é: ‘ #742; ‘12:. ‘3 3.3 gm???» _..- ‘9 w I jiiiii , I; l.’;..'I'Iei-.{zii..i. ; :i- - "Ei.Ii.ii?,gf%I,"‘I' Ii”III iiiuI Iii‘iiiI‘iIII .i'rii ‘, i“ “W? "WE“ II}? MI] I i' ('7 , ;_ fifiraufi “£2?” .— .’ w..- mo;— w‘il 1’"IIIIIJI H “II II; a 3 {in Y1 {“r‘v n 2," 139.4 f ““044 3&9’9. 2;}. A‘. I) :ESIJQQQ , 23.. ,, . -L‘n *1 3- figs . :‘u \:. a .- H cure. ...- - y. A u -. - ' :3...‘€-‘? I"- -. .. .fi' ”‘V.' , ; u - --. -. . ¢ .. I _.'|q.i| DIIiI’i‘ .‘ n ‘. . " . m.» with mi ' .1 ~ . . S'i‘izt‘iéiir‘wzp i. . . i " i . ; .i it: :1in, i - 'i-.}i'i"il“-".:€~'.\; II“. in! 1.9M .Iirgm . .' I . iiiW'Iii/iii O . . M _,,,-_,- A . - ‘ A: . - - '. ' ' - ___.L— "' , ... ..- ”an, , -. . ... . -- Us..- 7 v... u . .- ." >. ‘ - - v v . L, _ “J4 ...- “5...: -' - r -. ' .. A: 3; 9M“ w-r- 1 .-.- I” V -. - v .- - ._...‘"'= “—- ‘ - ';_ 0o- - ...- . . -- .12.: 94-: .v. . 4‘ -— ~o-.—. . mun—u... arc... co-g-n 1.- ' . 1:35. “I”? IJIIV V!“ 1 - . ..., rt... - r ‘ -' .I II: - i; ' .'_",’._. \'I\\')OI {Ii 1.. .v . ‘. ~. '.. out: I III I VIII" I "I VIIIII’ “III“ I ‘ II! I‘llr I’j MIR"; . ,3 1.} ' 1 :LF’ {'3' ' ' 'II.‘ 12"? ' 'H' Ii "I“ ,‘II :I at .5: 1 . . ‘ '3. 4 i (II), '.i “III Hi ‘ '31' I '“wi‘d: . . I“ I'II"I|i:;"1IIIIII a r ‘ I ‘ O u a u "M' -> t 3.433% " an...» - ' ' , '— .., 9615-9- "’ ~ n {I :v. ' .. .. ' -»¢ - _ o . m Lad-r." _ I '- - ‘~ % u- ‘ . . d .“..v ' —, - I I ‘ ‘fim ".(l .. , ' .. .. . . - .- . u... .. , ,, .. "W . .. . _ _ _ "L. , .. _ - . DD< u'- .0. . - _ ',‘ . ~ ' "' " ‘ ' . .- . .p . . . “—40—..— ‘I. ”it“. Io]- I‘II I '; 1III‘IIiliI‘i i'”I"MI.'rJi ' 1:“; I I “ii i’l'ln II'IM Vii!" ‘ii ’i i”l HI WWW .11 i. I! “I'M“ ii“ ”IIIII MEI tIJ ”MI .; i:i M Iiib... If“ WI I iii II ii; liI’" “III [III] [II Wu. :' _.._.— :~...:..: - .: ‘37:.“ ..- ‘— m». 11%;}: SEC—— ~:‘ - - ~00— A ...-. - —-. -- A --, F23. - .r u . I‘ w "‘_. _7:‘=:.—u:“ ~ :— 3w. .‘ .' u h; ’ n n; u . . M . _ a ‘ p.- ‘ u— OuI-IL *4 o .3”, 'v-fl m. ~. v' 4 3“. . ' 1.31: I III?“ If I: "($1. iiiiz‘iI; . 1y >4 ‘ ‘gi * F1 ii I , I 35:23“; '1 i I \ if!“ LI-Euhl v '1‘ i I, i ‘i “I (1‘; ”In“ «III LI: ...:-« .: -iii‘iiiiiiiiiiiiiiii‘i“if ‘i 'iiiew‘ i ”i; "I“ I IIMIIIIIIt I, IiIItIIIIii III'iIIIJII‘i ii .. a” w- -._.._ .1”: -._.,‘.._.‘-:‘ .3:”-‘ -.—: w“ ._a-.- w I”. ' lli' “IE I? IIIIIII III’I‘IIIII ii il {Iiiii M». E III“: min“, ”I . I” .H WI; l. H M - , a: ‘as _ ' ":9'E:E* m . ‘ —— mm‘w IJ‘ N1; ‘ p, ' ‘ ' Universrty This is to certify that the thesis entitled THE PREPARATION AND CHEMISTRY OF NI(II) METALLOCYCLES presented by Mei-In Melissa Liu has been accepted towards fulfillment of the requirements for Ph . D . Chemistry degree in WM 9% Maéorv professor Date Nov. 11+, 1977 0-7639 THE PREPARATION AND CHEMISTRY OF NI(II) METALLOCYCLES BY Mei-In Melissa Liu A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1977 ABSTRACT THE PREPARATION AND CHEMISTRY OF NI(II) METALLOCYCLES By Mei-In Melissa Liu Six phosphine nickel cyclopentane complexes (P¢3, diphos, ¢2PCH2¢, PCY3' Me¢2P, and n-BuBP), two phosphine nickel cyclohexane complexes (P¢3, diphose) and an acyclic bis(triphenylphosphine)di-n—butyl nickel (II) complex were prepared by the reaction of suitable lithium reagent (1, 4-dilithiobutane, 1, 5-dilithiopentane, or n-butyl lithium) to the appropriate dihalobisphosphine nickel (II) complex at low temperature. The mode of decomposition of metallocycles under thermolysis and photolysis were examined. The relative ratio of the gaseous products formed in the solvents effected decomposition Of metallocycles was studied. It was found that strongly coordinating solvents promoted carbon- carbon.fragmentation products. The coordination number of complexes in 31 solution was determined by P nmr spectroscopy and/or molecular weight determination. These complexes decompose to give ethylene, cyclobutane, and 1-butene. Both the theoretical analysis and experimental results provided the conclusion that the ratio of the gaseous products is a function of ligand structure and coordination of the complex. Three coordinate nickel- locycle (P1 species) give 1-butene, four coordinate nickellocycle (P2 species) give cyclobutane while five coordinate nickellocycle (P3 species) give ethylene. ACKNOWLEDGEMENTS The auther wishes to sincerely thank Dr. Robert H. Grubbs for his conatant guidance and support, enthuiasm and friendship, which have all contributed to my development as a chemist. Tb Dr. James F. Harrison and Dr. Akira Miyashita, the auther would like to thank for their helpful discussion and technical assis- tance. I also would like to thank the DA 60 nmr group for the many hours they spent giving experimental assistance during the early stage of my nmr work. Gratitude is also extended to the Department of Chemistry, Michi- gan State University for financial assistance in the fOrm of teaching assistantship and research assistantship from the National Science Foundation in.the form of a grant administered by Michigan State University. Finially, I wish to express my deepest thank to my parents for their encouragement, and my husband, Joe , for his perspiration, inspiration and motivation. ii TABLEOFCONTENTS Page AcncNowLEDGmENIS........................ii LISTOFTARLFs.........................i" LISTOFFICURES........................Vi CHAPTER 1: PREPARATION AND ANALYSIS OF NICKEL(II) MMOCYCI’E I . I I I I I I I I I I I I I I I I I I 1 I ntrOd-uc ti on I I I I I I I I I I I I I I I I I I I I I I I 1 Results and Discussion. . . . . . . . . . merimenml I I I I I I I I I I I I I I I CHAPTER 2: EFFECT OF HEAT, LIGHT, SOLVENTS AND OLEFINS ON - THE DECOMPOSITION OF NICKEL(II) METALLOCYCLES . . . 26 I ntrduc ti on I I I I I I I I I I I I I I I I I I I I I I I I 26 Results and Discussion. . . . . . . . . . . . . . . . . ... 28 marimenml I I I I I I I I I I I I I I I I I I I I I I I I 43 CHAPTER 3: ETHYLENE AND CYCLOPENTANE FROM NICKELCII) HETALLOCYCLES................... ”'6 IntroduCti-on I I I I I I I I I I I I I I I I [+6 Results and Discussion. . . . . . “7 ExPerimental. . . . . . . . . . . CHAPTER 4: APPLICATION OF EXTENDED HfiCKEL CALCULATION TO DIE REACTION OF NICKEL(II) METALLOCYCLES . . . . 69 I ntrmuc tion I I I I I I I I I I I I I I I I 69 Calculation......... .......70 Results and Discussion. . . . . . . . . . 75 LIST OF REFERENCES. . . . . . . . . . . . . . . . . . . . . . . 78 iii 10. 11. 12. 13. 14. 15. LIST OF TABLES The yield and analysis data of tetramethylene nickel metal- locycles and III. . . . . . . . . . . . . . The yield and analysis data of pentamethylene nickel matal- locyCIes I I I I I I I I I I I I I I I I I I Products (76) formed from the thermal decomposition of tetra- methylene nickellocycles and III in solid state . . . . . . . Products (96) formed from the thermal decomposition of penta- methylene nickellocycles in solid state . . Products (96) fromed from the photolysis of I a’ Ic’ and I at 0 C in solution state. . . . . . . . . . Products (76) formed from the photolysis Of Ia in SOlid Smite I I I I I I I I I I I I I I I Products (96) formed from the photolysis and IIa and IIb in solid state. . . . . . . . . Products (75) formed from solvent effects on tionofIa at9°°CtiC. . . . . . . . . . . Products (76) formed from solvent effects on tion Of ICI I I I I I I I I I I I I I I I I Products (%) formed from solvent effects on tion of Ib and Ie at room temperature . . . Products (75) formed from solvent effects on tion of IfI I I I I I I I I I I I I I I I I Products (76) formed from solvent effect on tion Of IIa I I I I I I I I I I I I I I I I Products (76) formed from olefin effects on tion of Ie at room temperature. . . . . . . Products (76) formed from olefin effects on tion Of IIa I I I I I I I I I I I I I I I I Products (76) formed from triphenylphosphine decomposition of Ia at 9'01 1C in toluene. iv f ,Ic, andIe thermolysis of the decomposi- the decomposi- the decomposi- the decomposi- the decomposi- the decomposi- the decomposi- effects on the Page 11 11 31 36 37 39 #2 42 16. 17. 18. 19. 20. 21. 22. 23. 24. Page Products (%) formed from tri-y-butylphosphine effects on the decomposition of Ic at ifi-iI C in toluene. . . . . . . 63 Products (9%) formed from tri-n-butylphosphine effects on the decomposition of If at -50°C in toluene. . . . . . . . 64 Products (96) formed from triphenylphosphine effects on the decomposition of IIa at 16'C in toluene. . . . . . . . 65 Cyclobutane formed from oxygen and olefin effected decom- POSition Of Ic I I I I I I I I I I I I I I I I I I I I I I 66 Parameter used in the calculation. . . . . . . . . . . . . 70 Coordinates of 1(square planar metallocycle). (HBP)2NO . 71 Coordinates of 2(tetradehral metallocycle). (H3P)2NO . . 72 Molecular orbital configurations and the associated elec- tronic states of the square planar nickel metallocycle . . 73 Molecular orbital configurations and the associated elec- tronic states of the tetrahedral nickel metallocycle . . . 74 LIST OF FIGURES Vacuum line system for acidolysis reaction. . . . . . . Vacuum line system for thermolysis reaction . . . . . . Vacuum line system for trap to trap method. . . . . . . Temperature de endant 31P nmr spectrum of Ia in toluene “1m adda’ P 3/ Ia = 14I1 (mele) I I I I I I I I I I I 31P nmr spectrum of II in toluene with (A). added,P;¢3/ II = 4.1 (B). added, P¢ / II = 1 (c). no added. ata 92'C 3 a 3 Temperature dependant 31R nmr spectrum of IIa in toluenewiflladded,P¢3/Ila=1.2 o cocoon... Temperature dependant 31F nmr sgectrum of Ila in toluene with added P¢3 / IIa = .O 31P nmr spectrum of I in toluene with (A). no added néBujP (B). added néBuBP (C3. added n.3u3P, heat up to 80'C . . . Effect of added P¢3 on the decomposition mode Of Ia . . Effect of added néBu3P on the decomposition mode of Id. Effect of added néBuBP onthe decomposition of Ic . . . Effect of added naBu3P on the decomposition of If . . . Effect of added P¢3 on the decomposition of IIa . . . . State correlation digram of square planar metallocycle and tetrahedral metallocycle. . . . . . . . . . . . . . vi Page 15 43 51 53 E3 3% 59 62 77 CHAPTER 1 PREPARATION AND ANALYSIS OF NICKEL(II) PETAILOCYCIES Introduction Oxygen, nitrogen and sulfutheterocycles have played major roles in the development of organic chemistry. It is becoming apparent that transition metallocycles may have considerable importance in transition metal organic chemistry. Tetramethylene metallocycles have been proposed as intermediates in a number of transition metal catalyzed (2 + 2) cyclo additions of olefins (sdheme 1) and retro cyclo addition of olefins (sobeme 2). \ L Q —-> M i 1 (Scheme 1) H+ll :J + M -—e M ———> (>11 .(Scheme 2) Most cases such as the cyclodimerization of methylenecyclopropanes catalyzed by nickel(0) (eq. 1) reported by Binger (1) in 1972 involve reactive olefins. Ni(COD)7 > Pentane, -15 (7’ W + M (eQ- 1) 25% 65% 1 2 Based on.the structures of the products the authors proposed that the reaction goes through a metallocycle as show in scheme 3. [:>== + Ni(C0D)2 . (Scheme 3) _9 (COD) Ni \ / / Q —-> Hall (2) found (eq. 2) in 1973 that ethylene and acrylonitrile produce cyclobutane carbonitrile when catalyzed by nickelocene or cyclo- pentadienyl nickel carbonyl. NC @— Ni-——© N . (eq- 2) .A "Mr \/ Grubbs (3, A) has suggested that tetramethylene metallocycles could be source of the metal-carbene in some tungsten olefin metathesis catalyst systems (scheme 4). "=CHZ [flux —> O —s (74/5 (schemeu) W==CHZ Some metallocycles have been trapped or isolated (5,6) as shown below (eq. 3). —_—CF HFC Z (Eto) P °Fe(co)9 F2 + 3 a J > (Etc); 2(<10)2Fe (eq. 3) HFC=CF2 F2 F Halpern (7) identified metallocycles as intermediates (eq. 1+) in the isomerization of cubane derivatives by rhodium(I) catalyst. They trap R R —> E? -—> D P: COOCH3 (eq_ 4) ped the rhodacycle by using a stoichiometric amount of the catalyst (eq. 5). [Rh(CO)2C_L)L> (eq. 5) CO —-"/ CO Osborn (8) has isolated a metallocyclic intermediate 3 from an Ir(I) catalyzed imerization of norbornadiene (scheme 5). acetone / -' 5 P¢ M (Ir(1,5 CDWC‘I]2 ' CW NED Cl’ ‘NDD 9/ (Scheme 5) 1+ Also a nickel type analogue of 3 (eq. 6) was identified by Blackborow at. 2.1..(9) @\ e Ni (atom) + + dipyridine —-> /Ni (eq. 6) p ‘ .19 C The structure of b was confirmed by nmr, mass spectrum and elemental analysis. Some metallocycles have been prepared as stable complexes. Plati- num metallocycle g have been prepared by both Grubbs' and Whiteside's groups independently and confirmed with x-ray diffraction by Grubbs (10). Whiteside (11, 12) has demostrated that platinum metallocycles possess unusual stability relative to acyclic analogue d/ and did not give C-C bond fragmentation on thermolysis. ¢3P\Pt ¢3P\ NR 113/ 933? ’ “I: c d 4/ N Whiteside (13, 11+) has also prepared and demostrated that metallocycles Q can apparently be formed by the addition of ethylene to reduce titanocene species, however, this titanium species 3 proved too unstable go for spectral analysis . 5 Diversi (15) and his coworker reported in 1977 that the reaction of RhI2(P¢3)(a5.C5Me5) 3 with di-grignard reagent Bng-(CHz)n-MgBr 5 (n = 5 or 6) led to the rhodacycloalkanes h and 3, respectively. React- ion of 3 and g (n = Llr) gave the rhodacyclopentane :1 and the rhodium(I)- ethylene complex is. 5 -C M I (it 5 e5 / /Rh\ Bng-(CH2)n-MgBr 5531’ I \H: 5 n=4, 5, or6 (55-0 e) 5 e 5M55>hO (7 -C5M 5):RhO ¢3P ¢3p 3-3 l 5-C Hzc=g= CH2 (‘7 5M85)\ ~ Rh Rh / \ P (7545Me5) / p953 953 k 3 In the solid state, 3 and i decomposed at 160'C under argon; no carbon-carbon bond cleavage was observed, the only volatile products being n-pentenes and n-hexenes from h, and ‘1' respectively. Similar de- composition modes have been observed in the case of some platinum(II) metallocycles. 6 The method of preparation of )5 and the simultaneous evolution of ethylene strongly suggest that the metallocyclopetane derivatives j is the precursor of is undergoing carbon-carbon bond cleavage. Efforts to isolate j have so far been unsuccessful, invariably leading to samples fl contaminated by lg. (15C Me ) H23: 2 5 5/Bh ———>(75-05Me5) Rh\P¢ + H20=CH2 (eq. 7) Recently, Grubbs (16) found a simple system, which combines the best properties of the platinum, rhodium, and titanium complexes and undergoes all of the major reactions previously observed for metallo- cycles: (a) Reductive elimination of cic—alkyl groups, (b) fi-hydride elimination, and (c) carbon-carbon bond cleavage. What factors encourage those reactions is an intersting subject for further study. Results and Discussion This chapter deals with the preparation, purification, and analysis of six phosphine nickel cyclopentane complexes (P¢5, diphos, ¢2PCH2P, PCYB' Me¢éP, and néBuBP), two phosphine nickel cyclohexane complexes (P¢3 and diphos) and an acyclic bis(triphenylphosphine)di—n- butyl nickel(II) complex. There are several reasons for choosing tertiary phosphine as ligands: (a) A large number of tertiary phosphine are inert to various substrates in homogeneous catalysis. (b) 'Dlere exist well-established 7 synthetic routes to phosphine ligands. (c) One theory suggests the tran- sition metal phosphine bond possesses the proper balance between r—donating and at-back bonding needed for stabilization of a metal- carbon bond. (d) Tertiary phosphines not only coordinate well on nickel but also readily dissociate again, making available the coordination site necessary for a catalytic system. The target complexes were prepared by the addition of a suitable lithium reagent (1,4-dilithiobutane, 1,5-dilithiopentane, or n-butyl lithium) to the appropriate dihalo bisphosphine nickel(II) complex in ether at -78'C. The reaction was slowly warmed to -10°C and the result- ing yellow solid was isolated by filtration (eq. 8, 9, and 10). = (mg), a. P2 PZNiCI2 + Li-CHZ-CHZ-CHz-CHZ—Li—s PzNi b' P2 = (¢2Pmi2¢)2 ( ) 0. P2 = Diphos eq. 8 _ Ia_f d. P2 — (PCY3)2 e. P2,: (PMe¢2)2 f. P2 = (Pn-Bu3)2 P mm + Li-CH -CH -CH -CH —CH2-Li-> PzNi: ) a. P2 (W3)2 2 2 2 2 2 2 (eq. 9) b. P2 Diphose II a,b (¢3P)2Ni012 + 2 n-BuLi__) ($213P)2Ni(Bu)2 (eq. 10) III All reaction should be carried out under purified argon and at low temp- erature below -10'C. If the reaction was carried out with a large excess of dilithium reagent or at higher temperature, other colored solids were produced. In the synthesis of bis(triphenylphosphine)tetramethylene nickel (II) or bis(triphenylphosphine) pentamethylene nickel(II) the use at a 8 large ratio of dilithium reagent to nickel dichloride resulted in either the precipitation of brown solid or no solid at all. The properties of the brown solid were not investigated. In the preparation of the yellow bis(diphenylmethylphosphine)tetra- methylene nickel(II) complex, raising the temperature resulted in the formation of orange particles. Thermodecomposition gave ethylene (56%) and cyclobutane (33%) only. Once isolated the orange complex was stable at room temperature and fairly stable in deoxygenated water. This could be a nickel mono- or diethylene complex. In the preparation of bis(triphenylphosphine)dinahutyl nickel(II) raising the temperature gave a dark brown solid. This dark brown solid was stable at room temperature but gave off butene upon addition of excess triphenylphosphine. An acyclic n-butyl ligand would have no ring strain to hinder 9-hydride elimination and could thus generate butene. The possible pathway shown as follow: (¢3P)2Ni\,v‘“ —-> (¢3P),,Ni.,-H —> (¢3P)nNHL\+ N (scheme 6) (¢3P)nNi-u‘ (¢3P)nNi-P¢3 + aasz (scheme 7) ¢, ' In the preparation of bis(tricyclohexylphosphine)tetramethylene nickel (II) (scheme 8). Raising the reaction temperature or adding excess dili- thium gives a variety of products which have been investigated by Dr. Miyashita in detail. Apparently the ratio of lithium reagent to nickel dihalide, the reaction time, and the temperature are critical factors in the synthesis of the target compounds. The use of fresh lithium ribbon in the preparation of dilithium reagent is also important. It may 9 (CYBP) 2Nix2 + LiJ—LLi ether . ~40°C~-20 C (CY3P):j] Ni ether,.8 hr. er, Rexcess Li-I__1-Li RT (23 c) (CY 3P) 2Ni(C2H (CY 32P) Ni LiJ_LLi (ether)?! white yellow solids dark brown solids (scheme‘8) improve the yield by reducing contamination. To get rid of contaminants (mainly lithium halide and nickel di- halide), the complexes were recrystalized from toluene, ether, or nphexane as appropriate solvent. For non-water sensitive complexes, ether is a good solvent (low boiling point, easy to handle, bigger crystals). For water sensitive complexes, toluene is good, since toluene dose not dissolve the main impurities at low temperatures. However, the boiling point of toluene is higher and crystalization is slower than in ether. Gaseous analysis was carried out in vacuum line at low temperature to avoid the production of thermaldecomposition products. . + o . - - + PZNl (CH2)n + H ( 30 c) -—> H3C (CH2)n_2-CH3 Solids (eq. 11) n = 4, 5 Nickel analysis was done using the glyoxime, This extraordinar- ily sensitive reagent for nickel depends on the fact that ak—dimethyl- glyoxime added to an ammoniacal solution containing a nickel salt 10 produces an intese scarlet coloration (16). This reaction can routinely detect nickel at a concentration of one part per million in water. With care, it is possible to detect one part in 30,000,000 (17). Tertiary phosphine can easily be oxidized to the stable tertiary phosphine oxide. This characteristic was used for phosphine analysis. Diphos tetramethylene nickel(II) is the most stable complex. The chelated ligand and five membered ring angle probably account for this stability. Its molecular weight suggests that it is a monomer. The molecular weight of bis(tricyclohexylphosphine)tetramethylene nickel(II) suggests that in solution one phosphine is completely dissociated from the nickel. Steric hindrance could be the cause since tricyclohexylphosphine is a bulkyl group. The molecular weights were determined by freezing point depression of benzene. 11 Table 1. The yield and analysis data of tetramethylene nickel metallo- cycles and III. Compound 96 Yield M. wt. P/Ni Cq/Ni Ia (¢3P)2Ni:j 36 626 (calc. 639) 2.0 (2.0) 0.98 (1.0) Ib (¢CH2P¢2)2NO 32 659 (calc. 667) 2.0 (2.0) 0.99 (1.0) Ic Diphos N13 25 512 (calc. 513) 2.0 (2.0) 0.96 (1.0) Id (CY3P)ZNi:) 1+3 343 (calc. 675) 2.0 (2.0) 0.97 (1.0) Ie (Me¢2P)2Ni:j 20 - 2.1 (2.0) 1.00 (1.0) I f (n—BuBP)2Ni3 13 - NDa 0.90 (1.0) III (¢3P)2Ni(n-Bu)2 21 - 1.9 (2.0) 1.80 (2.0) aND = not determined Table 2. The yield and analysis data of pentamethylene nickel metallocycles. Compound 96 Yield P/Ni C 5/Ni IIa (¢3P)2Ni ) £10 2.0 (2.0) 0.97 (1.0) IIb Diphos N11 > 28 2.0 (2.0) 1.10 (1.0) 12 Experimental General Methods: All reaction involving organometalic compounds were carried out under prepurified nitrogen or argon using an activated Q reactant column (18). Diethyl ether, toluene, and benzene were distilled from sodium benzophenone ketyl. Absolute ethanol was used in systhetic work after bubbling with argon. The solid complexes decompose in the air with the production Of smoke. Bistriphenylphosphine tetramethylene nickel(II), bis(benzyldi~ phenylphosphine) tetramethylene nickel(II), and diphos tetramethylene nickel(II) are not sensitive to water. The other complexes are very sens itive to air or water, however, so all manipulations were carried out under deoxygenated anhydrous conditions. The 1H nmr spectra were run on a Varian T-60 or a Varian A-56/60 D nmr spectrometer. Chemical shifts are reported in parts per million down field from tetramethylsilane (TMS). The 31F nmr spectra were determined by a multinuclear nmr DA 60 with phosphoric acid as external standard. Gasses were analyzed by gas chromatography on a Perkin Elmer 900 or a Varian 1400 equipped with flame ionization detectors. Products were identified by comparison of retention times to those authentic samples. identity was considered established when equal retention times on two columns with different stationary phases (19) were obtained. Product yields were determined by the response relative to an internal standard with consideration of a cofactor or by a vacuum line equipped with a barometer. Molecular weight was determined by a cryoscopic method. 13 Synthesis of bis(triphenylphosphine) nickel(II) dichloride: The synthetic procedures were modified from Venanzi (20). Triphenyl- phosphine used was recrystalized from ethanol. To a hot solution of 52.6 g (0.2 mmole) 0f triphenylphosphine in 500 ml of glacial acetic acid was added with stirring to a solution of 23.8 g (0.1 mmole) of nickel(II) chloride hexahydrate, 20 ml of water, and 250 ml of glacial acid. The Olive green microcrystalline precipitate, when kept in contact with the mother liquor overnight, gave dark blue crystals of bis(triphenylphos- phine nickel(II) dichloride (80% yield) which were vacuum filtered, washed with glacial acetic acid and dried in vacuo. Preparation and standardization of 1,4.dilithiobutane: (21) In a 500 ml three-necked round bottom flask fitted with an argon line, 250 ml dr0pping funnel, oil bubbler, and stirring magnet were placed 100 cm of shaved lithium ribbon and 250 ml of dry, oxygen free diethyl ether. To the reaction was added, dropwise over two hours, a solu- tion of 25 ml of 1,4-dibrom0butane in 150 ml of ether. The reaction was initially allowed to proceed at room temperature for a few minutes, then it was cooled to 0'C. Following the addition of the dibromobutane, the mixture was stirred for 30 minutes. The solution was allowed to settle overnight in a refrigerator; The liquid was then separated by filtration and stored under nitrogen at O'C. The 1,4-dilithiobutane was stable for approximately one week at 0'0. The concentration of 1,4-dilithiobutane solution was determined by removing a 10 ml aliquot of the filtered solution and adding it to 2 m1 of chlorotrimethylsilane under nitrogen. This generated quantitatively 1,4-bis(trimethylsilyl)butane. Durene (1,2,4,5-tetramethylbenzene) was 14 added as a standard (cofactor = 1) and the solution was analyzed by glc (8 ft 3% QF-1 at 80'C flow rate 30 ml/min). Preparation and purification of bis(triphenylphosphine)tetramethylene nickel(II): The reaction was carried out under positive N2 pressure in a 100 ml side-arm round bottom flask equipped with a teflon-coated stirrer bar. Bis(triphenylph0sphine) nickel(II) dichloride (4 g, 6.2 mmole) and dry, oxygen free ether (20 ml) were placed in the flask. After cooling the reaction mixture to -78°C (dry-ice-ethanol bath), 62 ml (12.8 mmole) of 1,4-dilithiobutane in ether was slowly added by syringe. After addition the reaction mixture was stirred for 10 minutes, then the bath was removed and the stirring maintained. 0n partial warming the black starting material turned to dark violet and then dissolved in the ether, forming a dark solution. 0n warming still further (—10'C) a bright yellow solid was produced. As soon as the yellow solid ceased forming, the reaction was filtered. Only the top parts of the suspension were taken for filtration. The bright yellow solids were washed with dry, oxygen free ether several times and then dried under vacuum. The dried yellow solids were dissolved in dry oxygen free toluene at -15.0 and then filtered. The resultant clear yellow solution was partially condensed to 90%.of the original volume under vacuum, and then 10-20% of n-hexane was added (stop adding n-hexane before the solution become cloudy). The solution was kept in a dry-ice box. Bright yellow crystals were obtained and dried under vacuum giving a yield of 36%:based on starting material. The 1H nmr spectrum at -10’C in toluene-d6 showed multipletes at 15 1.67 and 1.87. A single 31 P peak was observed at -42.4 ppm over the temp- erature range of -10.0 to -90'C. The gaseous analysis from acidolysis, nickel analysis, and phosphine analysis showed P/Ni = 2.0 (theo = 2.0), CL/Ni = 0.98 (theo = 1.0). Molecular weight determined 626 (calc. = 639). Gaseous products from acidolysis of nickel(II) complexes: The solid nickel complex was transfered into a small Schlenk tube having two joints and one side-arm, and the weight of the sample was measured. The Schlenk tube was connected to a vacuum line (see Fig. 1) after one joint was fitted with a bent tube containing concentrated sulfuric acid. The tube was evacuated and the whole system was closed. The sample was decomposed with concentrated sulfuric acid at -30'C to 0.0 by turning the bent tube up. Gasses formed from the acidolysis were trap- ped using a Cold finger in a liquid nitrogen bath. The pressure was measured with a Hg manometer. Gaseous products were also analyzed with glc or mass spectrometer. .___fl=vacuum outlet l? thermal bath ’9 Mg manometer solid sample ‘\ l—— \ cold finger (L3 Figure 1. Vacuum line system for acidolysis reaction 16 Phosphine analysis of bis(triphenylphosphine)tetramethylene nickel(II): The residue which resulted from the acidolysis was oxidized with hydrogen peroxide at -10'C to giving a white solid (triphenylphosphine oxide). This white solid was extracted with ether several times and the ether solution was neutralized with sodium carbonate. The ether was evaporated off and the triphenylphosphine oxide was weighed until a cons- tant weight was obtained. Nickel analysis : (22) To the solution which resulted from the decomposition of bis(tri- phenylphosphine)tetramethylene nickel(II) with sulfuric acid was added 150 ml of water. A white precipitate (triphenylphosphine and/or triphenyl- phosphine oxide) formed immediately. To facilitate the removal of this white solid, the mixture was first heated to 70‘C, then cooled to room temperature, and finally vacuumed filtered to give a clear solution. Tb the clear solution at 60°C was added 20 ml of a freshly prepared 1% solution of dimethyl glyoxime in absolute ethanol. Concentrated ammo- nium hydroxide was added dropwise very slowly until no more of the bright red flocculent precipitate formed. The mixture was digested for one hour, cooled to room temperature, and vacuum filtered onto a tared fritted glass crucible. The solid was dried for several hours at 130.0, its weight determined, and the percent nickel calculated using the fellowing equation. % Ni _ Wt. of ppt. (g) x 20_.313 Wt. of sample (g) 17 Synthesis of benzyldiphenyllhosphine : Lithium ribbon (35 cm, 3.5 g, 0.507 mole) was placed in a 500 ml two-side-arm round bottom flask containing 100 ml of tetrahydrofuran. The flask was fitted with a dropping funnel containg 80 ml of tetrahydrofuran and 34 ml (0.188 mole) of diphenyl chlorophosphine. The lithium ribbon was stirred using a magnetic stirring bar; The tetrahydrofuran solution of chlorodiphenylphosphine was added to the lithium over a two hour period. An ice bath was periodically used to prevent the temperature from rising too high. After addition was completed the reaction was stirred for four more hours and then allowed to settle. The liquid was transferred to another 500 ml two-side-arm round bottom flask which contained a stirring bar; The flask was fitted with a 125 ml dropping funnel containing 80 ml of tetrahydrofuran and 27 ml (0.235 mole) of benzyl chloride. This was degassed with nitrogen for 15 minutes in the addition funnel. The benzyl chloride solution was added dropwise over a two hour period. After addition of sufficient benzyl chloride to give a reddish brown solution, the addition was st0pped, the addition funnel removed, and the reaction stirred for half hour. One hundred milliliters of degassed saturated ammonium chloride solution was added. This was left stirring overnight. The top layer (tetrahydrofuran) was transferred to another side arm round bottom flask containing calcium chloride (degassed by evacuation and diffusion of nitrogen). The aqueous solution was washed with three 75 ml portions of oxygen free benzene which was then added to the flask containing calcium chloride. All removals were made by syringe. The combined organic layers were condensed and the remaining solvent (about 100 ml remained) was removed by vacuum distilled (with the pot heated to about 100 degrees). The system was filled with argon and cooled 18 and 400 m1 of isopropyl alcohol was added to it and allowed to stand overnight. After filtration the solid was recystalized from isopropyl alcohol again, and gave benzyl diphenylphosphine (yield 5M%, m. p. 72L 73‘0) (23). The 1H nmr spectrum showed peaks at 3.37 (2), 7.07 (5), 7.25 (10). Synthesis of bis(benzyldiphenylphosphine) nickel(II)dichloride: To a hot solution (60’C) of 2.0781 g (7.53 mmole) of benzyl phos- phine in 40 m1 of galcial acetic acid was added with stirring a solution of 0.8928 g (3.75 mmole) of nickel(II) chloride hexahydrate in 2 ml of water and 25 ml of glacial acetic acid. The resultant pink solution was kept in a refrigerator overnight where upon dark violet needle crystals formed. The product was vacuum filtered, washed with glacial acetic acid and dried under vacuum to give an 85% yield of bis(benzyldiphenylphos- phine) nickel(II) dichloride. The IR spectrum showed Ni-Cl bands at 250 cm-1, 340 cm-1. Preparation and purification of bis(benzyldiphenylphosphine1 pegggme- thylene nickel(II): Ib A solution of (2.6 g, 3.8 mmole) bis(benzylphenylphosphine) nickel (II) dichloride in 10 ml of dry oxygen free ether was placed in a 100 ml side-arm round bottom flask equipped with a teflon-coated stirring bar and maintained under argon. After cooling the reaction mixture to -78'C (dry-ice-ethanol bath), 80 ml (8 mmole) of 1,4—di1ithiobutane in ether was slowly added by syringe. After addition was complete the reaction mixture was stirred for 10 minutes, and then allowed to warm partially while stirring. 0n warming the dark violet solid redissolved in the ether 19 On further warming (-15°C) a bright yellow solid fell out of solution. As soon as precipitation was complete the solid was filtered and washed several time with ether. The yellow solid was washed with deoxygenated water (not sensitive to water), to get rid of lithium salts. After being washed with other again the product was dried under vacuum and recrystlized from ether (yield 32% of Ib). The gaseous analysis from acidolysis, nickel analysis and phosphine analysis from phosphine oxide showed P/Ni = 2.0 (theo = 2.0), Cu/Ni = 0.99 (theo = 1.0). Synthesis of ethylen bis(diphegylphosphine) nickel(II)diChloride: A solution of 5.15 g of nickel(II) chloride hexahydrate in 200 ml of ethanol was added to 8.02 g of 1,2-bis(diphenylphosphine)ethane in 600 ml ethanol (50' C). After the reaction had stirred, orange feathery needlelike crystals formed. The solution was allowed to cool and the orange cystals were filtered under vacuum, washed with ether and dried under vacuum giving an 82% yield of ethylene bis(diphenylphosphine) nickel(II) dichloride. Preparation and urification of ethylene bis(diphenylphosphine) tetra- 'p§thylene nickel(II): Ic ' All procedures were carried under purified argon. A solution of 4'g (7.6 mmole) bis(diphenylphosphine) nickel(II) dichloride in 20 ml of oxygen free ether was placed in a 100ml side-arm round bottom flask equipped with teflon-coated tirring bar. The reaction was cooled to -78°C (dry-ice-ethanol bath) and 54 ml (13 mmole) of 1,4-dilithiobutane in ether was slowly added by syringe. 20 After addition the reaction mixture was stirred for 10 minutes, then the bath was removed and the stirring maintained. 0n warming, the orange solid dissolved in the ether giving a dark brown solution. On further warming and stirring a bright yellow solid came out of solution. As soon as precipitation was complete the product was filtered, washed several times with ether until a negative lithium test was obtained (or washed with deoxygenated water and ether) and dried under vacuum. The dried yellow solids were recrystalized from ether and dried under vacuum giving 25% of ethylen bis(diphenylphosphine) tetramethylene nickel(II). Analysis P/Ni = 2.0 (theo = 2.0), Cu/Ni = 0.96 (theo = 1.0). The 1H nmr showed a broad peak at 1.1-1.8 range (overlapping the ethane on the phosphine). The 31P nmr showed peak at -45.5 ppm. Lithium test: Ferric salts react with periodates to yield a precipitate of a ferric periodate complexs This precipitate is solubale in excess perio- date solution and also in excess potassium hydroxide solution. The resulting alkaline solution of the ferric periodate complex is a selec- tive reagent for lithium, since it gives a white precipitate (LiKFeIO6), even from dilute solution and in the cold. Sodium and potassium give no precipitate: ammonium salts, all metals of group I to IV and Mg should be absent. Sensitivity: 0,1 ug lithium, concentration limit = 1 : 100,000. The ferric periodate reagent is prepared by dissolving 2 g of potassium periodate in 10 ml of freshly prepared 2 M potassium hydroxide solution and diluting with water to 50 ml. Then added 3 ml of 10 percent ferric chloride solution and diluted to 100 ml with 2‘! potassium hydroxide solution. The reagent is stable. 21 Procedure: Place a drop of neutral or alkaline test solution in a micro test tube, and add 1 drop of saturated sodium Chloride solution and 2 drops of the ferric periodate reagents Simultaneously carry out a blank test with a drop of distilled water. Immerse.both tubes for 15-20 sec- onds in water at 40-50.C. 'A white (or yellow white) precipitation indicates the presence of lithium; the blank remains clear. Diphenylmethylphosphine: Tb a stirred solution of 50 ml (0.278 mmole) of chlorodiphenyl- phosphine in 90 ml of ether at 0'0 was added 135 ml of methyl lithium (0.279 mmole) solution over a two hour period. The ice bath was removed and the mixture was stirred for an additional thirty minutes. The org- anic layer was filtered (or decanted) and fractionally distilled (b.p. 108-110’0, 0.15 mm.Hg vacuum), giving 36 g (66.6%) yield of diphenyl- methylphosphine. The 1H nmr showed 1.7 (3) and 7.5 (10). Bi§(diphegylmethylphosphine) nickel(II)dichloride: (24) A solution of 12 g (60 mmole) of diphenylmethylphosphine dissolved in 200 ml hot ethanol (degassed) was mixed with a solution of 7.13 g (30 mmole) of nickel(II) chloride hexahydrate dissolved in 200 ml of hot ethanol. After mixing, the solution was violet. The solution was kept in a refrigerator overnight and the violet crystals which formed were filtered and dried under vacuum, giving 13.6 g (91.3% yield) of bis(di- phenylphosphine) nickel(II) dichloride (m.p. 1500C). 22 Bis(diphenylmethylphosphine)tetramethylene nickel(II): Ie A solution of 51.6 g bis(diphenylmethylphosphine) nickel(II) di- chloride in 20 ml of ether was placed in a 1000 m1 side arm round bottom flask fitted with a stirring magnet, gas outlet valve (oil bub- bler) and a dry-ice-ethanol bath. The flask was first purged of air by flushing with argon and start stirring. An excess of 1,4—dilithiobutane in ether (400 ml, 0.056 mole) was added dropwise at -78'C. After add- ition, the bath temperature was allowed to increase until a light yellow precipitate came out. The precipitate was recooled to -78'C and filtered under argon using cold jacketed fritted filter. The light yellow solids were recrystalized from ether giving 20%iyield of bis(diphenylmethyl- phosphine)tetramethylene nickel(II). Gaseous analysis by acidolysis, nickel analysis, and phosphine analysis (25) by acetoacetate ligand exchanged gave P/Ni = 2.06 (theor = 2.0), Cu/Ni = 1.0 (theor = 1.0). The 31F nmr showed -4.3 ppm (-90.C, ether). BiS(tri-n-buiylphospbineQ,tetramethylene nickel(II): If A solution of 3 g (0.0056 mmole) of bis(tri-n-butylphosphine) nickel(II) dichloride dissolved in 10 ml of ether was cooled to -78'C. To this was added 51 ml of dilithium butane solution (0.22 M’in ether). The reaction mixture was kept under -35RC for several hours (with stir- ring) until a yellow precipitate formed. The solvent was evaporated under vacuum at low temperature (-30°C). The product was extracted with pentane several times until only lithium chloride (white solid) remained. The pentane solution was condensed to half its original volume at low temperature (-30'C) and kept in a dry- iceAbox for two weeks. The resulting yellow crystals were dried under 23 vacuum at low temperature. Giving about a 13% yield of bis(tri-n-butyl- 31 phosphine)tetramethylene nickel(II). The P nmr at +1.5 ppm (toluene, —90°C) . Preparation of bis(triphepyiphospiine) di-n-butyl nickel(II): III All steps were carried at below -15°C. A solution of 5 g (7.6 m mole) of bis(triphenylphosphine) nickel(II) dichloride in 70 ml of dry oxygen free ether was placed in a side arm flask equipped with a teflon coated stirring bar. The solution was cooled to -78'C, and 8 ml (2.4 M) of nebutyl-lithium-n-hexane solution was added very slowly. After add- ition was complete the mixture was warmed up to -20'Ca:-10'C (with stir- ring). The yellow red solids which formed from the dark solution were filtered under -20'C using a frit equipped with a cooling jacket (filtration and procedures must be done quickly and at low temperature). The solids were water sensitive and too unstable to recrystalize from toluene. The solids were repeatedly waShed with cold ether and.then dried under vacuum at -10'C in the dark. Giving a 21% yield of bis(tri- phenylphosphine) di-n-butyl nickel(II), analysis P/Ni = 1.9 (theor = 2.0), Cu/Ni =1.8 (theor = 2.0). Deutero chloric acid was generated by slowly combining 10 ml deu- terium oxide and 5 ml of acetyl chloride under nitrogens The deutero chloride solution was then added to bis(triphenylphosphine) di-n-buty nickel(II). The gas above the resultant solution was removed from the flask by syringe and the butane in the gas was isolated by glc. The butane was then analyzed by mass spectros00py. The mass spectrum exhi~ bited a parent peak of m/e 59 which corresponds to butane-d1. 24 Preparation and standardization of 1,5-dilithigpentane: In a 500 ml three-neck round bottom flask fitted with an argon line, 250 ml dropping funnel, oil bubbler, and stirring magnet were placed 100 cm of shaved lithium ribbon and 200 ml of dry oxygen free diethyl ether. A solution of 30 ml 1,5—dibromopentane in 150 m1 of ether was added dropwise over two hours. The reaction was carried out at O'C. Following the addition of the dibromopentane, the mixture was stirred for 30 minutes. The liquid was separated by filtration and stored under nitrogen at 0°C. The concentration of the 1,5—dilithiopentane was determined by removing a 10 ml aliquot of the filtered solution and adding to it 2 ml of chlorotrimethylsilane under nitrogen. Durene (1,2,4,5-tetramethyl- benzene) was added to the quantitatively generated 1,54bis(trimethyl- silyl)pentane as a standard (cofactor = 0.8) and the solution was anal- yzed by glc (10% carbonwax on 60/80 mesh Chromosorb W at column temp- erature 110'C and flow rate 15 ml/min). Pre ration and purification of bis(triphegylphosphine)pentamethylene nickel(II): IIa A solution of 6.53 g (10 mole) of bis(triphenylphosphine) nickel (II) dichloride in 30 ml ether was placed in a 300 ml side-arm round bottom flask equipped with teflon-coated stirring bar. After cooling the reaction mixture to -78‘0, 77 ml (0.26 p) of 1, 5-dilithiopentane in ether was added dropwise by syringe. When the addition was complete the reaction mixture was stirred for 20 minutes and then allowed to warm. 0n warming, the black starting material dissolved forming a dark brown homogeneous solution. On further stirring yellow orange solids formed 25 and were then filtered (below -20'C) and dried under vacuum. The yellow orange solids were dissolved in dry oxygen free toluene at -10°C. After filtration n-hexane was added to the clear yellow solu- tion which was then placed in a dry-ice box; .The yellow brown crystals which then formed were dried under vacuum. Giving a 40% yield of bis(tri- phenylphosphine)pentamethylene nickel(II). The 1H nmr spectrum Showed multiplets at 0.96 and 0.73. The 31F nmr spectrum showed a peak at -33.6 ppm. The gaseous analysis from acidolysis, nickel analysis, and phosphine analysis showed P/Ni = 2.1 (theor = 2.0), C5/Ni = 0.97 (theor = 1.0). Preparation and urification of the ethylene bis(diphenylphosphine)penta- methylene nickel(II : IIb An ether solution of ethlene bis(diphenylphosphine) nickel(II) di- chloride was cooled to below -45‘C with stirring, and to this a ether solution of 1,5_dilithiopentane was added by syringe.- After the 1,5—di- lithiopentane had been added (1.5~2.0 mole excess) the pot temperature was allowed to increase. 0n warming the orange solution gave yellow solids. Rapid filtration (below -20°C) of the yellow solids was followed by washing with coold ether. The product was dried under vacuum at -20°C ‘V-10'C in the dark. The yellow solids were dissolved in dry oxygen free toluene at -10'C (starting material and lithium halide are not solubale in toluene at this temperature). After filtration a clear yellow solu- tion was obtained. Tb this was added 10-20%10f nshexane. The solution was cooled to -50‘C giving a 28% yield of ethylene bis(diphenylphosphine) pentamethylene nickel(II) as yellow crystals which were dried under vacuum at -10‘C in the dark. Analysis P/Ni = 2.0 (theor = 2.0), Cj/Ni = 1.10 (theor = 1.0). The 31F nmr showed at ~45.5 ppm. CHAPTER 2 THE EFFECT OF HEAT, LIGHT, SOLVENTS AND OLEFINS ON THE DECOMPOSITION OF NICKEL(II) METALLOCYCLES Introduction The instability of transition metal-carbon bonds is a problem of fundamental importance in organotransition metal chemistry. A theory (26, 27) proposed by Whitesides, Filioppo and Yagupsky et. al., deter- mining the stability of transition metal-carbon bonds stresses the importance of a low energy decomposition pathway being present, such as la-H elimination. Whitesides, Filippo and.Yagupsky et. al. ascribed the stabilizing effect of a ligand to its ability to block the coordi- nation site. Stabilization also results when the M-C-C-R dihedral angle is constrained to be far from the zero degree angle that seems to be optimal fer metal hydride elimination. Thermal decomposition of platinum metallocycles gave only those ,9-hydride elimination products (12) which had been obtained from acyclic analogues. Thermal decomposition of nickel(II) metallocycle in toluene probably occurs by several pathways judging from the analysis of the de- composition products (28). Those pathways (other thanB—hydride elimi- nation) which involve C-C bond cleavage give ethylene, cis nickel-carbon bond elimination gives cyclotutane. What factors give those pathways the opportunity to compete with B-hydride elimination is a questions which can only be answered by further investigation of the thermalchem- istry and photochemistry of nickel(II) complexes. 26 27 Recently Grubbs (29) found that the coordination number of a complex appears to be the major factor controling the mode of its de- composition. One other theory, originally proposed by Chatt and Show (30) and later modified by Yamamoto (31) , used the electronic of constituents to explain the stability of metal-alkyl bonds. In the case of 2 decomposed in the solid state is 275 KJ mole-1. In the presence of olefins, however, this parameter is reduced to ca. 65 NJ mole-1 and an 18-electron inter- mediates }_ can be isolated (31) (scheme 9). \ ,Et Ni Ea 275 KJ mole"1 z \Et 3 bipy + Ni + C4H10 O a 0 new \ INi’Et 6 . -1 ' 0" \Et Ea 5“ "1°19 > (biPY)Ni(Cl-Iz=Cl-Dt) «:C,,H10 18-electron intermediate (scheme 9) A correlation can be drawn between the electron deficiency of the olefin and the ease with which the metal-alkyl bond will cleave. As the electron deficiency of the olefin is increased, a higher rate for the metal-carbon bond cleavage is observed. There is an energy gap between the bonding (FR-M orbital and a vacant d orbital. Promotion of an electron from the bonding 6R-M orbital to a vacant d orbital would result in the split- ting of the R-M bond. Thus, if the energy gap were decreased, promotion 28 of the electron would become easier, and the cis-metal-alkyl bond would be more susceptible to cleavage. In this chapter strong coordinating olefins were used to investigate the effect of electronic factors on the mode of decomposition. Results and Discussion (1). Thermal decomposition: Samples of nickel(II) complex (solid) were allowed to decompose in a Schlenk tube connected to vacuum line. The temperature of the oil bath was raised slowly until the complex had totally decomposed and the gases produced at each temperature were measured. The product composi- tion was determined by go. The decomposition temperatures in Table 3 and 4 refer to the temperature at which the gases evolved most rapidly on heating in vacuum. Most of the complexes turned black at or above the decomposition temperature. The four types of products obtained from the decomposition of solid state metallocycle and III(see Table 3) were as follows: (a). carbon-carbon bond break: ethylene (b). hydride-abstration: butane (c). nickel-carbon cis elimination: cyclobutane (4).]3-hydride elimination: butenes Observing the product ethylene percentages in Table 3 we see that I had the highest (13.2%) and Ia had the lowest (0.7%) Percentage. d All the compounds had similar decomposition temperatures. Electronially tricyclohexyphosphine in more basic than triphenylphosphine. Stericly tricyclohexyphosphine is a bulky phosphine which easily forms a three 29 coordinate species by losing on phosphine. According to Whitesides, Filippo and Yagupsky et. a1. theory (26, 27) butenes should be the major product when the ligand is the tricyclohexyphosphine and indeed the highest percentage of butene's on If1 (87%). The free phosphine lost from the nickel may coordinate on another molecule of bisphosphine complex. The facts suggested that ethylene formation would be promoted by a more basic ligand or by a higher coordination species. Ic and 16' have electronically similar groups on phosphine. There is.a steric diff- erence, Ic has a chelated diphos ligand while Ie has two non-Chelated ligands. Examination of the products shows that both complexes gave about the same amount of ethylene. They also gave similar amounts of butene's, but Ic decomposed at a higher temperature than I e' The amounts of butane and cyclobutane fermed were quite difference for the two comp- ounds. The free phosphine methyl group in Ie could be a good hydride source leading to the formation of butane. The formation of cyclobutane from Ic could be promoted by the chelation. III gave typical thermal decomposition products (12) with no ethylene or cyclobutane observed. In this case nickel(II) complexes C-C bond cleavage and Ni-C cis elimination are due to the nickel(II) complex having cyclic structure. Electron donating ligands can promot C-C bond encourage Ni-C cis elimi- nation. Complexes having a greater tendency to form three coordinate species, also tend to give more butene products. Six membered ring metallocycles (see Table 4) undergo 61C? 610' bond cleavage and Ni-C cis elimination. IIa gives much more/s-hydride elimination product than Ia' This compatible with the first hypothesis. The M-C-C-H dihedral angle in the six membered ring_compound is much closer to the angle required forte-hydride elimination than it is in the o.on::-:::::: : o.on : o.ooH «AnarcvazmAamsv HHH rim oh man no o.nm r.m o.mwo.or Oazmfimsmév CH 9: fine 6.? : 9% mm 0.30.03 Oazmfimsozv oH r.oa r.HH a.ar : : «.ma c.00a mmuwzmflamsov oH ~23 0.3 5? ohm : Om 6.90.03 02 morass oH ~.~ 9m ofim . arm 06 0.: Prawn szAmummroE pH co m; as mgr a; so Pawn 8H Oazmflmsv nH capsmumno ocopsmumuo. 0:337.“ ogsnoaomo Egan ocoarafim ogpgomsoe unsomsoo Amvaoscoum .oemsm efiom ea HHH was roachooaaoxodc osoahfiosofleop mo soapfimomaooov dogmas. 65 Son.“ 850% ARvmpozoOHm .m canoe 31 ---- a.o --..- o.oe o.a n.e - n.a a.a a.ns a.o n.N o.mm 0.3: H.: m.wfi h.o N.o m.H «.0 n.o H.o checks; .2 cores pd. 0.9.6.0: oazmfimsv mHH Amvmeosoowa osdpmnodsoa vasomsoo ca moaohooaaoxOHs ocoahrvosspcom mo.soapflmomsoooo flushes» one .opepw vaaow scum dosnom ARVmpOSUon .3 magma 32 a five membered ring complex. II also gives (gt-C and (Si-Cr bond cleavage b products and a Ni-C cis elimination product but there is no suitable six: member ring complex for comparison. (2). Photodecomposition: Most of the complexes are not thermally stable so two sets of samples were prepared under identical conditions. One set of samples was irradiated with ultraviolet light (450 watt) while the other set was wrapped in aluminum foil and used as a control. The big change accompanying photolysis was an increase in C-C bond cleavage reactions (Table 5, 6, and 7). Five membered ring nickel(II) complexes either in solution (Table 5) or in the solid state (Table 6) gave more ethylene with photolysis. Six membered ring nickel(II) complexes in the solide state (Table 7) gave more dté’and dicrcleavage products with photolysis. This phenomenon is probably the result of a bonding electron being promoted to a C—C antibonding orbital by ultraviolet radiation (with more discussion in last chapter). (3). Solvent effects on the decomposition of nickel(II) metallocycles: Ia was decomposed in toluene, acetonitrile, or in toluene with added ligand (triphenylphosphine or tricyclohexylphosphine) (see Table 8). Examination of the product ratios, shows that the yield of ethylene increased as the solvent was Changed from toluene to acetonitrile to pyridine. Ethylene became the major product in pyridine and in toluene with added phosphine (Soheme 10), toluene/P¢3. toluene/PCYB. These re- sults probably reflect a change in coordination number. The coordinat- ing ability of the solvent increases as its basisity increases going 33 Table 5. Products(%) formed from the photolysis of Ia: Io, and If at 0'0 in solution state. Products(%) Compound Solvent Light ' 02m. 04118 + can“, Ia ¢-H UV 21 79 Ia ¢-H - 6 94 1a 95—an UV 11+ 86 Ia ¢—CH3 - 5 95 IC ¢-H UV 59 Lu 1c ¢-H - - 99.9 Ic 93-0113 UV 75. 5 24. 5 1C 95-0113 uv 31 69 I f 95-11 UV 31 69 If' ¢-H - 29 71 I f ¢-CH3 UV 19.5 81 . 34 Table 6. Products(%D formed from the photolysis of Ia, Ic, and Ie in solid state. Products(%) Compound Temp. Time (hr) Light C2H4 C4H8 + CLnflio Ia RT 25 UV #1.9 58.1 Ia RT 25 - 20.4 79.5 Ic RT 25 UV 37.4 61.9 IC RT 25 - 27.2 72.5 1e 5°C 8 UV 45.8 54.2 Ie 5‘0 8 - . 26.1 73.1 Table 7. Products(%) formed from the photolysis and thermolysis of Ila and 11b in solid state. Products(%) Compound Temp. Light CH4 C214» C3H6 C1.310 C5*‘10*“’5’*12 11a 5’0 UV 4.2 26.6 2.0 2.3 64.8 11a 110 0 - 0.1 6.5 0.1 2.7 90.3 ' 11b 5‘0 UV 14.7 40.0 8.0 17.4 19.5 Mb 125‘0 - 0.9 15.3 1.3 2.0 81.0 35 from t0p to bottom in Table 8. ¢3P\Ni:j .1311—9 (¢3P)3N13 —-> = ¢3P/ . l D (Scheme 10) As the concentration of the higher coordination species is increased, the yield of ethylene increases. When IC chelated diphos ligand was decomposed in toluene, pyridine or toluene with added triphenylphosphine or tri-nébutylphosphine, the mode of decomposition (see Table 9) was quite different from that observed with Ia' Cyclobutane was the major product when the decomposition was carried in toluene, pyridine, or toluene with added triphenylphosphine. In an attempt to alter or other- wise affect the mode of decomposition, higher concentrations of the more basic tri-n-butylphosphine (n—Bu3P:Ic = 200 a 1) were used. This changed the major product from cyclobutane to 1-butene. One possible explanat- ion is that the chelated complex Ic does not easily form a higher coor- dination species. As larger excesses of the more basic tri-nébutyl- phosphine were added, the reaction went, tri-n-butylphogphine acting as a base can do an intermolecular hydride abstraction on the hydro carbon ring (scheme 11). The pathway is still not fully understood however. The probable reason why Ic’ when decomposed in toluene with a large excess of triphenylphosphine, does not give 1-butene as the major product is that triphenylphosphine is not as basic as tri-n-butylphosphine. 36 Table 8. Products(%) formed from solvent effects on the decomposition O C of Ia at 9 C11 C. Products(%) Solvent - ‘ Ethylene Butane Cyclobutane Butene Toluene 4.6 5.4 68.8 21.1 Acetonitrile 16.3 - 83.7 - Pyridine 71.8 6.7 21.5 ‘ - Tcluenc/¢3P * 71.7 2.9 25.3 - Toluene/0132@ 86.5 1.7 11.8 - *¢3P : Ia = 20 , 1 (mole) @CYBP : Ia = 20 : 1 (mole) 37 H a 8m .... H . *mmsmé coca. 86.3. N366 ox {mm 6.0 o. 36.1: K966828369 1.--} 6.: 1!.-- 0.8 9m 6.m 698.1: mmEcscfice m; 6. H N.N 6.66 md hm 6.36.3 8638. 64 . Nun 6.6 lama m5 ma 8.63 as SSE 0.6 Hi. N. 3“ Que fig 9N 3.03 am 98369 62345 Imumwo campsmsmlv macadamia mGSSAOHohU 633m ocmahfim mHSmHmASme pso>Hom Amvflcsccam .OH mo 2033098006 65 so 3.06996 .5998 son,“ 6050M Amvmposvowm .m 6.3.69 AmHosv fl » on u 0H u *mmsmlc .............i 0. E ....... u 0.0 m .2 N66 *mmsm.s\csc§ce 0H 0.6 6.: 0.2 6.00. . hos 6.00. 08020 0H 0.0“ 6.0 6.00 6.6: «.0 0.6 00038. 0H 0.0 0.6 6.6“ 6.00 6.9 0.6H 00825 6H 04 04 0.0a 0.6.6. 0;. 6.0 0:880. 6H campsmnmumfio osmysmumlp 623sz 08359328 633m osmgfim vam>Hom 653500 $860083 68.0.0on53 589 0.6 0H 626 AH mo 830890806 05 so 306.30 8853 sow.“ 605.8.“ @1856on .3 63.09 0.0 6.0 0.60 0.00 0.00 6.00 0.6 600602 6.0 0.0 0.00 0.00 0.60 0.66 0.6 00606.0 0.0 6.0 0.60 0.00 6.00 6.06 0.6 00000000 0.0 6.0 0.60 6.00 0.60 0.60 0.0 000000000000 mamascmnmumfio msmvcwmumlp 600300000..." 0003000003060 6530.800 00036560 600650: pco>aom 0008888 .6: mo 00000006098006 65 no 300.0660 000800.006 00on 0005.08 ARvmvosdon .NH 0369 39 AGHOSV H « ¢ n... %H u mmzmlct. 0.6 6.0 0.06 0.0 0.06 0.06 0.0 *06smws\csms0ca 6.0 0.6 0.66 0.0 6.00 0.00 0.6 00000000000< 0.0 0.6 . 0.06 0.0 0.00 6.00 0.6 00000000 mampzmumumwo campsmumnp 60060030010 00030093060 608.005 60.06?an 6.0368983 0.00658 ARvmvosvon .MH mo 083069008066 05 so 60006.06 £8506 690.0 6050.0 Amvflozvohm .00 63.6.0. (P:N1Q2{ ____> (3:165) H-PBuB'n —->W PBu3-n (Scheme 11) The modes of decomposition of I and Ie in pyridine and toluene b are shown in Table 10. Ethylene became the major product when Ie was decomposed in toluene with added tri-n-butylphosphine. This behavier is quite different from that of the IC but it is similar to that of Ia' The decomposition of the nickel(II) metallocycle IIa (six member ring) was also investigated (see Table 12). The formation of ethylene incre- ased as the solvent was changed from acetonitrile to pyridine, tri-n- butylphosphine and finally triethylamine. When IIa was decomposed in triethylamine, strongly coordinating solvent, ethylene become the-major product. (4). Olefin effects on the decomposition of nickel(II) metallocycles: According to the second theory, complexation of an electron defi- cient olefin (A) to the nickel may induce charge transfer from the bond; ing M-C orbital to a vacant metal d orbital and increase cyclobutane formation. 41 ¢3P\ g . ¢3P\ 31 961°” 953? / . Complexation of an electron rich olefin (B) to the metal should induce electron flow in the other direction, favoring ethylene formation, similar to that observed in photodecomposition. The results are shown in Table 13 and 14. Tetracycanoethylene, acrylnitrile, and ethylene tri- chloride were used as electron deficient olefins. Cis-Z-pentene and cyclohexene were used as electron rich olefins. 0.0 6.0 6.60 6.60 0.00 6.00 0.6 00000000000 6081050 0.0 0.0 6.60 6.60 0.60 0.06 0.6 -000000000.0.0.0, n6 0.. 0 +0.00 00.8 m.NN m.mm 9m 0030.00.06.84. msmpcomwmnmfio .onmpcmmrmup msmpcmmTH mumpsmmoaohu mumpsmm mamahnvm msmspmz 0206000 0500000900 .mHH mo sofipfimoA50006 650.:0 6906660 0206000 sonw 605006 Afivmpozdonm .00 magma 42 0.0 6.6 0.0 0.0 0.0 6.66 0000000101000 mumsaop o.mm m.00 m.n o.wm n.m w.6 \wsoahspmosmomngma 0cmpsmlmnmflo mumpsmVNIp mqmpsmwfi msmwsnoaoho mumpsm osmahnpm 0000000 60300060000 .00590000500 SOON 06 0H mo c0000momsooou map no 6900600 ocflmmao 200% 665006 A&vmposdon .M0 60969 43 Experimental All experiments were carried out under a deoxygenated argon atmos- phere, or in a vacuum. Gasses evolved during thermolysis were analyzed with a mass spectrometer. The composition of the gases evolved during thermolysis or photolysis of a complex was determined on a gas chroma- tograph Varian 1400 or a Perkin Elmer 900 equipped with flame ionization detector. The latter was also hooked up with an Autolab System I inte- grator. An 8 ft stainless column (1/8 in 0.1).) containing ether 7% parafin wax on.Al or 120/150 mesh Duropak was used. The column temp- 203 erature were 80 and 75°C, respectively. The carrier gas (He) flow was 30 ml/min. Thermal decomposition: A 15 ml Schlenk tube containing 0.1-0.4 g of complex was connected to avacuum line equipped with a mercury manometer, and was evacuated (Figure 2). The Schlenk tube was placed in a oil bath with thermometer. The temperature of the oil bath was raised very slowly (1'C/min), and the gas evolved at each temperature was recorded. 4 .ii;_.vacuum outlet cold finger manometer V) Figure 2: Vacuum line system for thermolysis reaction Photodecomposition: A 10 ml quartz tube containing 0.1 g of complex in the solid state or a 50 ml'pyrex tube containing 0.1 g of complex in 10 ml of solvent was irradiated under argon using a 450 watt super high-pressure mercury lamp with overal UV wavelenth. The sample tubes were positioned at the same distance from the lamp in a big ice box or water bath. The evolved gas (on top) was analyzed by gc. Solvent and olefine effect on the decomposition of nickel(II) metallo- 2223a: . Solvents were dried by standard methods and stored under dry oxygen free conditions. The complexes were prepared as previously des- cribed. All product gases which had dissolved in the solvent were col- lected by repeated trap-to-trap distillation as follows (see Figure 3). =fl: . : , vacuum outlet Decomposition system Liquid nitrogen Figure 3: Vacuum line system for trap to trap method The gas and 50 to 30%lof the solvent from the decomposition system in tube A were vacuum distilled into tube B, then a new empty tube A was connected to the joint after tube A was removed. All of the gas in 45 tube B (and 10% of the solvent) was distilled into tube A', and tube B was then dried in vacuum. Finally all of the gas in tube A' was disti— lled into tube B without solvent. Gas samples for go analysis were taken from tube B . CHAPTER 3 ETHYLENE AND CYCLOPENTANE FROM NICKEL(II) METALLOCYCLES Introduction In the first chapter the only well characterized metallocyclo- pentane complexes considered were made from more reactive olefin such as norbornadiene. The formation of unsubstituted metallocyclopentane from ethylene has been observed in two cases (14, 33). In one case ethylene was introduced into a system containing CPZTiN=NTiCP2, which was at a temperature below -30°C. This produced a reaction mixture with chemical properties which suggested the presence of metallocy010pentane (see scheme 12). CPZTiN=NTiP2flaiC§e-> CPZTig c: > & BrJ_1€&r (Scheme 12) Recently, 1977, Schrock gave strong evidence for the formation of a tantalum metallocyclopentane (33) by reaction of m with ethylene ‘V (eq. 12). The product 3 was isolated in 9% yield and characterized by 1H nmr, and chemical reactions 47 Cl (:1 = CH CP-Té-H CHZ CH? > Cl-T'aa + i (eq. 12) 5c 0 n . in/ 4/ This chapter will focus on the reverse reaction (formation of ethylene from unsubstituted metallocycles) and some oxidative addition reactions of metallocycles. Results and Discussion (1). 31F nmr of nickel(II) metallocycles: The 13C nmr of carbons in subject metallocycles could not be obtanied because of low solubility. The 1H nmr of Ic and Id were domi- nated by the resonance of the ligand. The 31F nmr chemical shift of Ia was observed at -42.5 ppm and a new complex was observed by 31P nmr after excess triphenylphosphine had been added to Ia' This new complex (chemical shift at -27.5 ppm) was isolated (29) and recystalized under low temperature as golden brown cystals. The analysis of this complex showed P/Ni = 2.98 (theor =3.0) and hydrolysis with sulfuric acid (-20°C) produced butane (98%). This indicates that the new species is still a metallocycle (eq. 13). (¢3P)2Ni:j + P¢3 (excess) ——'> (¢3P)3Ni:) (eq. 13) 31 When the solid was redissolved, it again showed a P signal at -27.5 ppm. The 31F spectrum of Ia in the presence of a 14.1 mole excess #8 of triphenylphosphine was temperature dependent. As the temperature was raised (Figure 4), the -27.5 ppm peak averaged with the free triphenyl- phosphine peak while the signal for Ia remained sharp. The 31F spectra of IIa showed similar behaviour. The chemical shift of IIa was observed at -33.3 ppm. Addition of one mole excess of triphenylphosphine (Figure 5) gave a higher coordination complex. The new peak at -23.5 ppm was assigned to the higher coordinated metallocycle on the basis of the observation of Ia' When a toluene solution of IIa was treated with an excess of triphenylphosphine at -10°C and then stored at -20‘C for a week golden brown cystals formed which were then isolated. The 31P signal of this complex was observed at the same position (-23.5 ppm) at -90‘C. As the temperature was raised, the —23.5 ppm peak equlibrated with the triphenylphosphine peak, presumably by a dissociat- ion mechanism (scheme 13). The rate of equlibration depended on the amount of triphenylphosphine added (Figure 6 and 7) with a higher conc- entration of triphenylphosphine giving a slower rate of exchange. (¢3P)3NO + ¢3P* 2:2 (¢3P)2NO + 953? + ¢3x>* Tl (¢3P)2(¢3;)Nio : [(¢3P)2Ni:> + ¢3P*] + ¢3P (Scheme 13) The 31P spectrum of Ic at ~90°C consisted of a single peak at -#5.0 ppm, Ic (Figure 8) indicating that neither exchange nor formation 49 of detectable amounts of higher coordination number complexes and de- composition products are only formed on heating to 80'C. The results are consistent with data in the previous chapter and serve to confirm that four coordinated metallocycles (P2 species) give cyclobutane while five coordinate metallocycles (P3 species) give ethylene. 10°C Figure 4. Temperature dependant 31F nmr spectrum of Ia in toluene with added,P¢3/ Ia = 1a.1 (mole). 51 (A) (B) y 1 Ila (C) t Figure 5. 31? nmr spectrum of II in toluene with (A). added,P¢2/ 11a = 4.1 (B). added,P¢3/ II8L = 1 (c). no added P5253 at - 'c 52 -27'c -72°c -81‘0 P¢3 -91.C Figure 6. Temperature dependant 31P nmr spectrum of IIa in toluene with added,P¢3 / IIa = 1.2 53 W '75 .C .89'0 Figure 7 Temperature de - . pendant 31 nmr ‘lrum in toluene with added P¢P / IISPECL, Of IIa 3 a "' 00 n-Bu3P (C) -B I. n “3P (A) \ -90'c Figure 8. 31P nmr spectrum of IC in toluene with (A). no added PBu3-n O (B). added PBu3 (c). added PBuB-n, heat up to 80 c 55 (2). Phosphine effects on the mode of decomposition of nickel(II) metal- locycles: W ' Since the coordination numbers of the complexes were a function of the structure of the phosphine, the ratios of the decomposition pro- ducts as a function of the P/Ni ratio were determined. Plots of the resulting gas composition as a function of added phosphine are shown iJIFigure 9, 10, 11, 12, and 13. From the molecular weights in solution, it can be determined that Ia and Ic are present as P species whereas I dissociated to a P1 2 d species. Cyclobutane is the predominate product for both P2 species at P/Ni = o. The dissociated Id produced 1-butene as the major product. As trialkyl phosphine is added to the trichlohexyphosphine complex the amount of cyclobutane increases to a maximum at P/Ni of 4. At this point and at higher P/Ni ratios the decomposition ratios are very similar to those resulting from triphenylphosphine complex at P/Ni = O and larger value and only for Ic at P/Ni = o. The 31? studies showed that I 0 did not form higher coordination number complexes when tri-n-butyphosphine was added. Consequently, this requires very high concentration of phos- phine appears to induce decomposition to linear butenes by a mechanism not fully understood. The decomposition curve of If shows the same tendencies as des- cribed above, but except data (P/Ni ratios)is lacking. The results combined with nmr data and molecular weight inform- ation are most consistent with the following scheme 14. P P PNi(C,+H8) g 7 PZNi(C4H8) ‘a 7‘ PBNi(C4H8) (scheme 14) J, t 1-butene cyclobutane ethylene 56 There are two modes of carbon-carbon fragmentation possible for 11a, 0 a ' 0 C-C bond cleavage, and C-Crbond cleavage, so the ethylene observed here may not be exclusively from a simple fragmentation. Futher work, such as labeling studies, is required for any conclusion can be drawn. 6H mo woos 000330900006 o5 no mam 0.30.6.6 mo poommm .m onzwfim Ao0osv 00 \ 600 .00000 8 M0 3 6 o I21 60) Susanna e? .I. 9:11 lllllllll D llllllllllll 000.305.0350 ... I .1 b osodnfifi . OOH (g retoux) notitsodmoo sea 6H Mo 0690 an mpfimodsoooo 05. so mmsmu: 6066.6 mo woommm .00 0.99.0.6 00095 00 \ 0-06.0 .8000 om 60 00 6 o b ¥ lle.’.,. p 00036.0.1/ X/. I 4----ummsmomm--nf: , x . x. \ \ x. \ w a /. on x / 0 D . 0. z. \ x. x x. z. z. 0. AVI: onoahnpm / woo.“ (% .retom) uotitsodmoo sea 00 .3 00000800800 05 so 0606.0 00000 no 30.0.06 .00 00360.0 Aoaoev oH \ mnnmu: .8306 .6010 0.00 06 0.6 0.0 0.0 I11 000030.06 11 6... III lllll .I Illllllllllll Illa: III ”gapOHOhU II I I p, .\. / X F .x. III .\ o 0A,! \ x M. /h .\ / I / / 0‘ o \.\ ...l. |.\ \ .A . urn. .l .l .| i. . Mmmmmua. m . 0:305 (% are-tom) uotqtsodu'too sea fi 000 .00 6o 80000200800 05. 0c 6606.0 00000 6o 0006.06 .60 060606 A0095 60 \ 6606.0 60000 06 . £0 + 0.6 . ow 0.0 .. o 9.030.000.0060 I. III... III a l Isl: 11013,-9 ’ID"/’ In 000 _ .om \ 1 0m 100 6080001500 .000 (95 .I’EIOIII) notipsodmoo sea 61 Figure 13. Effect of added P953 on the decomposition of 11a 000000 000 \ 66.0 00000 oco 00050 / 63 Table 15. Products(%) formed from triphenylphosphine effects on the decomposition of Ia at 9° 031°C in toluene. Products(%) 913 P, added/Ni . M0131. ratio Ethylene Butane Cyclobutane Butene's 2.0 17.2 5.3 67.2 10.3 5.0 44.3 0 53.1 2.6 8.0 64.7 0 31+.Z 1.1 10.0 70.5 0.9 28.3 0.3 15.0 72.2 1.7 26.1 0 20-0 71.7 2.9 25.3 0 Table 16. Products(%) formed from tri-n—butylphosphine effects on the decomposition of lo at 14 Ct1 C in toluene Products(%) n-Bu3P,added/Ni Molar ratio Ethylene Cyclobutane Butene v s 0 2.7 89.5 7.2 8 1 . 6 81 . 6 15. 9 20 2.6 60.9 32.7 LK) 1 . 9 38 . 6 53 - 5 60 1 . 2 15. 2 81 . 8 160 0 . 6 7 . 4 91 . 2 200 0.6 7.0 92.1 61+ Table 17. Products(%0 formed from tri-n-butylphosphine effects on the decomposition of If at —50°C in toluene. *V—v’ ~——f Products(%) n-Bu P,a.dded/Ni Mo ar ratio Ethylene Cyclobutane 0 85.0 15.0 5 91-5 8.5 10 94.0 . 5.9 15 96.0 4.0 20 97.0 2.9 25 98.0 2.0 60 100.0 0 65 00.0H 00.0 00.0m 00.0 00.00 00.: 00.H 00.0H “0.0 00.00 00.: 0:.fi: 00.: 00.H 00.0w 00.: mm.mm 00.0 00.mm m0.: H0.0 N0.mm 00.0 00.00 00.0 00.0w 00.0 00.0 00.00 00.0 00.00 00.0 00.0H 00.0 00.0 “0.00 00.0 00.00 00.0 0a.0 00.0 00.0 Hm.0m 00.0 00.00 0m.0 00.0 H0.m fi0.0 00.00 00.0 00.0: H0.0 H0.0 00.: 0 m.mcmpsmm occuaomoaoho mqmpnom onupcmmnomH mcoaanpm mnmsvmz Ofipmh HdHoz 0500000 .000 ARVmpozvon .ososaop c0 ooofi #0 0HH mo coflwfimogsooou map so mpoowmo osfinmmocmazsonmflhp somw 065909 onmpOSUOHm .mH mapme 66 (3). Cyclopentane formation Organonickel compounds are widely used as reagents and catalysts in organic synthesis (34). Takahashi, et. al. (35) recently reported that a low valent nickel-bipyridy1 complex is a powerful reagent for the cyclo coupling of dihalogenoalkanes,as shown in scheme 15. They suggested that the cyclo coupling reaction might proceed through nickel cycloalkane intermediates. X-(CH2)1+-X Ni-biPY 1;. Ni'biPY (Scheme 15) 2 . (111213302 C] O Diphos nickel cyclopentane (IC) has been isolated : upon treament with oxygen and activated olefin it liberats cyclobutane more than 91%. Table 19. Cyclobutane formed from oxygen and olefin effected decomposition of Ic' Products(%) Products(%) Reagent Temp. Solvent Ethylene Butane Cyclobutane Butene’s Oxygen 26'0 — 1.1 0.4 91.2 7.3 Oxygen 15.0 Toluene 0.4 0.1 95.0 3.2 Oxygen 15°C Benzene 0.5 0.3 96.0 2.3 figfiéni- 15 c - 0.3 0.6 91.0 6.7 When dibromomethane was added to an ether solution of IC at 0°C and allowed to reat at room temperature for twenty four hours,cy010pentane 67 was obtained in 70% yield (Scheme 16). P . D 6— \Ni:) RTCH—iri > 0 (Scheme 16) , r In the case Of Ia' the yield of cyclopentane about 20% was relatively low owing to the low stability of Ia' This method may have a poten- tial application in the synthesis of cyclopentane derivatives by using dibromomethane derivatives as reagents . 68 Exnerimental The 31F spectrum was run by multinucleur nmr spectrometer DA-60 with D20 external lock, and phospheric acid as external reference. The sample was placed in a sealed vacuumed nmr tube transfered at low temp- erature and under an inert atmosphere. . Decomposition reactions were carried out on a vacuum line as illus- trated in the previous chapter. The decomposition gases were collected by trap to trap distillation and the components were analyzed by go. They are shown in Table 16, 17, 18 and 19. Cyclopentane synthesis: Dibromomethane was added at 0°C under argon to an ether solution of metallocycle (type 1) with 1 : 1 ratio of dibromo- methane to Ia or 10' After addition the reaction was allowed to react at room temperature for twenty four hours. The gas volume and compo- nents were analyzed by gc. CHAPTER 4 APPLICATION OF EXTENDED HfiCKEL CALCULATION TO THE REACTION OF NICKEL(II) METALLOCYCLES Introduction The Hfickel theory has been widely OXploited in chemistry. Woodward and Hoffmann have developed the concept of symmetry-based selec- tion rules for chemical reactions. They have applied their approach to electrocyclic reactions (36), sigmatropic reactions (37, 38), and con- certed cycloaddition reaction (39). Hoffmann also applied these calcu- lations to the bonding capability of transition metal carbonyl fragments and the structure and chemistry participation of transition metal in concerted cycloaddition reactions using the WoodwardéHoffmann scheme. Their methods is to draw a molecular orbital energy.level diagram for the reactants and then for the products and finally to correlate the individual final molecular orbitals using symmetry restrictions imposed by an assumed geometry for the transition state. If the correlation connects a filled bonding orbital in the initial state with an empty anti- bonding orbital in the final state, the reaction is then described as thermally disallowed but photOChemically allowed. As pointed out by Eaton (42), a more elegant approach to this is to construct a state correlation digram. In this chapter the extended Hfickel calculation was used and state correlation digram was constructed for the isomerization between square planar (H3£)2Ni(04H8) and tetrahedral (H3P)2Ni(04H8). 69 70 Calculation A basis set of valence atomic orbital for Ni consisting of 3d, 43, and 4p, single Slater-type orbitals were used for the 4s and 4p wave function, while the 3d wave function was taken as a contracted linear combination of two Slater-type wave function. The diagonal H matrix elements, H were as shown in Table 20. ii! The geometrical approach used here is to perform the calculation at a resonable orientations. The bond lengths used in the calculation were obtained from currently available X-ray data (43). The coordinates are listed in Table 21 and Table 22. Table 20. Parameter used in the calculation (44) (45). Orbital Orbital exponent H11, ev 1 2 Ni 3d 5.75 2.00 ' -13.2 4s 1.50 -10.7 4p 0.86 - 6.3 P 38 1.18 -19.00 3p 1.08 -10.04 C 25 1.63 -19.5 2p 0.88 - 9.9 H is 1.00 -13.60 71 H i H .H I _ I a C 14 .2 o )pY’ . § 4- ‘~w '15 ‘918 i \ H H Vx -1- ,Table 21. Coordinates of 1 (square planar metallocycle). (HBP)2Ni::) Coordinate (A0) Coordinate Atom, # X Y Z Atom, # X Y Z Ni 1 O 0 0 H 12 2.48 1.01 0.90 C 2 -1.48 1.48 0 ' H 13 2.48 1.01 -0.90 C 3 -0.81 2.90 0 P 14 -1.61 0 0 C 4 0.81 2.90 O H 15 -1.94 0 1.34 C 5 1.48 1.48 O H 16 -2.76 1.16 -1.34 H 6 -2.48 1.01 0.90 H 17 -1.11 -1.16 -1.34 H 7 -2.48 1.01 -0.90 P 18 1.61 0 0 H 8 -1.34 3.86 0.90 H 19 1.11 -2.74 1.34 H 9 —1.34 3.86 0.90 H 20 1.94 -2.74 -1.34 H 10 1.34 3.86 0.90 H 21 2.76 -2.74 1.34 H 11 1.34 3.86 -0.90 72 -2- Table 22. Coordinates of 2., (tetrahedral metallocycle). (H3P)2NiO Coordinate (A0) Coordinate Atom, # X Y Z Atom, # ‘7 Y '2— Ni 1 0 O O H 12 1.29 0.90 -2.56 C 2 1.48 O 1.40 H 13 1.29 -0.90 -2.56 C 3 2.90 O 0.80 P 14 -1.30 1.85 O C 4 2.90 0 -1.80 H 15 -0.83 3.19 1.16 C 5 1.48 O -1.48 H 16 -O.83 3.19 -1.16 H 6 1.29 0.90 2.56 H 17 -2.67 1.46 0 H 7 1.29 -0.90 2.56 P 18 -1.30 -1.85 ‘ O H 8 3.86 0.90 1.34 H .19 -2.67 -1.46 0 H 9 3.86 -0.90 1.34 H 20 -O.83 -3.19 1.16 H 10 3.86 0.90 -1.34 H 21 -0.83 -3.19 -1.16 H 11 3.86 -0.90 -1.34 73 Table 23. Molecular orbital configurations and the associated electronic states of the square planar nickel metallocycle. Molecular orbital configuration State E (ev) 2 2 1 1 1.3 ”b2 9a1 81:1 9131 A1 4.12585 46% 9a§ 86% 961 1'331 4.31319 2 413% 9a? 8b1 ' 913% 1'3A2 5.00864 41% 9a11a 8b1 913% 1 ' 3A2 5. 58234 46% 9ai 8b";1 913% 1 '3132 5.76969 4bi 9a? 8b? 96% 1’3A1 6.46514 48% 9a? 8b% 913% 1 ' 3A2 6.93198 41); 9a1 8b? 9b: 1 ' 2 7.11933 46% 9a? 86% 96% 1'3A1 7.81478 74 Table 24. Molecular orbital configuration and associated electronic states of the tetrahedral nickel metallocycle. Molecular orbital configuration State E (eV) 4a: 6b: 10a} 6b: 1'382 0.47849 4a§ 6bi 10a? 68% 1'3A2 . 0.54897 4a% 6b? 10a? 6b§ 1'381 1.49688 4a§ 6b? loai 6b% . 1'382 5.04815 4a§ 6bi 10a? 6b: 1'3A2 5.11863 4a% 66% ioaf 66% 1,331 6.06654 4a§ 68f 10a§ 681 1’381 5.20841 4a§ 6bi 10a? 6b} 1'3A1 5.27889 75 Results and Discussion The isomerization of a metallocycle (type I) to a bis-olefin- metal complex reduces the oxidation state but retains the same coordi- nation .mber of the metal. From previous chapter the triphosphine metal compexes are coordinately saturated the conversion to a lower oxidation state would require the loss of a ligand (scheme 18). The lower coordination mber intermediate (V) required for the phosphine exchange in the nmr is an appealing precusor for the formation of ethyl- ene. This suggests that the ethylene and cyclobutane are produced from isomeric coordinate complexes. P2103 -—-> [:1- sq. pl. UV slow i ,>\ PBS/O \ ~——> [V] ———> PZNiW -—-> = P (Scheme 18) The isomerization of a square planer complex to the isomeric complex (V) could take place through a higher coordination complex (IV) as an intermediate. The isomerization could also be induced with UV irradia- tion. The conclusion is true if direct isomerization take place very slowly or the direct isomerization is not allowed. The possible choice of V is tetrahedral. The three highest occupied molecular orbitals and the three lowest 76 unoccupied molecular orbitals were used to calculate nine low lying excited states as shown in Table 23 and 24. The state correlation dia- gam was constructed with singlet triplet spliting as 1 ev. As shown in Figure 14, the ground state of the square planar metallocycle correlates with the ground state of the tetrahedral metallocycle by symmetry allow- ence and the same way for excited states. The ground state of tetra- hedral metallocycle is a triplet. These states correlate with square planar excited states. In the square planar metallocycle state diagram there is a large gap between the ground state and excited states. This would indicate that the isomerization of square planar metallocycle to tetrahedral metallocycle will be slow and the isomerization of the ground state of the square planar metallocycle to triplet ground state of tetra- hedral metallocycle will not be thermally allowed but photOChemically allowed. This calaulation result is compatiable with the experimental results and suggects that the higher coordination species fermed are un- stable in the square planar configuration and isomerize to the tetrahe- dral configuration (Scheme 19). Photo irradiation accomplishes the same transformation by exciting the square planar ground state to the excited states thereby making the transformation of square planar metallocycle to tetrahedral metallocycle allowed. P\ P 20 W > to P p’ X, P P p :NiO I! I P P (Scheme 19) 77 Tetrahedral metallocycle Square planar metallocycle 04A] ' 103A1 -763.2 ev 1A) Figure 14. State correlation digram of square planar metallocycle and tetrahedral metallocycle LIST OF REFERENCE 13. 14. 15. 16. 17. 18. 19. 20. 21. LIST OF REFERENCES Binger, P., Angew Chem., Int. Ed. Engl., 11, 309 (1972). Hall, H., C. Smith, and D. Plorde, J. Org. Chem., 38, 2084 (1973). Grubbs, R., and T. K. Brunck, JACS 94, 2538 (1972). Grubbs, R., P. Burk, and D. Carr, JACS 97, 3265 (1975). Green, M. L. H., "Organometallic Comp's" Methuen, Londen (1968). Adv. Organometallic Chem., 8, 29 (1970). Halpern, J., P. Eaton, and L. Cassar, JACS, 91, 2405 (1969). Osborn, J., et. al., JACS, 95. 597 (1973). Blackborow, J. R., R. H. Grubbs, and A. Miyashita (unpublished data). Grubbs, R., H. Eick, and C. Biefield, Inorg. Chem., 12, 2166 (1973). Whitesides, C., J. White, and J. McDermott, JACS, 95, 4451 (1973). lEcherglott, J. X., J. F. White, and G. M. Whitesides, JACS, 98, 6521 1976 . ‘ Whitesides, 0., and J. McDermott, JACS, 96, 947 (1974). McDermott, J., J. White, and C. Whitesides, JACS, 98, 6529 (1976). Diversi, P., G. Ingrosso, and A. Lucherini, J. C. S. Chem. Comm., 52 (1977). Tschugaeff, L., Ber. 38, 2520 (1905). Chapman, A. 0., Trans. Chem. Soc. III, 213 (1917). Product of BASF corporation. 7. 5% parafin wax, Dur0pack (n-octane/parasil C). Venanzi, L., J. Chen. Soc., 719 (1958). West, H., and E. C. Rochow, Naturwissen Schaften, 40, 142 (1953). 78 22. 23. 24. 25. 26. 27. 28. 29. 31. 33. 35- 36. 37. 38. 39. 41. 42 43. 79 Belcher, H., and A. Nutten, "Quantitative Inorganic Analysis", AMC MacDonald, Butterworth and Co., Ltd., London, p. 131 (1970). Jamborski, et. al., J. O. M. Chem., 27, 619 (1970). Pignolet, L. H., W. D. Horrocks, and R. H. Holm, JACS, 92, 1855 (1970). Methyldiphenylphosphine as free phosphine from ligand exchange with acetoacetate and the analysis using glc (5% SE-30 colume at 180 C and dodecane as internal standard). Yagupsky, 0., C. K. Brown, and G. Wilkinson, J. C. s., A 1392 (1970). Whitesides, G. M., J. S. Filippo, Jr., E. R. Stedronsky, and C. P. Casey, JACS, 91, 6542 (1969); 92, 1426 (1970). Burk, P. L., Ph. D. thesis, Michigan State University, East Lansing, Michigan, 1975. Grubbs, R. H., A. Miyashita, M. I. Liu, and P. L. Burk, JACS, 99. 3863 (1977). Chatt, J. and B. L. Show, (a) J. C. S., 705 (1959),(b) ibid, 1718 (1960). Yamamoto, T., A. Yamamoto, and S, Ikeda, JACS 93, 3350 (1971). Mclain, s. J., C. D. Wood, and R. R. Schrock, JACS, 99, 3519 (1977). Semmelhack, Cf. M. F., "Organic Reaction", Wiley, New York, 1972 Vol. 19, 115 Takahashi, S.,Y. Suzuki, and N. Hasihara, Chem. Letter, 1363 (1974). Woodward, R. B., and R. Hoffmann, JACS 87, 395 (1965). Woodward, R. B., and R. Hoffmann, JACS 87, 2511 (1965). Woodward, R. B., and R. Hoffmann, JACS 87, 4389 (1965). (a). Woodward, R. B. and R. Hofmann, JACS 87, 2046 (1965). (b). Woodward, R. B., and R. Hofmann, JACS 87, 4388 (1965). Mihai Elian and R. Hofmann, Inorg. Chem. Vol.14, No.5, 1058 (1975). Mango, F. D., and J. H. Schachtschneider JAC 89, 2484 (1967). Eaton, D. R. JACS 90, 4272 (1968). Biefild, C. G., H. A. Eick, and R. G. Grubbs, Inorg. Chem. Vol.12, No. 9, (1973) ()0 1+4, Rosch, N,, and R. Hoffmann, Inorg. Chem. Vol.13, No. 11, (1971*) 45. Nelson, J. H., et. al. JACS 91, 7005 (1969). IIIIIIIIIIIIII u\“TAIWANfittiflfifixiuflnmigfiimx 5 686