime during the reaction of 25 to 23. Under these conditions, 4 would decompose once it was formed. This problem could be avoiding by very carefully excluding 0» when one is dealing with the hydrazine 23. The product was identified by comparison of spectral data to that of the diazene 4 that was produced by the method of Lisa McElwee-White. Equation {3 LI 7 As N—N Ne KOH/i PrOH Cuca N—Co2CaHs a N~cozceHs A R=CO5C O2CaHs, 22, 23 4 Pe ER —-N= _- 25 gy 23 a The overall yield of the synthesis from spiro-dioxolane 9 to the bisdiazene 4 was 5.1% via compound 22, and 3.1% via compound 25. The reactions are all fairly simple, which is one of the reasons that these routes to 4 are desirable. 33 TOTAL REACTION SCHEME Br re Br MOPBA, _)=0 oo Hs © 56% CHg 32.% r. Br NBS 87% , N—-CO2C Hs 13, N™COacaHs, Br Gnd/or to 63 Create 1 “CRC ats: 1 Ce eCatts 34 RN=aNR NoH = Coca N—cozlaHs x—coztats COC HH. *~coacals N~cozczHs A. R=CO2CoHs 2 2! eaz = 22 ih een KOH/iProH .cuct oe 2.NHaGH w—cozeats nu ; N Nt N —~Cozc2Hs 3Step Yield A ReCOplally 22 Bo kk Noe “ay i -n= - Bb 23 18% ro 36 Equdtion 14 LH] 2.KOH7iProH 4.NH40H ; 5. hy N feats fees 3 COacaH. ™ CO2C Hs The direct photolysis of the bisdiazene 4 produces mainly 5,5'-spirobis(bicyclo 2.1.0 pentane) 6, which may have interesting electrochemical properties (Equation 15). One might be able to generate the dication 7, which is analogous to the tetraradical 1, but would have a closed shell electronic structure. This could be accomplished by electrochemical reactions of 6. Another possible way to generate the dication 7 would be the reaction of the dibromide 26 with magic acid (SbF,) (Equation 15). The dibromide 26 could perhaps be produced by allowing the allylic bromination of the mono- Olefin 13 to proceed on both allylic positions. Then one could hydrogenate that product, hydrolyze the esters, and oxidize to give a dibromo-diazene. The direct photolysis of that product could yield the dibromide 26. 35 CONCLUSION The results of these experiments are not only useful from the standpoint that the reaction scheme produces the bisdiazene 4 in relatively large quantities, but also from the standpoint of the versatility of the route. The best method of synthesis of the bisdiazene 4 would be to add PTAD instead of diethyl azodicarboxylate in both of the Diels-Alder reaction steps. This is due to the high yield of the hydrolysis/oxidation of the di-PTAD adduct, which was initially produced in Lisa McElwee-White's synthesis of the bisdiazene 4 (Scheme III). That reaction goes in a yield of about 90%.° The yields of the other reactions involved would probably be relatively unchanged by the substitution of PTAD for diethyl azodicarboxylate. One could generate the precursors to the mono-radical 3 for studies of the frontside radical attack on a C-C bond (Equation 14) with only minor modifications of the synthesis of the bisdiazene 4. This could be accomplished via reactions at the allylic positions of the mono-olefin 13. Then one could hydrogenate, do the hydrolysis/ oxidation reaction, and then photolyze to yield the desired compound (Equation 14). 37 Equation 15 N=N P24 direct. (07 photolysis x / 4 N = — SbF, 1 KOH/iPYOH 1, 2NBS Br. Br 2Cculloa Br. . 2,fHJ 3NH40H N h C02C ats i —CO2CoHs 3S CO2cans "“COacaH, 26 9. 10. 11. 12. 13. 14, 38 REFERENCES Simmons, H.E.; Fukanaga, T.-d. Am. Chem. Soc., 1967, 89, 5208-15. Schweig, A.; Weidner, U.; Hill, R.K.; Cullison, D.A. J. Am. Chem. Soc., 1973, 95, 5426-7. McElwee-White, L; Goddard, W.A.; Dougherty, D.A. manuscript in preparation. McElwee-White, L.; Doughery, D.A. J. Am. Chem. Soc., 1982, 104, 4722-2), McElwee-White, L.; Dougherty, D.A., unpublished work. Gassman, P.G.; Yamaguchi, R. J. Am. Chem. Soc., 1979, 101, 1308-10. » Semmelhack, M.F.; Foos, J.S.; Katz, S. J. Am. Chen. SOC., Hines, J.N.; Peagram, M.; Whitham, G.H.; Wright, M. Chem. Comm., 1968, 1593-94. Snyder, C.H.; Soto, A.R. J. Org. Chem., 1967, 29, 742. Carey, F.A.; Sundberg, R.J. Advanced Organic Chemistry: Part B (New York: Plenum Press,1977), p. 247. Schaefer, J.P.; Endress, L. Org. Syn., 47, 31. Reich, H.J.; Shah, S.K. J. Org. Chem., 1977, 47, 1973-76. Sharpless, K.B.; Young, W.M.; Lauer, R.P. Tet. Lett. 1973, 1979. Salmond, W.G.; Barta, M.A.; Cain, A.M.; Sobala, M.C. Tet. Lett., 1977, 20, 1683-86. 39 EXPERIMENTAL SECTION General Experimental Procedures. All solvents used were ACS grade except for ethanol (absolute ethanol was used). All were not further purified except for tetrahydrofuran (freshly distilled from lithium aluminum hydride), and acetone (dried over 4 A molecular sieves). Boiling points are uncorrected. PMR spectra were obtained using a Varian EM-390 spectrometer. 136 NMR data was obtained using a JEOL FX-90Q spectrometer. Mass spectral data was obtained through the aid of the Caltech Analytical Facility. All NMR chemical shifts are expressed in parts per million downfield from internal tetramethylsilane. Preparation of trans-2,3-Di(bromomethyl)oxirane. Synthesized in 59% yield from trans-1,4-Dibromo-2-butene according to the method of reference 7. Preparation of cis-4, 5-Di(bromomethyl )-2,2-dimethyl-1,3-dioxolane: Synthesized in 56% yield from trans-2,3-Di(bromomethyl)oxirane according to the method of reference 7. Preparation of Spirodioxolane 9. Synthesized according to the method of reference 7 with Slight modifications. Ina 1-liter three-neck flask fitted with two addition funnels (capped with serum caps) and a reflux condensor (capped with 40 a stopcock) was placed a suspension of sodium hydride (19.02¢g of a 60% suspension in mineral oil, 476mmol, Alfa) in 300m1 of tetrahydrofuran. In the addition funnel, a solution of cis-1,4-di(bromomethyl )-2,2-dimethyl-1,3-dioxolane (52.89¢, 184mmol, containing approximately 29% of the trans isomer) in 60 ml of tetrahydrofuran was placed. In the other addition funnel, 60ml of tetrahydrofuran was placed. After placing the sytem under argon, cyclopentadiene (16.94m1, 257 mmol, freshly prepared by cracking dicyclopentadiene) was injected into the addition funnel containing only tetrahydrofuran. Then the solutions were added to the sodium hydride suspension over 35 minutes at 25°, Spontaneous heating occurred. During this time, the color of the solution went from beige to purple to dark blue to black. After the addition was completed, another 1.75 ml of cyclopentadiene (72mmol) was added all at once. The mixture was allowed to stir at room temperature for 2 hours. Then water and ether were added, and the organic layer was washed with water, dried with magnesium sulfate, and concentrated by rotary evaporation. The product was purified by distillation on a Kugelrohr at a pressure of .05 torr and a temperature of 60°-110°. The yield was 11.53g (60mmol, 32%) of a clear green liquid that was slightly impure. The product was identified by comparison to that prepared by Semmelhack et. al.. 44 Preparation of the Diels-Alder Adduct 10 of Diethyl Azodi- carboxylate to Spirodioxolane 9. A solution of the spirodioxolane 9 (11.53g, 60mmol) and diethyl azodicarboxylate (11.59¢, 67mmol, Aldrich Chemical Co.) in 150 ml of carbon tetrachloride was placed in a 250 ml round bottom flask fitted with a reflux condenser. This was heated to 55° for 47 hours under nitrogen. The solution, now much less yellow than it was initially, was filtered through basic alumina with dichloromethane as eluant, and then concentrated to give 18.08g (49mmol, 82% yield) of a very viscous, very light green liquid. This was characterized by PMR and 136 NMR. PMR (CDC1.)s 1.27 (triplet with a high center, 9H, i ketal CH, and 2 ester CH 3 3 1.73 (multiplet, 4H, 2 CH 's), 1.52 (singlet, 3H, 1 ketal CH), 5's near ketal), 4.17 (quartet, 5H, 2 CH,'s on esters and i bridgehead H), 4.53 (multiplet, 2H, 2 CH-0), 5.07 (broad singlet, 1H, 1 bridgehead H), 6.50 (multiplet, 2H, vinylic H's). 2c nur (CDC1,): 14.29 (2 CH,'s on esters), 23.39 (CH on ketal), 26.12 (CH. on ketal), 56.12 (CH, near ketal), 36.90 (cH, near ketal), 62.42 (CH, on ester), 62.25 (CH, on ester), 67.45 (bridgehead C), 68.81 (bridgehead C), 71.35 (spiro-C), 79.27 (CH near ketal), 80.05 (CH near ketal), 109.23 (C19), 134.77 (vinylic ©), 137.43 (vinylic C), 158.10 (-C-OEt), 158.69 (-C-OEt). 42 Attempted Catalytic Hydrogenation of the Diels-Alder Adduct 10. The adduct 10 (1.21g, 3.31x 1072mol) was placed in ethyl acetate with a catalytic amount of palladium on carbon. The mixture was subject to hydrogenation in a one-atmosphere hydrogenation apparatus. Only starting material was recovered as verified by PMR. Diimide Reduction of the Diels-Alder Adduct 10. A solution of the adduct (18.08g, 49mmol) and hydrazine hydrate (25ml, 515 mmol, Matheson, Coleman and Bell Chemical Co.) in 55 ml of ethanol was placed in a 100ml round bottom flask equipped with a reflux condenser. The solution was stirred at 52° for 24 hours while On was bubbled into it. Then the ethanol was removed by rotary evaporation. The residue was taken up in diethyl ether and washed three times with water until the pH was neutral. The organic layer was dried with magnesium sulfate, and concentrated. The yield was 15.08g (41mmol, 83%) of a white solid. The product was characterized by PMR and 13¢ NMR. PMR (CDC1,): 1.27 (triplet with a high center, 9H, 2 CH,'s of esters and 1 CH. of ketal), 1.47 (singlet, 3H, i CH, of ketal), 1.60-2.00 (multiplet, 8H, CH,'s not on 3 esters), 3.93 (broad singlet, 1H, bridgehead H), 4.15 (quartet, 2H, 1 CH, on ester), 4.17 (quartet, 2H, 1 CH, on ester), 2 4.50 (broad singlet, 1H, bridgehead H), 4.60 (multiplet, 2H, 2 CcH-0). 1c nmr (CDC1,): 14.49 (CH,'s on esters), 43 23.53 (CH, on ketal), 26.29 (CH, on ketal), 27.23 (2 peaks- both CH,'s resulting from hydrogenation), 35.03 (CH, near ketal), 35.68 (CH, near ketal), 57.77 (bridgehead C), 58.10 (bridgehead C), 62.13 (CH, on ester), 62.26 (CH, on 2 ester), 64.53 and 67.58 (one is probably due to spiro-C), 79.87 (CH-0), 80.06 (CH-0), 109.44 (pc), 157.40 (broad, both C-OEt). Benzaldehyde Exchange Reaction with Hydrogenation Product 11. A solution of the hydrogenation product 11 (.36g, 9.7x107 mol), 6 benzaldehyde (.108 ml, 1.06x107?mol), and 7x10 “1 of concen- trated HCl in 7 ml of carbon tetrachloride was placed in a round bottom flask.. The reaction mixture was stirred at 62° for 5 hours allowing vapors to escape slowly. Then water was added. The organic layer was separated off, washed with water until the pH was neutral, and then concentrated. The yield was .35g (8.49x1074 mol, 86%) of a white solid 12. The product was characterized by PMR. PMR (CDC1,)s 1.1 (triplet, 6H, CH,'s of esters), 1.5 (multiplet, 8H, CH,'s not on esters), 3.7-4.1 (multiplet, 5H, both CH,'s of esters and 1 bridgehead H), 4.1-4.8 (multiplet, 3H, 2 CH-O and 1 bridgehead H), 7.0-7.5 (multiplet, 4H, phenyl H's), 7.5-8.0 (multiplet, 1H, phenyl H). AY Attempted Elimination of Benzaldehyde Product 12. A solution 6 of 12 (.35g, 8.4x107 mol) and 525x10° 4 liters of n-butyl lithium (1.6M in hexane, 8.4x10 “mol) in benzene was put in a two-neck flask under nitrogen. The flask was equipped with a reflux condenser. The solution was stirred at 51° for 5 hours. Then water was added, and the organic layer was separated off and concentrated. The PMR indicated only starting material was present. Then the .17g (4.1x107 01) of starting material left was put in benzene. 2.5ml (4x107?mol) of n-butyl lithium was added. After 10 seconds the solution turned orange, and then it turned yellow. The reaction was quenched with water immediately. The PMR indicated that the esters were not present, but the benzaldehyde part was still there. PMR (CCl) ): 1.32 (broad singlet, t-butanol), 3.7-4.1 (broad, about 1H, bridgehead H), 4.3-4.9 (multiplet, 3H, CH-O and a bridgehead), 7.0-7.5 (multiplet, 5H, phenyl H's). Preparation of the Diol 14 from the Hydrogenated Adduct 11. A solution of the Diels-Alder hydrogenated adduct 11 (12.252, 33mmol), 91 ml of water, 91 ml of methanol, and 1 ml of concentrated HCl was heated at 92° for 45 minutes in an open flask from which vapors were allowed to escape slowly. The mixture was allowed to cool. It was parti- tioned between dichloromethane and water. The organic layer was dried and concentrated to give 8.92g (27mmol, 82%) of the crude diol. This was not further purified. 45 The product was characterized by PMR (CDC1,): 1.26 (triplet, 6H, both CH, 's of esters), 1.46-1.93 (multiplet, 8H, CH,'s other than on esters), 2.37-2.83 (multiplet, 2H, OH's), 3.90-4.37 (multiplet, 8H, 2 CH,'s of the esters, 2 bridge- head H's, and 2 CH-O's). Preparation of the Spiro-di-tosylate 15 from the Diol 14. The crude diol 14 (8.92g, 27mmol) and p-toluene sulfonyl chloride (21.40g, 107 mmol, Aldrich) were dissolved in 70 ml of pyridine, and allowed to stand in the refrigerator at 4° for 3 days. A white solid precipitated out with time (the pyridine-HCl salt). The mixture was poured into ice-water and extracted 3 times with dichloromethane. This was then concentrated. The product was an orange viscous liquid. This was taken up in ether, and decanted, leaving an orange residue. The residue was taken up in dichloromethane, and filtered through Florisil. Both the ether extract and the product of the filtration were concentrated, yielding 12.62g of a white solid (20mmol, 75%). The spiro-di-tosylate was characterized by PMR (CDC1,): 1.18 (triplet, 3H, CH, on an ester), 1.22 (triplet, 3H, CH, on an ester), 1.43-2.23 (multiplet, 8H, CH,'s not on esters), 2.40 (singlet, 6H, CH,'s on the tosylates), 46 3.87-4.40 (multiplet, 6H, CH,'s on the esters, and the bridgehead H's), 4.53-4.90 (multiplet, 2H, CH-OTos), 7.13-7.40 (doublet, 4H, tosylate H's), 7.40-7.54 (doublet, 4H, tosylate H's). Attempted Elimination of the Spiro-di-tosylate 15 with Potassium tert-Butoxide in Dimethyl Sulfoxide. A solution of the spiroditosylate 15 (.10g, 1.6x107 mol) in .2 ml of deuterated dimethyl sulfoxide was added via syringe to an NMR tube containing potassium tert-butoxide (04g, 3.6x107% mol, MCB Chemical Co.) in .2 ml deuterated dimethyl sulfoxide. The reaction proceed for 2 hours at room temperature. The reaction was watched by. ‘PMR. No reaction was evident by PMR, although the solution changed from colorless to black. After 2 hours the mixture was poured into water, and extracted with hexane. The hexane extract contained the diol 14, as verified by PMR. The Elimination Reaction of Spiro-di-tosylate 15 with Sodium Ethoxide. Sodium ethoxide was made by adding metalic sodium to ethanol (enough sodium was added fora large excess of sodium ethoxide for the next part of the reaction) in a two-neck flask equipped with a reflux 4? condenser and an addition funnel. The entire set up was under argon. The addition funnel contained the spiro-di- tosylate 15 (.10g, 1.6x107¢ mol) in .50 ml of ethanol. This solution was then added to the sodium ethoxide solution. The reaction was stirred at 73° for 23 hours. The solution became orange, and a solid precipitated out. Then most of the ethanol was removed by rotary evaporation. 2% NH) C1 solution was added. This was then extracted with dichloro- methane, and then with ether. The organic layers were com- bined, dried with magnesium sulfate, and concentrated. The crude product was run through basic alumina with carbon tetrachloride, and then with dichloromethane. The product was in the dichloromethane layer. There was .02g (6.1x107°mol, 38% yield) of a yellow liquid, that was identified as com- ound 17. The product was characterized by PMR and mass spectrometry. PMR (CDC1,): 1.17 (triplet ,3H, CH. of ether), 1.33 (triplet, 6H, CH, of esters), 1.87 (multiplet, 6H, CH,'s not on esters or ether), 3.40 (quartet, 2H, ether CH,), 3.60-4.50 (multiplet, 7H, CH,'s on esters, bridgehead H's, and CH-OEt), 5.63 (doublet, 1H vinylic H), 5.98 (doublet, 1H, vinylic H). Mass spectral mol. wt. 338.1836 (calc. 338.1840). 48 Attempted Elimination of the Spiro-di-tosylate 15 with Lithium Diisopropyl Amide. A solution of diisopropyl amine *m0l, MCB Chemical Co.) in .5 ml of tetra- (.02ml, 2.0x10 hydrofuran was placed in a 10 ml two-neck flask (capped with a serum cap) equipped with an addition funnel containing the ditosylate 15 (.1g, 1.6x107 mol) in .5 ml of tetrahydro- furan, and a condenser with an argon inlet. The diisopropyl amine solution was kept at -78°, N-butyl lithium (1.6:M in hexane, .O1ml, 1.6x107 mol) was added to the diisopropyl amine. After 10 minutes the spiroditosylate 15 was added, and the reaction was stirred for 20 minutes at -78°, Then 4 more diisopropyl amide(1.6x10° “mol) was added. The temperature was raised to room temperature. There was no visible change after stirring under argon for 18 hours. The reaction was heated to 35° for 10 minutes. Then ice was added, and the solution extracted with ether. The organic layer was concen- trated. The PMR indicated that only starting material was present. Attempted Elimination Reaction of the Spiro-di-tosylate 15 with 1,5-Diazabicyclo 5.4.0 undec-5-ene. A solution of the spiro-di-tosylate 15 (.10g, 1.6x107 mol) and 1,5-diazabicyclo- 5.4.0 undec-5-ene (.05ml, 4.0x107 "mol, Aldrich) in 1 ml of benzene was placed in a 10 ml two-neck flask equipped with 49 reflux condenser. The reaction was stirred at 75° for 18 hours under nitrogen. Then ice was added. The solution was extracted with ether. The ether layer was washed with water until neutral, dried with magmesium sulfate, and concen- trated. The yield was .06g. The PMR indicates that the product is proabably the elimination between the tosylates, which leaves one tosylate on the molecule. PMR (CDC1.,): 1.23 (triplet, 6H, ester CH,'s), 1.43-2.00 (broad singlet, 6H, CH,'S), 2.20-2.50 (multiplet, 2H, allylic CH), 2.42 (singlet, 3H, CH,on tosylate), 3.70-4.30 (multiplet, 6H, ester CH,'s and bridgehead H's), 4.60-4.90 (multiplet, 3H, vinylic H), 5.10-5.45 (multiplet, 4H, vinylic H), 7.20- 7.50 (multiplet, 2H, tosylate H's), 7.55-7.95 (multiplet, 2H, tosylate H's). Preparation of the Dixanthate 18 from the Diol 14. A solution of sodium hydride (.05g, 1.25x107?mol, in 60% in mineral oil, washed with hexane) in .5 ml of tetrahydrofuran was placed in a 10 ml two~neck flask equipped with a reflux condenser, and placed under argon. The diol 14 (.13g, 4.0x107 mol) in .5 ml of tetrahydrofuran was added to this over 5 minutes at room temperature. Hydrogen gas was released. The solution was colored beige. After 5 minutes of stirring, carbon disulfide (.10ml, 1.7x107-mol) was added. The color went to dark peachy orange. After 2 hours of stirring at room temperature, methyl iodide (.iml, 1.6x107?mol) was added. 50 The color changed to reddish peach. After $# hour, ice was added. The product was partitioned between ether and water. The organic layer was dried with magnesium sulfate, and concentrated. The product is a viscous yellow liquid. The yield was .15g (75%). The product was characterized by PMR (CDC1 1.27 (triplet, 6H, CH,'s on the esters), 1.80 3) (broad singlet, 4H, CH,'s not near xanthates), 2.07 (singlet, 2H, CH, near xanthate), 2.14 ( singlet, 2H, CH, near xanthate), 2.53 (singlet, 6H, CH,'s on the xanthates), 4.20 (quartet, 6H, CH,'s on the esters and the bridgehead H's), 6.20 (multiplet, 2H, CH-OXan). One Method of Pyrolysis of the Dixanthate 18. A sample of the dixanthate was heated under argon at 130° for 4 hours in a round. bottom flask. A green liquid formed on the walls. This green liquid was separated off leaving the pure xanthate in the flask. The green liquid was characterized by PMR (CDC1,)s 2.73 (singlet). The dixanthate was heated another 2 hours at 190°. Bubbling was noticed and the reac- tion mixture became darker. The mixture was run through basic alumina with dichloromethane. It appeared to contain some starting material and a new product, which was later identified by comparison with the product of the new method of pyrolysis to be the product of the elimination of one xanthate (although this is not really clear). 51 Another Method for the Pyrolysis of the Dixanthate 18. A sample of the dixanthate was placed in a Kugelrohr bulb in a Kugelrohr. The pressure was lowered to .05 torr. The same green liquid impurity as before started coming over. The temperature was raised to 130° for 45 minutes, and more green liquid came over. This fraction was removed. The temperature was raised to 200° for 40 minutes. The fraction that came over was still the green liquid impurity. The reaction bulb contained only the starting material. Then the temperature was raised to 200° for 20 minutes at a pressure of .10 torr. A clear liquid came over . This was identified by PMR to possibly be the product of the elimination of one xanthate. PMR (CDC1,): 1.27 (triplet, approximately 6H, CH,'s on the esters), 1.80 (broad singlet, 4H, CH,'s not near xanthate),1.90-2.20 (multiplet, 2H, CH, near xanthate), 2.30 (singlet, approxi- mately 2H, CH, on xanthate), 3.60-4.30 (multiplet, 6H, 3 CH,'s on the esters and the bridgehead H's), 5.35-6.20 (broad multiplet, 2H, vinylic protons). This material was then heated in a flask under argon at 200° for 1 hour. The peaks at 2.30 and 2.53 decreased in size in the PMR, and no new vinylic protons showed. 52 Attempted Preparation of the Dibromide 19. A solution of L mol) and lithium the spiroditcsylate 15 (.50g, 7.9x10° bromide (.24g, 2.8x107?mol) in 10 ml of acetone was placed in a 25 ml flask equipped with a condenser. This was stirred at reflux under argon for 48 hours. White crystals pre- cipitated out, and the solution turned bright orange. The precipitate was filtered off, and water was added to the solution. The product was paritioned between ether ana water. The organic layer was dried with magnesium sulfate, and concentrated. The reaction mixture contained some aldol condensation product of acetone, some mono-olefin 13 resulting from the substitution reaction of the Br, and the subsequent elimination of Bros and some starting material. There was also an unidentified compound with one peak at 6.85 (broad singlet) in the PMR. The acetone aldol addition product was removed by distillation on a Kugelrohr (pressure was .3 torr, temperature was 250.55°), The remaining product was run down a dry Florisil column, eluting with a carbon tetrachloride to dichloromethane gradient. The unidentified compound came off in the early fractions with a PMR (CDCl 1.20 (singlet, 2H), 1.82 (singlet, 1H), 3) 1.98 (singlet, 1H), 2.11 (singlet, 1H), 2.28 (singlet, 2H), 2.42 (singlet, 1H), 6.85 (broad singlet, 1H). 53 Preparation of the Diselenide. A solution of diphenyl 4mol, Aldrich) in 2 ml of absolute diselenide (.10g, 3.2x10_ ethanol was placed in a three-neck flask equipped with a reflux condenser and an addition funnel, which contained the spiroditosylate 15 (.20g, 3.2x1074mo2) in .8 ml of ethanol. The system was placed under argon. While stirring, sodium borohydride (.03g, 7.9x107 mol) was added in portions over 10 minutes to the solution of the diphenyl diselenide. Bubbling occurred, and the solution changed from yellow to colorless, indicating ‘the formation of the sodium phenyl selenide. Then the spiroditosylate solution was added over 10 minutes. The reaction was stirred for 2 hours at room temperature. Then .05 ml of 10% HCl was added. This was extracted four times with hexane. The hexane extracts were washed with 10% HCl, saturated NaCl solution, and dried with Na,S0,- This solution was then put into the refrigerator at 4°, and a solid formed. The solution was decanted off. The solid was yellow. Its PMR gave only peaks in the phenyl region. The mother liquor contained some more of the solid, and an orange liquid which was characterized by PMR (CDC1.)s 1.27 (triplet, 10 H,-ester CH,'s and CH,'s near Se), - 1.82 (broad singlet, 4H, CH,'s not near Se), 2.22-2.42 2 (multiplet, 2H, CH-Se), 3.90-4.40 (quartet, 6H, CH,'s on 54 esters and the bridgehead H's), 6.90-7.80 (complex multiplets, integration not calculated due to the presence of the solid with PMR peaks in this region, phenyl H's). Preparation of the Mono-olefin 13. A solution of the spiroditosylate 15 (9.73g, 1.53x107"mol), and sodium iodide (28.61¢, 1.91x10°-!mol, J.T. Baker Chemical Co.) in 100 ml of acetone was placed in a 200 ml flask equipped with a reflux condenser. This was stirred at reflux under argon for 88 hours. The solution was very dark brown (due to I, formation) with a white precipitate (Tos-0-Na). Dichloro- methane was added. The resulting solution was washed with 10% sodium sulfite solution until the I, color was removed. Then the solution was dried with magnesium sulfate, and concentrated. The crude yield was 4.82g (107% due to comtamination with the aldol condensation product of acetone). This was then run through dry Florisil ina column 4 cm in diameter, packed 20 cm high. It was eluted with chloroform. The early fractions contained the product. The yield was 3.70g (82%) of a slighly viscous yellow liquid. The product was characterized by mass spectrometry, PMR, and 136 NMR. The mass spectral molecular weight was 294.1569 (calculated for C 0),N, was 294.1569). 1stiz2 PMR (CDC1,): 1.23 (triplet, 6H, CH's on the esters), 1.77 (broad singlet, 4H, CH,'s), 2.27 (broad singlet, 4H, 55 allylic CH,'s), 3.97-4.30 (quartet, 6H, CH,'s on the esters and the bridgehead H's), 5.60 (broad singlet, 2H, vinylic H's). 13¢ NMR (CDC1,): 14.29 (CH,'s of esters), 27.09 (CH,'s), 35.47 (allylic CH,), 56.98 (allylic CH), 61.79 (CH,'s on esters), 66.21 (bridgehead C's), 77.19: (spiro-C), 128.15 (vinylic 0), 129.19 (vinylic C), 157.26 (-CO,-). Preparation of the Allylic Bromide 20 from the Mono-olefin 13. A solution of the mono-olefin 13 (2.56g, 8.71x107-mol), n-bromosuccinimide (1.53¢, 8.59x107-mol, Aldrich), and .10g of dibenzoyl peroxide (M.C.B.) in 100 ml of carbon tetrachloride was heated at reflux for 2 hours. The end of the reaction was signalled by a precipitate floating at the surface. The mixture was cooled to 0°, filtered, and concentrated. The crude bromide was filtered through Florisil, eluting with ether. This was then concentrated to leave a light brown viscous liquid (2.81g, 87%), which was characterized by PMR (CDC1,): 1.27 (triplet, 6H, CH, 's on esters), 1.87 (broad singlet, 4H, CH,'s), 2.13-2.57 (multiplet, 2H, allylic CH,), 3.67-4.60 (quartet, 6H, CH,'s on the esters and the bridgehead H's), 5.00 (multiplet, 1H, CHBr), 5.67 (multiplet, 1H, vinylic H), 6.13 (multiplet, 1H, vinylic H). 56 Elimination of Hydrogen Bromide from the Allylic Bromide 20 to give the Spirodiene 16. A sample of the allylic bromide (2.142, 5.70x1072mol) was dissolved in 75 ml of ethyl ether in a 200 ml flask capped with a stopper. The stopper was removed momentarily, and solid potassium tert-butoxide (1.782, 1.59x107“mol, MCB) was added all at once. The mixture was stirred for 4 hours at room temperature. Then 5% HCl was added. The solution was extracted with ether,and then CH,Cl,. These layers were combined, washed with water, dried with magnesium sulfate, filtered through Florisil, and concentrated. There was a yield of 1.16¢ (3.97x107?mol, 69%) of a reddish brown slightly viscous liquid, which was characterized by PMR (CDC1. ): 1.30 (triplet, 6H, CH,'s on the esters), 2.07 (singlet, 4H, CH,'s), 3.80-4.50 (quartet, 6H, CH,'s on the esters and the bridgehead H's), 5.96-6.67 ( complex pattern characteristic of the conjugated diene in this type of ring system, 4H, vinylic H's). Preparation of the Diels-Alder Adduct of the Spirodiene 16. A solution of the spirodiene 16 (1.09g, 3.73x107mol) ana diethyl azodicarboxylate (.73 ml, 42x10? mol, Aldrich) in 50 ml of carbon tetrachloride was heated to 70° for 64 hours. The carbon tetrachloride was removed, and the remaining product was filtered through basic alumina with dichloro- 57 methane to yield 1.05g (2.25x107?mol, 60%) of a white solid, compound 21. The product was characterized by PMR (CDC1,): 1.30 (triplet, 12 H, CH,'s on the esters), 1.45-2.10 (multiplet, 4H, CH,'s), 3.80-4.40 (multiplet, 10 H, CH,'s on the esters and the bridgehead H's), 4.73 (broad singlet, 1H, allylic bridgehead), 4.87 (broad singlet, 1H, allylic bridgehead), 6.57 (multiplet, 2H, vinylic H's). cviimide Reduction of the Diels-Alder Adduct 21. A solution of the adduct 21 (1.05g, 2.25x107-mol) and hydrazine hydrate 2 (1.09 ml, 2.25x10 “mol, MCB) in 21 ml of ethanol was heated to 61° for 18 hours with 0, bubbled through the solution. 2 The ethanol was then removed. Ether and water were added. The product was partioned between ether and water. The ether layer was dried with magnesium sulfate, and then concentrated to give 0.9262 (1.98x107mol, 88%) of a white solid 22. The product was characterized by mass spectrometry, PMR, and 136 NMR. The mass spectral molecular weight was 468.2203 (calculated for CoyHyo0gNy was 468.2220). PMR (CDC1.,): 1.27 (triplet, 6H, CH,'s on the esters), 1.30 (triplet, 6H, CH,'s on the esters), 2.90 (broad singlet, 8H, all CH,'s not on esters), 4.00-4.60 (multiplet, 12 H, 2 ester CH,'s and bridgehead H's). 136 NMR (CDCl, ): 14.49 (CH,'s), 26.90 (CH,'s not on esters), 27.29 (CH,'s not on 3 58 the esters), 60.50 (bridgehead C), 61.15 (bridgehead C), 62.45 (CH, 's on esters), 62.71 (CH,'s on esters), 157.27 Preparation of the Bisdiazene 4 from the Adduct 22. The adduct 22 (473 mg, 1.01 mmol) was dissolved in 110 ml of isopropanol, and put in an addition funnel. The solution had argon bubbled through it for 12 minutes. Potassium hydroxide (1.11g, 24.1 mmol) in 50 ml of isopropanol was added to a three-neck flask equipped with a reflux condenser, addition funnel (containing the solution of 22) and a gas inlet. The system was flushed 15 times with argon using a firestone valve. The KOH/isopropanol was brought to a reflux. The solution of the adduct was added dropwise. The resulting solution was heated at reflux for 3 hours. Then the solution was concentrated to give a brown paste. During this time, as little air as possible was admitted. The paste was taken up in ethanol. A saturated solution of cupric chloride was added until a brown precipitate formed. The precipitate was filtered off through a fine frit filter. The product, a reddish brown solid, was then dissolved in 125 ml of 25% NH, OH solution to give a bright blue copper ammonia complex. The solution was extracted 6 times with 60 ml of dichloromethane, dried over Na,80,, and concentrated to give 59 a brown paste. This was sublimed at 110° and .03 torr to Zive 83.25mg (46.7%) of a white solid, which was then recrystallized from dichloromethane/hexane. The product was identified by comparison te that produced by Lisa McElwee-White according to Scheme III (note: this reaction was also done by Lisa McElwee-White). PMR (C.Dg): 0.28-1.10 (multiplet, 8H, CH,'s), 3.93 (broad singlet, 2H, bridgehead H's), 4.17 (broad singlet, 2H, bridgehead H's). 130 NMR (CD,Cl,): 19.42 (CH,), 20.33(CH,), 74.14 (spiro C), 76.09 (bridgehead C), 76.94 (bridgehead C). Preparation of the PTAD Adduct 24 of the Spirodiene 16. A solution of PTAD (.72g, 4..09x1072mol) in enough dichloro- methane to dissolve the PTAD was added to the spirodiene 16 (1.16¢, 3.97x107-mol) while swirling the reaction mixture at room temperature. The red color of the PTAD disappeared on contact with 16. The product solution was filtered through silica gel, eluting with ether. The product was a yellowish solid, compound 24 (1.78g, 96% yield). The product was characterized by PMR (CDC1,): 1.30 (triplet, 3H, ester CH), 1.35 (triplet, 3H, ester CH), 1.57-2.27 (broad doublet, 4H, CH,'s not on esters), 3.90-4.44 (multiplet, 5H, CH,'s on the esters, and one bridgehead H), 4.44-4,87 (multiplet, 3H, bridgehead H's), 6.54 (broad Singlet, 2H, vinylic H's), 7.44 (broad singlet, 5H, phenyl H's). 60 Diimide Reduction of the PTAD Adduct 24. A solution of the adduct 24 (1.88¢, 4.02x107?mol) and hydrazine hydrate (2.00 ml, 4.13x107“mol, MCB) in 70 ml of ethanol and 25 ml of dichloro- methane was heated to 65° for 28 hours while 0, was bubbled through the solution. Then the ethanol was removed. Dichloromethane and water were added, and the product was partitioned between them. The organic layer was dried with magnesium sulfate, and concentrated to give 1.66¢ of a yellow solid 25 (88%). The product was characterized by PMR (CDC1.): 1.25 (triplet, 3H, CH, on ester), 1.27 (triplet, 3H, CH, on ester), 1.53-2.23 (multiplet, 8H, CH,'s not on the esters), 3.93-4.53 (mutiplet, 8H, CH,'s on esters, and bridgehead H's), 7.13-7.60 (multiplet, 5H, phenyl H's). The product was also characterized by mass spectrometry. The mass spectral molecular weight was 469.1952 (calculated for Co 3H 706N~ was 469.1961). Preparation of the Bisdiazene 4 from the Adduct 25. The adduct (457 mg, .97 mmol) was dissolved in 100 ml of isopropanol, and put in an addition funnel. The solution had argon bubbled through it for 12 minutes. Potassium hydroxide (1.07g, 23.2mmol) in 80 ml of isopropanol was added to a three-neck flask equipped with a reflux condenser, addition funnel (containing the solution of the adduct), and a gas inlet. The system was flushed 15 times with argon using a 61 firestone valve. The KOH/ isopropanol solution was brought to a reflux. The adduct solution was added dropwise. The resulting solution was heated at reflux for 3 hours. Then the solution was concentrated. As little air as possible was allowing in during all transfers. The resulting paste was taken up in ethanol. A saturated solution of CuCl, was added until a brown precipitate was formed. The precipitate was filtered off using a fine frit filter to give a grayish brown solid. This was then dissolved in 175 ml of a 50% saturated NH, OH/ 50% water.solution to give a bright blue copper ammonia complex. The solution was extracted 5 times with 50 ml of dichloromethane, dried over Na5S0)5 and concentrated to give a brown solid. This was sublimed at 110° and .07 torr to give 31.37mg (18%) of a yellowish solid. The product was identified by comparison with that produced by other methods.