UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Gallazanes and related compounds Penland, Allen David 1971

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1972_A6_7 P45.pdf [ 3.75MB ]
JSON: 1.0061858.json
JSON-LD: 1.0061858+ld.json
RDF/XML (Pretty): 1.0061858.xml
RDF/JSON: 1.0061858+rdf.json
Turtle: 1.0061858+rdf-turtle.txt
N-Triples: 1.0061858+rdf-ntriples.txt
Original Record: 1.0061858 +original-record.json
Full Text

Full Text

GALLAZANE3 Aril) RELATED COK?OUHi)S ALLEN DA^II) PENIAT'D B.Sc. (Hons.) University of British Columbia. 1969 A THESIS SUBMITTED IN PARTIAL FULFILi/EEI?! OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE In The Department of Chemistry V/e accept this thesis as conforming to the required standard The university of British Columbia July 1971 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver 8, Canada ABSTRACT This work involved preparation of cyclical dimeric or trimeric gallazanes of general formula: (RNHGaHg) where n= 2 or 3 and R = Et, Pr", Pr1, Bu", EuV s t Bu , or Bu . The effect of larger R group on ring size ( n value ) was deter mined. Some, deuterated analogues of these compounds were also prepared. These were (EtNHGaD ) , (BuSffiC-al)^, and (PrHffiGaP )2. Attempted preparation of j#NHGaH2 resulted in isolation of Ji^NH.GaH^.Me^. Reactions were undertaken with jfeH. GaH2. NMe.j and it partially deuterated analogue ^KITGaD9.IIf-Ie^, and shown to involve proton transfer through a '+-centre transition state. Additional work on the effects of R group on the nitrogen within the gallazanes involved preparation of dimeric gallazanes of general formula ((CE2)y.lT.GaH2) where x = 2,3,k or 5. Additional work on double ring systems involved preparation of analogous alazanes of general formula ((dl )xN.AlH2)n where x = 2,3,^,5 and n = 2 or 3. Similar borazanes were likewise prepared and were of general formula: ((CH2)xlI.BH2)n where x = 2,3,'+, 5 and n = 2 or 3. Adducts of general formula: ' (CH^NH.EMe^ where E = B, Al, Ga, In, were also prepared. Upon pyrolysis these adducts yield methane plus materials of the general formula: ((CH )2N.EMe2)-j where E = Al,C-a, In. Characterization of these materials as well as gaseous reaction products was accomplished by infrared spectroscopy. Additional data was obtained by 60MHz and 100MHz ' H nmr as well as mass spectrometry. Molecular weights were determined cryoscopically in benzene and analyses for galluim, aluminum or hydrolysable/ hydrogen carried out by standard means. - iii -Table of Contents: '• Page I. Title Page i II. Abstract 11 III. Table of Contents iii IV. List of Tables v V. List of Figures vi VI. Acknowledgement ix VII. Introduction 1 VIII. Experimental 6 A, Experimental Techniques 6 (a) Desiccation ... 6 (b) Reaction-Filtration Apparatus 10 (c) Molecular Weight 1(d) Spectroscopy . 15 (e) Elemental Analysis 16 B. Preparative 17 (a) Preparation of Gallium Trichloride ............ 17 (b) Preparation of Lithium Gallium Hydride.,.,...'. 19 (c) Preparation of Trimethylamine Gallane 20 (d) Preparation of Alkylamino Gallazanes 21 Preparation of Ethylamino Gallazane .......... 22 (e) Reaction of Me^NGall-j with aniline 23 Reaction of j&TKOal^NMe^ with Methylamine 23 (f) Preparation of Cyclic Imino Gallazanes 25 Preparation of Aziridino Gallazane ........... 25 - IV -Page (g) Preparation of Cyclic Imino Alazanes 25 Preparation of Pyrrolidine Alazane 27 (h) Preparation of CyclicImino Borazanes 27 Preparation of Purrolidino Borazane 27 i (i) Preparation of Aziridino Borazane j 29 (j) Preparation of Aziridine Gallium Trimethyljand Aziridino Gallium Dimethyl 30 (k) Preparation of AziridinePreparation of Azetidine 32 IX. Discussion •• 33 Part 1. Alkyl Cyclogallazanes 3Trimeric Cyclogallazanes 5 Dimeric Cyclogallazanes ^2 I.r. Spectra of Cyclogallazanes 51 Part. 2 Reaction of tfe^NGaR"^ with Aniline 53 Part 3 Imino Gallazanes 58 Part k Imino Alazane s 66 Part 5 Imino Borazanes 72 Part 6 Reactions of Imine Bases with EKe^ 77 X. References 81 LIST OF TABLES: Table 1 Analytical data for cyclogallazane compounds 2h Table 2 Analytical data for imino cyclogallazanes 26 Table 3 Analytical data for imino cycloalazane compounds.... 28 Table h Analytical data for imin© trimethyl and imino metal dimethyl compounds........ 31 Table 5 Ions of high m/e in mass spectrum of (Pr^THGaB^^... 50 Table 6 Infrared spectra of some cyclogallazanes in benzene solution 52 - vi -List of Figures Page Figure 1 Drying Pistol.' 7 Figure 2 Sublimer | 8 Figure 3 Me^NC-aH^ Sublimer .• 9 Figure k Vacuum Line, Part A... J 11 Figure k' Vacuum Line, Part E 12 Figure 5 Filtration-Reaction Apparatus 13 Figure 6 Molecular V/eight Apparatus . 1^ Figure 7 Gallium Trichloride Apparatus '.. 18 Figure 8 Hydrogen elimination scheme for gallazanes, alazanes and borazanes 3k Figure 9 Conformations of Trimeric Gallane Species 36 Figure 10 lOOMc/s *H n.m.r. spectrum of EtNHGaH9 in.benzene solution..... 38 Figure 11 'H n.m.r. of some cis and trans trimers ^.0 Figure 12 100 Mc/s 'H n.m.r. spectrum of EtMGaH2 in benzene solution. ^1 Figure 13 100 Mc/s 'K n.m.r. spectrum of EtNHGaD2 in benzene solution • ...' ^3 Figure 1^ 100 Mc/s >H n.m.r. spectrum of i-PrNHGaH2 in benzene solution.. ,. kk - vii -Page Figure 15 60MHz 'H n.m.r. spectrum cf neat a PrilTHGaHj and b Pr^NHGa,]^ k6 Figure 16 Conformations of Dimeric Gallane Species ^7 Figure 17 60Mc/s 'H n.m.r. spectrum of neat sec-BuNIIGaH2... ^9 Figure 18 60Mc/s 'H n.m.r. spectrum of ^NHGaH2 .NMe-j in benzene solution Figure 19 Infrared spectra of: a aniline; b MeNHGaH2; c aniline & MeNITGaH2 56 Figure 20 Structure of Aziridino G-allazane 59 Figure 21 60Mc/s 'H n.m.r. spectrum of (CH2)2NGaH2 in benzene solution 61 Figure 22 60Mc/s 'H n.m.r. spectrum of (CH^lTGaJ^ in benzene solution 63 Figure 23 60Hc/s 'H n.m.r. spectrum of (CH2)it,NGaH2 in benzene solution 6k Figure 2k 60Mc/s 'H n.m.r. spectrum of (CH?) NGaII? in benzene solution.. 65 Figure 25 60 Hc/s 'H n.m.r. spectrum of (CH2)2NAIH2 in benzene solution 67 Figure 26 60Mc/s »H n.m.r. spectrum of (CH2)2NAIH2 in benzene solution 68 Figure 27 60Mc/s 'H n.m.r. spectrum of (CH2)-jNAIH2 in benzene solution. 70 - viii -Page Figure 28' 60Mc/s 'IT n.m.r. spectrum of (CH^ItfAIHg in benzene solution 71 Figure 29 60Mc/s 'H n.m.r. spectrum of (CE2)^K3H2 in benzene solution 73 Figure 30 60Mc/s *H n.m.r. spectrum of (CH^^NBH^ in benzene solution 7^ Figure- 31 60Mc/s 'H n.m.r. spectrum of (CH2)^IJBH2 in benzene • solution 75 Figure 32 60Nc/s 'H n.m.r. spectrum of (CH2_)2N.BH2 in benzene solution 76 Figure 33 60Mc/s 'H n.m.r. spectrum .of H"3B.NH(CH2)2 in benzene solution 78 Figure 3^ 60Mc/s 'H n.m.r. spectrum of Me^B.NH(CH2)2 in benzene solution 79 Figure 35 60Mc/s 'H n.m.r. spectrum of Me^Ga.I\fH(CH2)2 in benzene solution 80 - ix -ACKNOWLEDGEMENT I would like to express my sincerest thanks to my research director Dr. Alan Storr, for his invalua.ble advice, guidance, and enlightening discussions throughout the course of this work. I would also like to thank Dr. B. S. Thomas for his help with some of the more difficult- preparative procedures encountered during the course of this work. INTRODUCTION The chemistry of gallium hydride has developed quickly since the discovery of the stable adduct, Me^N.GaH^, trimethy1amine gallane (1). Previous to this there had been a long search for uncoordinated gallium hydride and its derivatives. Free gallium hydride, although originally believed to be a temperature stable dimer digallane. Ga0H (2), has recently been shown to be a viscous polymeric liquid which disproportionates at -15°C into gallium and hydrogen (3). On the basis that, this material was benzene insoluble, these workers suggested that it was not dimeric, but rather, polymeric like aluminum hydride. IR spectroscopy showed the -1 -1 characteristic strong oGa-H at 1980 cm and AGa-H at ca. 700 cm for this compound. In addition, analysis showed a gallium to hydrogen ratio of one to three, proving that this was the long sought after (4) hydride of gallium. By a procedure analogous to that used for the preparation of gallium hydride, monochloro gallium hydride (GaH^Cl)^ was prepared and characterized as polymeric (5). Subsequently dichloro gallium hydride (GaHC^^ was prepared (6) by a different route and shown to be dimeric rather than polymeric. Lithium gallium hydride, LiGaH^, was first isolated by Finholt, Bond and Schlesinger (7) by the reaction: Et 0 4LiH(s) + GaCl3(s) ——- LiGaH^ + 3LiCl(s) . This compound is the only complex metal gallium hydride which is stable 2 at' room temperature, and then only as an ether solution. Two other unstable analogues, both disproportinating at below -15°C, are AgGall. (8) andTl(GaH,)„ (9). The reaction of LiGaH, with water causes 4 4 3 4 vigorous evolution of four moles of hydrogen. Hence anhydrous conditions are necessary for preparation and storage of this compound. The GaH^ moeity forms complexes with a number of organo compounds of the group V and group VI elements, in addition to adducts formed with the hydride ion (H ), as found.in LiGaH^. The preparation of these compounds is summarized in a recent review on gallium hydride and derivatives (10). Trimethylamine. gallane, Me^N.GaH^, is, in comparison with other gallium hydrides, fairly temperature stable and can readily be sublimed at room temperature. It can be prepared easily by the reaction of excess lithium gallium hydride with timethylamine hydrochloride in the following manner: LiGaH. (s) + Me0NHCl(s) »- Me„N.GaH (s) + LiCl(s) 4- Ht<8^ 4 3 3 3 This compound was the first metal hydride to have sufficient vapor pressure to enable the gas phase IR spectrum to be recorded (11). The gas phase IR spectrum exhibited strong absorptions due. to \5 Ga-H at 1853 cm ^ and SGa-H at 758 cm These assignments were confirmed by deuteration of the protons on the gallium atom. The shift of the Ga-H stretching and deformation, vibrations to lower frequency was by a factor of , as expected. Trimethylamine gallane has been shown by tensiometric titration to add a molar equivalent of trimethylamine gas and form a 2:1 adduct (11). Upon warming to room temperature this material reverted back to the starting material with evolution of trimethylamine gas. Dimethylamine gallane was prepared recently by transamination of trimethylamine gallane with dimethylamine gas (12). Me2NH(g) + Me N.GaH (s) =• Me3N(g) + Me^H.GaH (s) 2Me2NH#'GaH (s) —»• 2H2 + Me2N - GaH2 H2Ga - NMe2 Over a period of a few weeks Me2NHGaH evolved one molar equivalent of hydrogen to give the gallazane shown in the second equation (above). It was shown that this adduct was dimeric in.benzene solution. From con sideration of the gas phase IR spectrum, it was concluded, however, that this compound was monomeric in the gas phase, having C2 symmetry (12). The transamination reaction with gaseous ammonia has recently been shown to proceed via hydrogen elimination to give a quantitative yield of the polymeric solid (NH^GaH,^)^ (13) according to the following reaction: Me3N.GaH3(s) +• NH (g) -a- H^N.GaH^s) + Me N(g) + E^g) . A similar reaction with methylamine gas gave a mixture of two isomers of trimeric (MeNH.GaH2) according to the overall equation: Me3N.GaH (s) + MeNH^g) —* MeNH.GaH^s) +'Me N(g) + H2(g).. The present study involved an extension of this series of gallazanes, (RNH.GaH^^, in an attempt to elucidate the various factors which govern the value of n, the degree of association. In addition to the use of primary alkylamines [R = Et, Pr11, Pr1, Bu11, Bu^", BuS, Bu1"], the transamination reaction using aniline was also investigated. The second part of this work was concerned with a study of the reaction of cyclic imines, [ (CI^) NH where x = 2, 3, 4 or 5] with tri-methylamine gallane. The imino gallane products [(CH ) NGaH„] , were expected to involve some double ring strain and an investigation of this effect was undertaken. A further extension of this latter study involved the preparation and characterization of similar boron [CH^^NBB^]^ and aluminum [ (CI^^AIH] compounds. The reaction of imine bases with diborane to yield adducts with the general formula, [CH^^NH.BH^, where x = 2, 3, 4, 5 was studied in 1956 by Burg and Good (14). Three of these adducts gave, on hydrogen elimination, materials of composition:-(CI^) N.BI^ [where x = 3, 4, 5]. However, the aziridine compound, x = 2, appeared to give ring-opened, polymeric products, and was not isolated. In 1969 S. Akerfeldt et al (15) prepared the adduct aziridine borane, as well as aziridino borazane. The latter compound was until then believed unpreparable. Simultaneously, a crystal structure of the adduct (CI^^NH'M^ was reported (16), in addition to a "*"H nmr and infrared study of both the adduct and the aziridino borazane (17). This latter study rejected the previous p.NH2 formulation of a ring opened product,! \ , in the preparation of the L_BH adduct. (18) The preparation of aziridino alazane and related cyclic imino alazanes has received some recent attention. The first preparation of the cyclic compounds dates back to 1962, when some Italian workers isolated the piperidino and pyrolidino alazanes (19). Their preparation of aziridino alazane was hampered by the fact that this material decomposed with some violence at room temperature in the absence of solvent. More recently, Ehrlich (20) discussed in detail the preparation and subsequent ring opening of this material; which he suggests is polymeric. The present study on cyclic imino boremones and alaiemev has a twofold purpose. Firstly, as indicated previously, to compare these compounds with the gallium derivatives; and secondly to reinvestigate and extend the previous studies. The final part of this work involved preparing the aziridino metal dimethyl derivatives, [ (CH^^^MM^^ where M = B, Al, Ga, In, in order to investigate the effect, on the degree of association, of replacing the hydrogens on the group III atom with methyl groups. EXPERIMENTAL . . A. Experimental Techniques (a) Desiccation All gases were dried first by fractionating under high vacuum through a trap at -20°C, to remove large amounts of water, and then condensed at-196°C into one limb of a drying pistol, see Figure 1, packed with a mixture of glass-wool and phosphorus pentoxide. The gas is passed through the phosphorus pentoxide by alternately cooling one limb and then the other limb. The dried gases are then stored at less than one atmosphere in large glass bulbs attached to the vacuum line. All solvents were dried and redistilled before use; diethyl ether over lithium aluminum hydride, benzene and cyclohexane over molten potassium. The amine ligands which were commercially available were dried by refluxing over CaR^ followed by distillation. Solid components were purified by sublimation, either by vacuum bulb-to-bulb sublimation or as with trimethylamine hydrochloride, sublimed to the cooled central finger of the apparatus shown in Figure 2. Trimethylamine gallane was sublimed, under dynamic vacuum from the flask to the large vertical tube, marked as A, of the apparatus, which was cooled to -80°C, shown in Figure 3. All glassware was washed with acetone, oven dried, evacuated and filled with nitrogen before use. All nitrogen used was Canada Liquid Air "L" grade, purified nitrogen. ^ The hydride and alkyl derivatives, because of their relative instability and extreme reactivity with oxygen or water vapour were Figure 1 Drying Pistol BI4 n. Figure 2 Sublimer Me3NGaH3 Sublimer Figure 3 all prepared and handled in either a high-vacuum system or a nitrogen filled dry box. The high vacuum system developed for the work is shown in Figure 4. A double-stage rotary oil pump (Welch Scientific Co.) and an electrically heated single stage mercury diffusion pump were used to -4 obtain a vacuum of greater than 10 mm of Hg. The dry box (Kewaunee Scientific Equipment) had a special fort chamber that could be evacuated by a double-stage rotary oil pump and then filled with dry nitrogen to ensure the purity of the atmosphere in the box.. The dry box is also connected to a circulating pump which circulates the box's atmosphere through a drying train containing molecular sieve (Fisher type 5A) and a copper furnace to remove any oxygen. _ (b) Reaction-Filtration Apparatus The apparatus shown in Figure 5 found extensive use in our wo The apparatus is evacuated,filled with dry nitrogen, and the reactants are placed in flask A. Additional reagents may be added during the cou of a reaction by rotating the dumper tube B_, the reaction mixture is stirred by a magnetic bar C_. The products, if gaseous may be removed b a Topler pump through one of the stopcocks, or if in solution can be filtered through the sintered disc p_ (medium porosity) by cooling or evacuating the receiver flask E^. (c) Molecular Weights Molecular weights were determined by the cryoscopic method, the dry box an accurately known weight of pure compound was dissolved i a weighed sample of pure benzene (about 10 ml). The benzene solution was poured into the molecular weight apparatus, see Figure 6, and remov Vacuum Line, Part A Figure 4 from Part A Figure 5 14 Molecular Weight Apparatus Figure 6 from the.dry box. A slow stream of pure nitrogen was flushed through the apparatus as it was cooled in an ice bath. The freezing point of the solution was recorded and compared with that of pure benzene solvent and with standard solutions of biphenyl in benzene solvent. The following empirical formula was used to calculate the molecular weights. [K^]X[weight of sample (gms)] molecular weight = [weights of benzene solvent (gms)]X [change in temperature (°C)] K.£ = freezing point depression constant 5.20°C per molal. (d) Spectroscopy : Infrared spectroscopy was used throughout this work for semi quantitative analysis and for structural determination of compounds. Infrared spectra were recorded on a Perkin-Elmer Model 457 spectrometer (4000 - 250 cm The observable range for both liquid and gas samples was between 4000 and 400 cm ^ because KBr windows were used. For gaseous or volatile samples a 10 cm gas cell was used with KBr windows. For liquid or solution samples a 0.05 cm fixed path length solution cell with KBr windows was used and a variable-thickness cell filled with pure solvents (usually benzene) was placed in the reference beam to. compensate for solvent absorption. Because of the instability of most of the hydride adducts prepared, all infrared solution cells were loaded in the dry box and a spectrum run as rapidly as possible. As with infrared spectroscopy, nuclear magnetic resonance spectroscopy, NMR, was used as a tool to investigate reactions and for structural determination. The instruments used were a Varian A-60 and Varian T^-60 both operating with a radiofrequency of 60 megacycles per 16 second and a Varian HA-100 which operates at a radiofrequency of 100 megacycles per second. Most samples were run in benzene solution with a concentration of about 0.1 M to 1 M. The benzene proton signal was used as an internal, standard and was defined as T = 2.840 p.p.m. Tetra-methylsilane, TMS, was used as an external standard on several samples and is defined as = 10.000 p.p.m. The NMR sample tubes were specially fitted with a flame-seal constriction and a B-10 quick-fit cone so that the samples could be loaded and sealed under an atmosphere of nitrogen. As with the infrared samples, the NMR spectra were run as rapidly as possible since steady decomposition at room temperature often impeded prolonged investigation. (e) Elemental Analysis  (i) Active Hydrogen: Active hydrogen was measured by placing a small weighed amount of compound in a round bottom flask in the dry box, attaching a stopcock adaptor and evacuating on the vacuum line. A small volume of degassed, dilute aqueous HNO^ solution was then condensed onto the solid at -196°C. The mixture was allowed to reach room temperature and left to react for about one hour with stirring. Me3NGaH3 + 3H+ — Me3N + Ga+3 + 3H2 The volume of hydrogen gas, non-condensable at -196°C, was then measured using a Topler pump. The amount of active hydrogen in the compound was then calculated. This aqueous solution was made up to.a known volume and an aliquot was used in the determination of gallium as indicated below. 17 (ii) Gallium (Aluminum): A measured aliquot of the solution prepared in section (i) was measured out into a beaker. The solution was first made neutral with dilute ammonia solution, then was made slightly acidic, pH 5-7, with dilute aqueous HC1. The solution was then heated to 80°C and!a slight excess of a 5% solution of 8-hydroxyquinoline in glacial acetic acid was added followed by an aqueous solution of saturated ammonium acetate until pre cipitation of Ga(C H NO) is complete. After digestion at 80°C for one hour, the yellow precipitate was collected in a filtration crucible and the precipitate washed, first with hot, then cold water. The precipitate was then dried at 120°C, weighed and its gallium content calculated from the formula Ga(CgHgNO)^ which is 13.89% galliumby weight. This method has been found to give accurate determinations for a minimum concentration of 10 mg of gallium in 50 ml of solution. Aluminum was determined similarly as its 8-hydroxyquinolate. B. Preparative (a) Preparation of Gallium Trichloride (23) GaCl^ Gallium trichloride xjas prepared by direct combination of the elements. Pure chlorine gas (Matheson Ltd.) was dried by passing through concentrated sulphuric acid in a bubbler and was then passed into the all glass apparatus shown in Figure 7. The gallium metal, about 15 gms, (Alfa Inorganics Inc.) placed in A soon melted on warming with a bunsen burner, and reacted with the chlorine, first to give a colourless liquid, gallium tetrachlorogaliate (21), Ga2Cl^ (melting point 170.5°C (22)). On adding more chlorine this liquid Ga9Cl, disappeared and the liquid gallium Gallium Trichloride Apparatus Figure 7 ca burned with a grey-white flame giving a volatile white solid, gallium trichloride GaCl^, (melting point 79°C). 2Ga(l)'+ 2Cl2(g) ^ (Ga+)(GaCl4") i + 1 (Ga )(GaCl4 )(1) + Cl2(g) ——* Ga2Cl6 The rate of flow of chlorine gas and rate of heating the molten gallium were adjusted so that most of the volatile GaCl^ was deposited in the cooled receiver boat C!. After all the gallium had reacted (essentially 100%) , any sublimate in A was driven into C_ by warming and then flame sealing the constriction at B_. The apparatus was then evacuated and flame sealed at F_. The crude halide was then resublimed into the ampoules E_ and then these were sealed at their constrictions. The gallium trichloride was found to remain stable indefinitely when stored this way. (b) Preparation of Lithium Gallium Hydride (7), LiGaH^ Et 0 4LiH + GaCl_ = *• LiGaH. + 3LiCl 3 room Temp. 4 An ampoule of GaCl^, was weighed and broken open in the dry box and placed in a conical flask. The gallium trichloride was then dissolved in diethyl ether and the ampoule washed several times to ensure quantit ative removal of GaCl^. The empty ampoule was reweighed and the weight of GaCl^ determined. The ethereal solution of GaCl^ and all the washings were now added to the nitrogen filled reaction-filtration apparatus (see Figure 5) and the solution brought up to about 150 ml. From the weight of GaCl calculated, (8.59 gms; 48.8 mmoles) ,20 the weight of about 16 molar equivalents of finely ground lithium hydride (7.45 gms; 938 mmoles) (Alfa Inorganics Inc.), enough for a four-fold excess, was weighed out under nitrogen into the dumper tube. The reaction flask was cooled to -50°C in an acetone-solid C0o bath and the dumper tube rotated upwards to permit the slow addition of LiH to the reaction flask over a period of about thirty minutes. A bubbler was attached to the apparatus so that the reaction could be carried out under a constant pressure of one atmosphere of nitrogen. The coolant was allowed to warm up to room temperature and the mixture was stirred for about fifty hours to ensure.complete reaction. The resulting reaction mixture was filtered through the glass sintered disc and a clear colourless filtrate resulted. This filtrate was then transferred, in the dry box, to a conical flask fitted with a break seal and an extended neck which was flame sealed for storage. The LiGaH^ ether solution was observed to be indefinitely stable if stored in ' all glass ampoules under a nitrogen atmosphere and cooled below 0°C. Lithium galliuw deuteride, LiGaD^, was prepared and stored in exactly the same manner as LiGaH^, only lithium deuteride, LiD, (Alfa Inorganics Inc.) was substituted in the preparation for lithium hydride, (c) Preparation of Trimethylamine Gallane (1), Me^NGaH^ Et 0 LiGaH. + Me NHC1 * Me NGaH + LiCl + H0 4 3 room temp. 3 3 2 A known amount of lithium gallium hydride (2.38.gms; 29.4 mmoles) in ether solution was placed in the reaction-filtration apparatus, see Figure 5. Slightly less than the stoichiometric amount of trimethylamine hydrochloride, Me_NHCl, (2.644 gms; 27.6 mmoles) (Alfa Inorganics Inc.) 2f dried and purified by sublimation, was placed in the dumper tube of the reaction vessel which contained a nitrogen atmosphere. The ether solution of LiGaH, was first cooled to -50°C in a drv-4 * ice cooled acetone bath, as the trimethylamine hydrochloride was added over a period of about 10 minutes. Then the solution was allowed to warm up to room temperature and stirred for about four hours' to ensure complete reaction. The solution was next filtered through the glass sinter and the receiver flask containing the clear ether solution was attached to the sublimation apparatus, see Figure 5. This apparatus was attached to the vacuum line and the ether was pumped off at -50°C. When most of the ether was removed, the residue was allowed to warm up to room temperature while the large bulb part of the sublimation apparatus was immersed in an acetone-solid CO^ slush bath. The pure trimethylamine gallane was vacuum sublimed as long needle like crystals into the cooled receiver. The. overall yield in going from gallium trichloride to trimethylamine gallane was about 60%. „i The deuterated compound, trimethylamine trideuterogallane, Me^NGaD^ was prepared in the same manner only lithium gallium deuteride was substituted for lithium gallium hydride. Trimethylamine alane , Me^NAlH^, was also obtained similarly from commercially available LiAlH^ and tri methylamine hydrochloride. (d) Preparation of Alkylamino Gallazanes (RNHGaH^^ As the procedures are similar for preparation of all the gallazane compounds, only the procedure for the ethylamino compound will be given as an example. 23 benzene. A weighed quantity of this; solution was removed from the cryo-scopic molecular weight apparatus and hydrolysed. The volume of hydrogen evolved on hydrolysis was then determined. The gallium content was determined gravimetrically by standard procedures. The deuterio derivative, (EtNHGaD^)^* was obtained by an exactly similar procedure to the above, but using Me^NGaD^ as the starting material. Experimental details for the other alkylamino gallazanes are summarized in table 1. (e) Reaction of Me„NGaH„ with aniline (C,HCNH„) J J p J z Aniline (.405 g, 4.352 mmoles) was condensed onto trimethylamine gallane (.573 g, 4.351 mmoles) at -196°C and allowed to warm to room temperature After complete reaction (about two days) the flask was cooled to -196°C and the volume of evolved hydrogen measured (Found: 92.5 ml; Calc. 97.8 ml). The mixture was then allowed to warm to room temperature and a trace of Me^N was gas detected. The white solid product, Me^NGaH^NH , was mono-' meric in benzene (Found: 224, Calc. 223) and gave the following analysis: Ga: Found: 31.9%, Calc: 31.2%. > H active: Found: 1.12%, Calc: 1.12%. Reaction of a two molal quantity of aniline led to an insoluble polymeric material. It evolved a 2 molal quantity of hydrogen as well as a molal quantity of Me^N. Reaction of c^NHGaR^NMe^ with Methylamine A measured amount of methylamine gas (42.8 ml) was condensed onto a weighed quantity of (JiNHGaH^NMe^ (.426 g, 1.878 mmoles) at -196°C and this mixture was then permitted to warm to room temperature. No hydrogen was evolved. The volume of trimethylamine gas was measured (Found: 92.4 ml, Calc: 91.8 ml) and.its purity was checked by gas phase infrared spectroscopy. This product, as well as the products resulting Table 1 Analytical data'for cyclogallazane compounds prepared by the reaction:-Me.NGaH. + RNH„ (RNHGaH.) + H_ + Me.N j j / 2. n l J Compound Phase Moles H2 per Moles Me3N Degree of assoc Analysis at 25°C mole RNH2 • per mole RNH2 iation, n Found % (RNHGaH?) requires 7J Ga Hydrolysable hydrogen Ga Hydrolysable hydrogen EtNHGaH2 Viscous liquid 1.01 1.01 2,92 60.1 1.73 60.2 1.73 PrnKHGaH2 Viscous liquid 0.98 1.02 2.64 53.5 1.53 53.7 1.54 BuI1NHGaH2 Viscous liquid 1.00 1.09 2.57 48.4 1.37 48.5 1.39 PrXNHGaH2 Mobile liquid 0.92 0.98 1.91 53.6 1.55 53.7 1.54 BuXNHGaH2 Viscous liquid 0.95 1.03 2.15 48.4 1.38 48.5 1.39 • BuSNHGaH2 Mobile liquid . 0.92 1.02 1.83 48.5 1.40 48.5 1.39 ButNHGaH2 White solid 0.97 1.02 1.83 48.4 1.37 48.5 1.39 from the reactions: aniline plus Me^NGaD^, aniline plus Me NGaH , methylamine plus (^NHGaD^NMe^, and methylamine plus (JlNHGaH^NMe^ were characterized by infrared and ^"H nmr spectroscopy. (f) Preparation of Cyclic Imino Gallazanes Since the procedure for the preparation of these "double ring strain" gallazanes is standard throughout the series, and since the I technique, and apparatus are essentially the same as those used in pre paration of the simple gallazanes, only a short, procedure for aziridino gallazane will be given as an example. Preparation of Aziridino Gallazane . Aziridino gallazane was prepared by condensing aziridine gas (23.8 ml; 1.50 mmoles) onto trimethylamine gallane (0.140 g; 1.60 mmoles at -196°C, and allowing the mixture to warm slowly to room temperature. After complete reaction (about 1 h) the flask was cooled to -196°C, and the volume of evolved hydrogen measured (Found: 23.6 ml, Calc: 23.8 ml). The mixture was again brought to room temperature and.the volume of trimethylamine gas was measured (Found: 24.4 ml, Calc: 23.8 ml). The purity of the Me^N was checked by its gas phase i.r. spectrum. The white, crystalline solid product was analysed for hydrolysable hydrogen and for gallium by the previously discussed methods. The analytical dat for the compounds prepared in this series are given in table 2. (g) Preparation of Cyclic Imino Alazanes The procedure for the preparation of this series of alazane compounds is standard throughout the series. Hence the pyrrolidino alazane preparation, only is given as an illustrative example. Tabie 2 Analytical data for imine cyclogallazane compounds prepared by the reaction:-. Me3NGaH3 • + (dft2)xNH ===== ((CTI2)x1lGaII2)n + H2 + Meyi Compound Phase Moles H2 per mole imine Moles MeoN Degree of association ,1 Analysis at 25*C per mole imine n. Found % Theory % 1 Ga IJydrol. j hydrogen Ga Hydrol. hydrogen 'CH2)2NGaH2 White solid 1.01 . 1.00 2.00 (2.56) i \ • 62.1 1.76 61.1* 1.76 (CH2)3NGaH2 White solid 0.99 1.01 2.00 5^.1 1.55 5^.5 1.56 ;CH2)^IGaH2 White solid 0.99 0.99 2.02 |^9.0 ! i 1.38 '+9.2 l.'H ;CH2)5NGaH2 White solid 0.99 0.98 1.89 |^3.9 1.26 1.28 * Degree of association immedtately after dissolving imine cyclogallazane in benzene. \ 27 Preparation of Pyrrolidino Alazane (CH„).NA1H„ t t- 4 2 The bis trimethylamine alane used in the reaction was prepared by condensing excess Me^N gas onto trimethylamine alane at -196°C. After equilibration pf this system at room temperature, the excess trimethylamine was removed at -20°C, leaving the bis adduct. Pyrrolidine (35.0 ml, 1.559 mmoles) was condensed onto bis trimethylamine alane (0.228 g; 1.542 mmoles) dissolved in 5 ml of dry benzene. This mixture was permitted to warm to room temperature. After the evolution of hydrogen had ceased, the flask was cooled to -196°C and the volume of hydrogen measured (Found: 35.2 ml; Calc: 35.0 ml). The benzene solvent and trimethylamine gas from the reaction were then removed at -20°C to leave a white crystalline solid in the reaction vessel. Analyses for aluminum and hydrolysable hydrogen were performed only on the aziridino alazane since most of these compounds had been previously prepared and analysed (23). Experimental data for this series of compounds is summarized in table 3. (h) Preparation of Cyclic Imino Borazanes The procedure for the preparation of these borazane compounds is standard for three of the derivatives, (CH ) NBH where x = 3, 4, 5 and therefore the preparation of pyrrolidino borazane only will be given. The preparation of aziridino borazane differs slightly and will be described later. Preparation of Pyrrolidino Borazane Pyrrolidino borazane was prepared by condensing pyrrolidine (100 ml, 4.45 mmoles) on a previously condensed sample of diborane (50 ml, 2.22 mmoles) in a 500 ml break-seal flask. The mixture was Table 3 Analytical data for iminC cycloalazane; compounds prepared by the reaction:-(Me3N)2AlH3 + (CH2)XNH - ((CH2)xKAlH2)n+ H2 + 2Me3N Compound Phase, at 2$*C Moles H per' mole imine Molecular .. j weight i j Degree of association n (CH2)2NA1H2 White solid 1.02 298 4.20 (3.14*) (CH2)3?LA1H2 White solid 1.00 263* 3.06* (CH^NAIIL. White solid 1.01 308 3.10 (CH2)5NA1H2 White solid 0.98 243 2.17 Analytical data for imino cycloborazane compounds prepared by the reaction:-HH6 + (CH2)XMI ((CH2)yHBH?)n + H2 Compound Phase at 25*C j Holes IT2 per i mole imine ! Molecular weight Degree of association, n CCH2)2KBH2 White solid — 165 3.00 (CH2)3NBJI*' White solid 0.97 134 1.9*+ (CH2)^JB.H2 White solid 1.03 166 2.00 (CH2)ra.H2 White solid 1.08 196 2.02 • * Private communication Dr. B. S. Thomas. allowed to warm to room temperature to form the liquid adduct. The bulb was then cooled and sealed off under vacuum. It was then placed in an oven at 128°C for 3 1/2 hours to pyrolyse the adduct. After pyrolysis was com plete, the flask was attached to the vacuum line, cooled to -196°C and the fragile break-seal ruptured with a bar magnet. The evolved hydrogen was measured (Found: 103 ml, Calc: 100 ml). The product was then warmed to room temperature and checked for non-condensibles. Experimental data for these compounds is given in the lower part of table 3. (i) Preparation of Aziridino Borazane This compound was prepared by condensing aziridine (100 ml, 4.45 mmoles) onto a sample of diborane (50 ml, 2.22 mmoles) at -196°C. About 5 ml of strictly dry diethyl ether was condensed onto this mixture and the mixture warmed to -130°C. At this point the mixture was per mitted, by means of a propane slush bath, to warm slowly to -78°C. The ether was removed giving a product, which when solvent free was a white crystalline solid. The infrared and "^H nmr spectra of this adduct agreed with those found in the literature (18). The adduct was dissolved in benzene and refluxed under an atmosphere of dry nitrogen for three to four hours. The aziridino borazane product was separated by removing the benzene solvent at -20°C. The IR and nmr spectra recorded for the aziridino borane obtained by this method, agreed with those found in the literature (18). Attempts to prepare this complex by a pyrolysis method using the reaction of aziridine with either Me^N.BH^ or diborane failed to give the desired product. These reactions were non-stoichiometric, yielding 40% of the theoretical hydrogen and 77% of the Me^N in the first case and only 54% of hydrogen in the last. The products in each of these cases gave liquid plus solid but were not soluble in benzene to any significant extent. (j) Preparation of Aziridine Gallium trfmethyl and Aziridino  Gallium dimethyl The adduct aziridine gallium trimethyl was prepared by condensi: aziridine (75.5 ml, 3.36 mmoles) onto gallium trimethyl (75.5 ml, 3.36 mm< at -196°C and warming to room temperature. The adduct was a clear mobile liquid which was stable to methane elimination at room temperature. The aziridino gallium dimethyl was prepared by pyrolysing a 0.413 g sample of the previously prepared adduct at 110°C for 5 hours in a break-seal bulb. After the five hour reaction time the bulb, now con taining a white solid (mp 184°C).was connected to the high vacuum line, cooled to -196°C, the glass break seal ruptured and the methane measured (Found: 56.8 ml, Calc: 58.8 ml). The product was then warmed to room temperature and checked for the presence of condensibles. The analytical data for the other compounds of this series is given in table 4. (k) Preparation of Aziridine ^NM Since commercial samples of aziridine were not available the preparation of this material was undertaken using the following route. The methods of Wenker (24) Leighton (25) and Reeves (26) were all tried but gave lower yields than the following method. 96% H2S04 (109-9 S» 1>0^ m°les) was added directly to a stirred sample of ethanolamine (65.7 g, 1.07 moles). This mixture was then heated to 100°C under water aspirator vacuum to give a quantitative yield of ethonolamine sulfate according to the following scheme: Table 4 Analytical data for imine metal trimethyl and imino metal dimethyl compounds prepared by the following: (CHg) NH IZZZZ Me3M.HH(CK2)2 Me3M.KH(CH )2 (Me2M.N(CH2)2)n + CH^ Compound Phase at 25*C Moles methane per mole imine ; j Pyrolysis : temperature Degree of association n Me3GaM-l(CH2)2 ; mobile liquid 110*C,5h Me2GaH(CH2)2 | white : ! solid 0.97 2.88 Me3EKH(CH2)2 • '. mobile liquid • 180 C,12h Me2EN(CH2)2 v/hite solid 0.68 polymeric solids and liquids Me3AlM(CH2)2 mobile, liquid evolves CH^ : at r.t. 60*,4h Me2AlK(CH2)2 v/hite . solid . " " 0.88 ! • | 2.96 iMeoInKH(CH )0 i i ? 1 mobile liquid evolves CH^ •• at r.t. 80'c,12h Me In!Sf(CII ) i 2 22 white solid ' 0.70 ' '3.00* *Private communication Dr. B. S. Thomas. - H0CH2CH2NH2 + H2SO The white solid product was ground with 95% EtOH, suction filtered and dried in a vacuum descicator over The ethonolamine sulfate was then placed in a 1000 ml round bottomed flask surmounted by a still head and water condenser set for downward distillation and overlaid with a 40% NaOH solution (95 g NaOH, 143 g H20). The flask was heated with an open flame and the distillate collected rapidly in a well cooled 500 ml receiver. Once distillation was complete, enough KOH to obtain a saturated solution was added and the flask stored in the fridge overnite. The upper organic layer was then removed and dried over CaH2/K0H. The product, when water and ethanol free was stored over CaH2 at +5° until required. (Yield ~15%). Azetidine (27) f CHH • Azetidine was prepared by the same procedure as above but starting with propanolamine instead of ethanolamine. The yield was about 1%. H20 + CH2-CH2 O.-SO3- N1I3 DISCUSSION Part 1 The ease of intramolecular hydrogen elimination from adducts of the type MeNH^, EH^, where E is B, Al or Ga follows1' the sequence B < Ga < Al as illustrated in figure 8. Note that 90° |(28) is required i for hydrogen elimination with boron* MeNH^GaH^ eliminates hydrogen at room temperature (13) , while MeNH^AlH^ eliminates two molar equivalents of hydrogen at -20° (29). Stone (30) explains this sequence in terms of the relative electro-negativity values of the atoms involved. 5+H H5- 6+H H5-| I V S Me-N >- G-H y Me-N > E ~H _ >- Me-N —E-H + H I \ * I \ \ A I 2 H H . H H H H In the above scheme the hydrogen attached directly to the nitrogen atom is considered to lose electron density on formation of an electron donor bond by the donor moeity. Hydrogen attached to the acceptor atom, E, simultaneously increases in electron density and an electrical strain is thus created in the adduct. The strain is relieved when hydrogen elimination occurs. On the basis that the differences between the Allred-Rochow (31) electro-negativities of the E atoms and that of hydrogen (at 2.1)increase in the order B, Ga, Al,the hydxidic character in EH^, and hence the ease of hydrogen elimination should decrease in the order Al through Ga.to B}as observed. The factors affecting the association of the products from -hydrogen elimination are believed to be the following (32). (i) Steric Effect - With the same donor and acceptor atoms B (2.01) R / H,B«-N-H 3 \ H BORANE 90°, 1 Hr. H R I I -B-N-It n BORAZANE 90°, 10 Hr. T R I B=N-BORAZINE Ga (1.82) / H_Ga<-N-H 3 \ K GALLANE r.t., 1 day H R I I -Ga-N-I I H H n GALLAZANE" DECOMP. (GALLAZINE) Al (1.47) H.A1«£— N-H 3 \ ALANE -20c (ALAZANE) -20° H R I I -Al-N-POLYMERIC NETWORK IMINO ALANE (ALAZINE) Figure 8 -35 increased size of R groups on the E atom cause a shift, to lower oligomers. (ii) Valency angle strain - Dimers contain more strain than trimers, but this is easier to tolerate with larger donor and acceptor atoms, (iii) Entropy - Prefers monomer over dimer and dimer over trimer. (iv) Nature of reaction intermediates. | The cyclogallazanes prepared in this study ranged from white solids to mobile liquids and all had satisfactory analyses for gallium and hydrolysable hydrogen; all were soluble in common organic solvents. As is evident from Table 1, increasing the size of the R group coincides with the formation of lower oligomers. Thus, steric interactions in cyclohexane-type trimers become too large and a preference for the angularly-strained, dimers, with lower steric requirements, becomes apparent. With both the trimeric and dimeric species the physical data (i.r. and '^H nmr spectra) indicate the presence of at least two con-figurational isomers in benzene solution. Trimeric Cyclogallazanes (RNHGaH^)^ A cyclohexane-type ring structure for trimeric cyclogallazanes, (RNHGaH^)^, is proposed on evidence collected from ^"H nmr data and from supplementary evidence from i.r. spectroscopy measurements. As observed with the methyl derivative, (13) at least two configurational isomers are present in benzene solutions of the new trimers. Figure 9. The most stable isomer, on steric grounds, is the one in which all three N-alkyl groups occupy equatorial positions on the ring. The next most^ stable isomer, sterically, is one in which one N-alkyl group is axial and the remaining two N-alkyl groups equatorial to the (Ga-N) ring. Ga R' N H Tsl — Ga H Ga N —R H CIS Ga R' .N H 'N~-Ga Ga N—H R TRANS Conformations of Trimeric Gallane Species Figure 9 -37 These isomers will be termed cis and trans respectively. (EtNHGaH^)^ - 1,3,5-Triethylcyclpgallazane, a viscous liquid at room temperature, is trimeric in benzene solution. The partial "*"H nmr spectrum of the benzene solution at 100 MHz (Figure 10) shows clearly the presence of a number of non-equivalent ^-CH^ groups. 'The signals from these groups consist of three well-defined triplets (J,.!™, ca. 7 Hz). The. HCOH pattern of signals suggests that the triplets A and B arise from ^-CH^ groups in similar environments whereas the triplet C, at higher field, appears unique. It is therefore tempting to assign triplets A and C to the trans-isomer.(ca. 2:1 ratio), and the triplet B to the cis-isomer. The triplets A and B, both assigned to equatorial p-CE^ groups, occur very . close together which is to be expected since little change in equatorial "CH^ environment will occur between the two isomers. These assignments would indicate that the trans-isomer is in greater abundance, which is somewhat surprising for a cyclohexane-type ring on purely steric arguments. Similar trimeric borazanes, (33) however, show this same preference for trans-isomer formation. An alternate explanation is to assign the triplet A to the cis-isomer and the triplets B and C (ca. 1:2 ratio) to a twist conformation similar to the one recently reported for the ethyleniminodimethyl-aluminium trimer (34). In the twist conformation one could again obtain yS-CH^ groups in different environments in a 1:2 ratio, the unique /3'CH^ group being attached to the nitrogen on the two-fold axis of the molecule. This alternate explanation would then indicate the sterically favoured cis-isomer in greater abundance. If the chair-type model is accepted for the trimeric gallazanes, it is interesting to note the appearance of the axial ^3-CH^ signal in (EtNHGaH^)^ at higher field than the equatorial 1 a 9.06 9.08. 9.27 Fig. 10 lOOMc/s 'H n.m.r. spectrum of EtNHGaH in benzene solution 39 signals. This is in contrast to the axial NMe signal in the trans-(MeNHGaH^)^ trimer, which appears at.lower field than the equatorial signals (13). It seems that this downfield shift for methyl groups axial to- cyclohexane-type rings is quite common, occurring in a variety of inorganic ring systems, (MeNHBH^)3» (35) (MeCH.S)^, (36) (MeCH.CH2)3, (37) and (MeCH.O)3, (38) two of which are shown in Figure 11. Perhaps, this phenomenon can be accounted for by invoking van der Waals deshielding due to 1,3-axial interactions. With the /?-CH3 groups of (EtNHGaH2)3 the proximity to axial hydrogens on the nitrogen atoms is evidently not sufficient to give this type of deshielding. The methylene protons in (EtNHGaR^).^ do not give well resolved signals but overlapping quintets are apparent in the "4l nmr spectra (J ?r J ) (Figure 12), rlCiUrl HNCH presumably arising from the axial and equatorial environments in the different isomers. The NH resonance is partly 'hidden' under the fl-CR^ signals in the hydride compound occurring at ca t 9.3, but it appears as a broad triplet (J . --7 Hz) at higher field (t 9.52) in the spectrum of the deuterioderivative, (EtNHGaD2)3, at 100 MHz (Figure 13). Signals due to GaH protons were not observed principally because of low concentrations but also perhaps because of nuclear quadrupole broadening (39, 40). The "4l nmr spectra of the remaining trimeric gallazanes (R = Pr11 and Bu11) are less clearly resolved, even at 100 MHz. The V -CH3 proton signals in • (Pr^JHGa^) 3 appear as a series of triplets (JHCCH ca. 7.2 Hz) centred at t-9.43, 9.44, and 9.46 again indicating the presence of at least two isomers. These triplets are tentatively assigned to cis- and trans-isomers, the triplet at higher field being assigned to the axial J*-CH^ group of the trans-isomer. The H nmr spectra of the n-butyl derivative are very complex, even at 100 MHz, and no assign ment is attempted. Dimeric Cyclogallazanes, (RNHGaH ) ! i i Dimeric cyclogallazanes, (RNHGaH^^ may exist as configurational isomers with the N-alkyl groups cis or trans on the ring [(Ila) and (lib) respectively]. A number of additional variations are possible if the (Ga-N)^ ring is nonplanar, which has been shown to be the case for numerous analogous substituted cyclobutane derivatives (41, 42). Non-planar configurations may be expected more especially in the cis-isomer, to relieve steric interactions between adjacent, bulky, R groups. (Pr NHGaH2)2 - 1,3-Di -isopropylcyclogallazane is a mobile liquid at room temperature and is readily sublimed. In benzene solution its molecular weight corresponds to a dimer. The .''"H nmr spectrum in benzene solution consists of a series of doublets in the "5"-CH^ region of the spectrum (Figure 14). The major doublets, D and E (Jor,„1T 6.3 Hz) at HL.Cn tr 9.14 and 9.15 are assigned to the cis- and trans-isomers of the dimer. The remaining small doublets in this region may be due partly to NH signals (JJJ^QJ 6.3 Hz) or to the presence of small amounts of other oligomers. Attempted fractional distillation, however, failed to separate any components and all fractions when dissolved in benzene gave similar spectra to that shown in Figure 14. The possibility of restricted rotation of the isopropyl groups in one isomer leading to both the major doublets A and B in the spectrum was investigated by obtaining spectra at a series of temperatures (0 - 60°). Although the separation between the two doublets decreased slightly at higher temperatures there was no OJ D E A 8 C I I I I I I L. 1 • ' 1 I III, 44 8.86 8.89 8.93 9.14 9.15 9.32 Fig.14 IOO Mc/s H n.m.r. spectrum of i— PrNHGaH2 in benzene solution. 45 indication.of a collapse to just one doublet and therefore the assignment of A and B to cis- and trans-isomers is preferred. The neat liquid (P^NHGaR^^ and its deuterio'analogue gave the novel *H nmr spectra shown in Figure 15. Here, for the first time, the GaH signals are clearly seen as broad resonances at tr 4.71 and 4.88. The signals are field dependent and indicate the presence of different i environments for hydrogens on gallium atoms. These signals are, of course, absent in the spectrum of the deuterio-derivative, thus confirming the assignment. In addition, the CH multiplet (Ju_nu = JTT_7„.T) , centred at f6.34, and the broad NH resonance at T 7.94 are clearly distinguished. The remaining doublets, A and B, (J 6.4 Hz) due to ft -CH groups are HL.CH '3 centred at T 8.15 and 8.32. Again a mixture of cis- and trans-dimers (Figure 16) is postulated and it is seen as fortuitous that the ratio of the y3-CH^ doublets is approximately 1:2. The presence of the trimer in the liquid form, which could give rise to this ratio, is discounted on the mass spectra data obtained for the deuterio-compound, (Fr^NRGaT)^)2' The ions of high m/e values are listed in Table 5 and correspond to the pattern expected from the dimer (Pr^HGaD^) ^ taking into account the 69 isotopic distribution of gallium atoms in the molecules [ Ga(60%), ^Ga(40%)]. Molecular-ion peaks, although weak, occur in the mass spectrum in addition to peaks due to the more abundant ions which have lost deuterium from gallium. The most intense peak in the spectrum occurs at m/e = 44 and may correspond to the propane ion C0Ho+. The spectrum gave no indication of the presence of trimeric units, and since it is unlikely for the dimer to be converted into trimer in going from vapour to liquid, a dimeric constitution for the neat compounds is 2.84 471 4.88 6.34 7.94 ' 8 32 60MHz 'H n.m.r. spectrum-of neat a Pr'NHGaH2 and b Pr'NHGaO, R Ga ,R Conformations of Dim eric Gallane Species. Figure 16 predicted. s (Bu NHGaH^)^ ~ 1,3-Di-s-butylcyclogallazane is a mobile liquid at room temperature. It is dimeric in benzene solution and in this solvent it has a *H nmr spectrum which exhibits two strong doublets (J ca. 6.6 Hz) HCCH at T 9.13 and 9.16 which are assigned to the 3 -CH^ groups in the cis- and trans-dimers. Signals due to the V -CH^ protons appearj at higher field but the triplets expected on a first-order basis are poorly resolved. The nmr spectrum of the neat liquid showed essentially the same pattern as the solution spectrum but once again, in addition, the GaH signals are clearly visible at T4.64 and 4.81 (Figure 17). (Bu NHGaH ) - 1,3-Di-isobutylcyclogalla zane is a viscous liquid at room temperature and in solution probably exists as a mixture of dimers and trimers. Branching of the hydrocarbon chain of the R group at the £ -carbon atom possibly reduces the steric interaction sufficiently to lead . to both dimer and trimer formation. Four well-defined doublets (Junnu ca. 6.6 Hz) at TT 9.27, 9.30, 9.31, and 9.38 appear in the high field region of the ''"H nmr spectrum in benzene solution at 100 MHz. These are assigned to p 8rouPs but no further assignment is attempted. (B^NHGaH^)^ - 1,3-Di-t-butylcyclogallazane is a white solid at room temperature, dimeric in benzene solution, and displaying three y'S-CH^ signals in its "^H nmr spectrum in this solvent. Two of these signals are close together at t 8.96 and 8.97, and a third, much weaker signal, occurs at higher field (f 9.15). The signals are all field dependent and therefore not due to coupling. The major signals are assigned to the cis- and trans-dimers, the third weaker signal, accounting for ca. 5% of the total integral, is possibly due to monomer in solution. 0 Hz > i l • I I • I i I I I I I I I I L_J 1 I I j I I I I I I ! 1 I ! L_ H i I III i I i i[ ij I 1 I I I I I I I I I L I , I I II I I I L_ 2.84 4.64 5.81 Fig.17 60Mc/s 'H n.m.r. spectrum cf neat sec-BuNHGaH2-Table 5 Ions of high m/e in mass spectrum of (Pr NHGaD ) m/e Relative Abundance 266 0.5 265 0.5 264 5.0 • 263 2.5 262 17.5 261 3.7 260 27.7 259 12.5 258 16.0 257 • 0.5 • 44 100.0 51 I.'r. Spectra of Cyclogallazanes (RNHGaH) 2 n I.r. spectra of the cyclogallazanes (RNHGaH„) , and their L n deuterio-derivatives (RNHGaD„) in some cases, in benzene solution were 2 n recorded in the range 4000 - 250 cm As observed previously with gallane derivatives (5, 11), the strongest absorptions were attributable to the Ga-H and Ga-D stretching and deformation modes. Selected absorption bands are listed and assigned in Table 6 for the ethyl and isopropyl deri vatives which are representative of the trimeric and dimeric cyclogallazanes respectively. As expected on amass effect the ratio V(Ga-H)/ -0(Ga-D) is close to 1.4. The NH stretching abosrptions are interesting in that three bands occur in this region for trimeric species but two bands only, for dimeric species. Presumably the different possible environments for the NH unit in the various cis- and.trans-isomers lead to the observed vibrations but is is noteworthy that the band at 3280 cm ^ in the ethyl derivatives is concentration dependent, decreasing in relative intensity on dilution. Perhaps hydrogen bonding of the type invoked recently by Brown et al (43), to explain the i.r. spectra of similar.cycloborazanes at various concentrations, could be operative, also, in these gallium systems. The i.r. spectra of neat (Pr^HGaH^)^ and its deuterio-analogue were also recorded. In each spectrum the NH stretching vibration occurred as a broad band at 3270 cm Similarly, Ga-H(D) stretching vibrations appeared as broad bands at 1875 and 1825 (1350) cm The -1 Ga-H(D) deformation modes occurred at 725 and 690 (510, 493) cm and absorptions attributable to ring vibrations came in the region 540 -590 cm"1. Table 6 Infrared spectra of some cyclogallazane derivatives in benzene solution EtM.GaH2 EtN-H.GaD2 Gall GaD Assignment 3338 w 3318 m 3280 s 3338 w 3316 m 3280 s N-H stretch 1875 vs 1825 vs 1350 vs • 1335 vs 1.37'+ Ga-F.(D) stretch 745 vs 502 vs 496 vs 1.404 Ga-H(D) defn. 580 s 550 s 510 m 542 s . 522 s Ring modes PrHlHGaH^ 3320 m 3283 s Pr1NHGaD2 3320 w 3283 m Gall GaD Assignment N-H stretch 1875 vs 1820 vs 1355 vs 1330 s 1.3 64 Ga-H(D) stretch 745 vs 508 vs J+97 vs 1.465 Ga-H(D) defn. 586 s 560 m 4-90 m 596 s 552 s 536 m Ring modes Bt^NHGaTTg 3307 v 3208 vs EutKKGaD2 3312 s 3264 s GaH GaD Assignment N-H stretch 1890 vs 1820 m 1318 vs 1.408 Ga-H(D) stretch 745 s 538 vs 521 vs 1.402 Ga-H(D) stretch 598 s 554 s Ring modes 53 Part 2 The reaction of aniline with trimethylamine gallane proceeded as indicated in the following equation: Me3N.GaH3(s) + <$NH2(g) -—» 4>NH.GaH .NMe (s) + H fg) The monomeric material, <^NH.GaH2.NMe3, giving the \l nmr shown in Figure 18, was somewhat unexpected since with the primary alkylamine reactions dis cussed in part 1, complete elimination of trimethylamine occurred with the production of a gallazane (Ga-N) ring species. In the present case it n appears that due to some electron withdrawing effect of the phenyl ring a cyclic gallazane was not formed. This effect seems to have reduced the donor properties of the lone pair on the aniline nitrogen atom, and hence prevents coordination to a second gallium and consequent ring formation. It was believed that introduction of a strong acceptor would remove the trimethylamine from the complex, ^NH.GaH^.NMe^, since a strong donor such as nitrogen always prefers a strong acceptor over a weak acceptor. . • The acceptor of choice was diborane since it is both a strong acceptor and would not undergo any unwanted side reactions such as might occur if the (oron trifluoride, BF , acceptor were used. However, the reaction of diborane with 4NH.GaH^.NMe^ resulted, not in production of the desired gallazane, d)NH.GaH2, but in decomposition into gallium, hydrogen, aniline as well as the expected trimethylamine borane. The ^ following sequence of reactions summarizes these experimental observations:" 2.84 'Fig: 18'60Mc/s 'H n.m.r. spectrum of (J)NHGaH2-NMe3 in benzene solution 55 Me N.GaH .NH$ + 1/2B.H, * Me.N.BH + (j^NHGaH 5 L L o 5 5 L 4)NHGaH2 > <j>NH + Ga + 1/2H It seems likely that when the ' r^NH.GaH^' is formed in the reaction, the donor strength of the nitrogen connected'to the phenyl ring is so reduced that formation of a stable cyclic gallazane does not occur. The monomeric unit is evidently unstable, when formed and decomposes to its components even below 0°C. It was of further interest to react ctNH.GaH^.NMe^ with methyl amine in the hope that displacement of trimethylamine would occur and yield a novel cyclic gallazane on hydrogen elimination according to the following sequence of reactions: '(J>NH.GaH2.NMe3 + MeNH^ *• Me^ + <j>NH.GaH2,NH Me <p NH. GaH2. NH2Me ' $ NH. GaH. NHMe' + H2 The actual mixture of products obtained was identified by nmr and infrared spectroscopy. Figure 19 shows the N-H stretching region for each of aniline and methyl-gallazane as well as the reaction mixture. It should be noted that the two upper spectra combine to give the lower spectrum. Hence, although trimethylamine was displaced as expected, the elimination of aniline and production of the familiar (MeNH.GaH,^)^ trimer occur as follows: Me3N.GaH2.NH<$> + MeNH£ + t}>NH2+ 1/3[MeNH.GaH2] Fig. 19 Infrared spectra of: a aniline; b MeNHGaH2 ; ,c aniline & MeNHGaH2 57 The products were identified also by means of their characteristic "*"H nmr spectra. It was of interest to then establish the mechanism of hydrogen transfer. The two most probable mechanisms for this transfer are illustrated below: H H H 0-N-—-Ga - NMe —> 0NH2 + GaH .NHMe H H H B. 0-N*3-GaH I t Me 5NHMe In the first mechanism, the proton which transfers to the aniline comes from the gallium. In the second mechanism a four centre intermediate is formed with the hydrogen atom for aniline production coming from the methylamine nitrogen. The deuterated compound, <f> NH. GaD2 .NMe^ was therefore prepared and reacted with methylamine. The infrared spectrum of-, the products did not display either a N-D stretch for aniline or a Ga-H stretch for the gallazane, thus eliminating mechanism A as a possible route to the products. It therefore seems likely that mechanism B is the actual mode of proton transfer. 58 Part 3 Imino.Gallazanes The reaction of aziridine, azetidine, pyrrolidine and piperidine with trimethylamine gallane yields compounds of. the type [ (CR^^NGaH^ ] where x = 2, 3, 4 or 5; following elimination of molar equivalents of i hydrogen and trimethylamine. Cryoscopic measurements on centrifuged benzene solutions indicate that all these materials arej dimeric (Table 2) in benzene. Recently, however, an x-ray crystallographic study (45) on a single crystal of aziridino gallazane produced by sublimation under about 5 - 7 cm of nitrogen pressure, resulted in the characterization of this compound as a trimer in which the (Ga-N)^ ring . is in the chair conformation (Figure 20). The mean dimensions, found were Ga-N 1.97, N-C 1.54, C-C 1.55A; N-Ga-N = 100°, Ga-N-Ga = 121, Ga-N-C = 116°; while the angles in the three membered rings were close to 60°. This structure, although confirming the predictions in part 1 concerning the configuration of the (Ga-N) ring, is somewhat unexpected in view of the cryoscopic molecular weight in benzene solution. The resolution of this apparent dilemma could be the following. It has been found that freshly dissolved samples of aziridino gallane, whether freshly prepared or not, give degrees of association of 2.55 to 2.65. Samples dissolved in benzene and stored for a few days give a degree of association of 2.00. Since the solid is trimeric, it would seem that the cryoscopic results indicate the gradual formation of dimer in the benzene solvent. It was also observed that a significant amount of insoluble material was formed on dissolving the solid. The following mechanism seems plausible: C1' C1 Structure of Aziridino Gallazane Figure 20 60 (Azir GaH2)3 (Azir GaH^ + (Azir'GaH ) (Azir GaH ) 2 x Thus the trimer gives unstable monomer which polymerizes, leaving the dimer in solution. Another possible mechanism appears to be the following: (Azir GaH2)3 (Azir GaH2)2 (Azir GaHn) 2 x where two competing rearrangements occur, one giving polymer, the other dimer. If the degree of association of gallazanes in benzene is not necessarily an indication of the association in the solid or neat liquid phase, possibly the neat nmr spectrum of isopropylamino gallazane (Figure 15) could be also rationalized in terms of a trans trimer configuration, in agreement with the observed intensity ratio of 2:1 for the . |3.-CH proton signals. The nmr spectrum of aziridino gallazane (Figure 21) shows only a sharp singlet, indicating a single isomeric constitution which is expected on the basis of a planar (GaN)2 ring with all hydrogens equivalent. Figure 22 shows the ^"H nmr spectrum of dimeric azetidino gallazane, with integrals of the two areas of resonance in the ratio of 2:1. The splitting observed is that expected on the basis of a planar (GaN)2 ring, a triplet for the four c< protons and a quintet for the two /3 protons. 1 1 1 1 1 1 1 1 1 1 1 ' 1 1 1 1 | 1 1 I 1 III! 1 1 1 1 [III 1 1 1 . 1 i i ' 1 1 ' 1 i ' . 500 1 ' I 400 1 1 1 1 • 1 | i 1 i. 1 1 1 1 300 1 'Ill' 1 200 • i 1 1 1 1 1 1 1 100 i I- i i 1 i 1 r i ) Hz J L I I I I I J L I I I l I I I I _I_J L I I I 2.84 ... 8.44 Fig.21 60Mc/s H n.m.r. spectrum of (CHANGahL in benzene solution >-H> J L <4> Figures 23 and 24 show the H nmr spectra of pyrrolidino gallazane (integration of 1:1 as expected for the four oc and four (3 protons) and piperidino gallazane (integration of 4:6 for ©C : f$ + o~ proton multiplets). The latter two spectra are no longer simple, with evidence of complicated spin-spin interaction. II 1 1 1 1 1 1 1 II 1 III! 1 1 1 1 1 1 1 1 i 1 1 I- 1 1 1 1 1 1 1 1 1 j ' 1 1 1 • 1 1 1 1 1 i 1 1 1 1 500 1 400 1 I. 1 1 I I 1 3( 1 ' 1 1 1 )0 1 1 1 1 1 1 » 1 200 1 i 1 1 III 100 1 1 1 1 1 c 1 )~H> II 1 II 1 II 1 II 1 1 1 1 i | II I I • 1 11 1 iii! 1 1 1 1 1 1 1 1 1 1 •I-I 1 1 1 1 500 1 i 1 l | II 400 1 1. | | 1 i 1 I II 1. 300 i ; 1 1 200 ! 1 1 1 Iii 100 1 1 1 1 1 I I i 1 Q Hz 2.84 \ 712 8.76 Fig.24 60Mc/s 'H n.m.r. spectrum of (CH2)BNGaH2 in benzene solution 66 Part 4 Imino Alazanes The reaction between ethylenimine (aziridine) and bis trimethyl amine alane was first attempted in 1962 by Marconi (19). These workers did not isolate the aziridino alazane product. A more recent discussion of this reaction product (20) suggests that ring opening of the aziridine ring occurs on solvent removal yielding an average degree of association of n = 10. The product prepared in this study gave initially the nmr spectrum of figure 25. Since this spectrum contains a high field triplet and evidence of a lower field quartet the previous formulation (20) of ring opening to give ethyl groups seems fairly conclusive. However, the spectrum a few hours later (Figure 26) showed an increased intensity of the high field triplet with respect, to the broad singlet for the aziridine rings. The following day, after storage at +5°CJ) :., the nmr spectrum showed the high field triplet to be even more intense than previously. These results indicate that ring opening occurs at a fairly steady rate at o o 5 - 25 C. The initial aluminum to active hydrogen ratio was found to be Al- H„ while the analysis of the same product left at room temperature for three days under dry nitrogen was found to be Al^ QQ H^ These results indicate that in the limit, complete aziridine ring opening could occur to give all N-ethyl groups in an insoluble polymeric product. It was of interest to see what the degree of association would » be if ring opening could be held to a minimum. Thus the degree of association of freshly prepared aziridino alazane was determined in 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 III | 1 1 I 1 | 1 1 II 1 1 1 1 1 1 1 1 1 1 1 . 1 . 1 1 1 1 500 1 . 1 1 II 400 M||l 1 3C I 1 1 )0 1 1 | 1 1 1 1 1 I-I 1 200 1 1 I 1 1 100 I 1 1 II 1 ) Hz Fig.25 6.0M.C/S 'H n.m.r. spectrum of (CH )2NAIH2 in benzene'solution Fig.26 60Mc/s 'H n.m.r. spectrum of (CH2)2NAIH2 in benzene solution , w .00 69 benzene solution with a minimum of delay. This worker was able to obtain a minimum value of n = 4.2 whilst a co-worker was able to obtain n = 3.14. These results suggest that the degree of association before ring opening sets in, is likely n = 3. This is.the expected degree of association in view of the results for the other alazanes of table 3. The product from the reaction of the bis trimethylamine alane with azetidine did not give up all its trimethylamine, some of which remained coordinated to it (Figure 27). Pumping at 0°C removed most of this trimethylamine to give the spectrum of Figure 28. The integration ratio is 2:1 for the or: <5 protons. The molecular weight is consistent with the formulation of this compound as a trimer. Similarly the 'Hi nmr spectrum of trimeric pyrrolidino alazane shows two areas of resonance in the ratio of 1:1. One resonance is centred at 1 = 7.10 and the other at T = 8.50. The piperidino alazane appears to be mainly dimer in benzene solution (n = 2.17) possibly resulting from the larger steric requirements of the piperidino ring. The nmr spectrum for this compound shows two resonances in the ratio of 6:4 at Tr = 8.61 and at = 7.18 respectively. These correspond to (&+ T~ and o< proton resonances. It appears that trimeric species are common with the imino alazanes and in this respect they differ from the dimeric imino gallazanes. Since the bond lengths of Al-N and Ga-N are known to be almost identical, the reason for this difference probably lies more in the nature of the reaction intermediate leading to these species than in steric or other effects. Possibly this difference is due to the relative ease with which aluminum can go 5-coordinate in the intermediate, but this is highly speculative as no mechanism utilizing a 5-coordinate aluminum has actually been demonstrated. 2.84 Fig 6.46 785 792 Fiq.27 60Mc/s *H n.m.r. spectrum of (CH2)3NAIH2 in benzene solution 72 Part 5 Imino Borazanes The fact that azetidino, piperidino and pyrrolidino borazane are dimeric in benzene is not surprising since boron-nitrogen systems generally prefer a monomeric or dimeric state to that of trimer. The nmr spectra of these three compounds are given in Figures 29, 30 and 31 and are all in agreement with the formulation of these compounds as having planar (B-N)^ rings and containing each a single isomeric form. The aziridino borazane prepared by the method of Akerfeldt (17) gave a singlet for the aziridino ring hydrogen in agreement with the literature (18) (Figure 32). The adduct, prepared by the Burg method (14) gave the nmr spectrum of figure 33 in agreement with the literature (18). The aziridino borazane has a trimeric constitution in contrast to the remaining members of this series. The reasons for this different constitution may be a result of the preparative route used to obtain the compound. II f ; • • i i i | i i i i 1 1 1 1 1 1 1 1 1 1 > 1 1 1 1 1 1 1 1 1 | 1 1 1 1 i | i | ' i I 1 " 1 i i 1 i i 500 1 400 1 i.l | M 1 . 3( Mill DO I MIM 1 200 MM 1 1 | | 1 1 1 1 100 I i 1 f I ) Hz 2.84 762 8.33 Fig;29 60Mc/s 'H n.m.r. spectrum of (CH2)3NBH2 in benzene solution CO i i i 1 1 i 1 1 i i —i—i—r i | i I i i 1 1 1 I III! i 1 1 i 1 7 ! • 1 1 1 1 500 i I i i [ i 1 | II i i i i | i I i i 1—1—| ! • i 1 ill 1 ' 400 1 3( Hr i | i .i • 1 1 I | i ' 30 1 I 200 1 1 I 1 1 i I r i 100 i I 1 III! 0 Hz Fig.30 60.Mc/s 'H n.m.r. spectrum of (CH2)4NBH2 jn benzene solution -4 1 1 1 1 1 1 1 1 1 i 1 1 1 1 • . 1 1 | 1 1 1 1 ) 1 | I i 1 1 1 I III | 1 1 I- i 1 I 1 1 ' 1 500 1 1 1 1 | I I 400 1 i. I | l 1 I i ! 1 300 1 1 1 1 1 1 1 I 200 1 1 1 I 1 1 1 1 1 100 l 1 1 i 1 I 1 0 Hi t 1 i 1 i 1 1 ! i 1 I 1 1 1 I i 1 1 1 I I 1 1 1 1 [ ! 1 1 1 1 llli Mil 1 i Ill; 1 | 1 - -1 1 I- 1 1 1 1 1 I 1 1 1 1 1 I 1 l| 1 1 1 1 1 II 1 1 1 1 1 III! ! 2.84 . 746 8.69 Fig.31 60Mc/s 'H n.m.r. spectrum of ((5hOft3H9 in benzene solution ~i—r 400 300 1 i 1 1 i I 1 1 1 l 1 1 1 1 I 1 ! j " 1 1 1 i 1 1 I 1 200 ! 1 'IM' 10Q 1 i i ! i I i . 1 OH s J L J__L .1. . . I L_ 9.63 l__l_L± I I ! I I I I ! I ! I Fig.32 60I\/Sc/s 'H n.m.r. spectrum of (CK,). N* BH! in benzene solution Part 6 Reactions of Imine Bases with EMe E = B, Al, Ga, In Reaction of EMe^ with aziridine gave, on methane elimination, compounds which were trimeric in benzene solution. Since the hydrido analogues previously prepared were trimeric as well, this result suggests that the groups about the E atom have little effect in determining the final degree of association of the complexes studied here. The compound Ke^B^(CE^)^ was not isolated, as the high temperatures necessary to achieve methane elimination from the adduct also cause polymerization. The nmr spectra of the two stable adducts Me^BNR^CH^^ (Figure 34) and Me^GaNH(CH^)^ (Figure 35) are characteristic but very different. The aziridine ring protons of the boron compound give rise to a singlet at low field - probably the result of nitrogen inversion or fast exchange reactions in solution. The higher field singlet is the resonance of the boron methyl protons. The aziridine ring protons of the gallium adduct, on the other hand, appear to be split into a multiplet. This complex splitting is believed to be the result of not only primary but also second order magnetic coupling of the hydrogen nuclei- on the aziridine ring,. ) . The three methane elimination products had simple and very similar nmr spectra. These spectra consisted of a lower field singlet for the aziridino protons and a high, field singlet for the methyl groups of the E atoms. rr— " r T 1 ! ' rr ': • i' ••I"' T 1 1 ( 1 1 I 1 1 1 1 1 1 1 1 I | 1. 1 ! ! 1 I- I 1 • i i • i. • 500 i 1 j ! i > 1 i ! 1 400 1 I 1 i ' i 1 -1 1 ' i i i 300 1 i I- 1 1 1 1 200 ' i 1 1 ! | i | 1 | I 1 100' i 1 i 1 c i Fig.33 6.0 Mc/s 'H n.m.r. spectrum of H3B-NH(CH2)2 in benzene solution 2.84 Fig.34 60Mc/s 'H n.m.r. spectrum of Me3B-NH(CH2)2 in benzene solution CO Bz 2.84 10.46 Fig.35 60Mc/s 'hTnmr. spectrum of Me3Ga-NH(CH2)2 in benzene solution 81 REFERENCES 1. N. N. Greenwood, A. Storr, and M. G. H. Vallbridge, J. Chem. Soc, 249 (1962) 2. E. Wiberg et. al., Z. Anorg. Chem,, 2J1, 114 (1943). 3. B..Siegel, J. Chem. Ed., 38, 496 (l96l). 4. N. N. Greenwood, and M. G. H. Wallhridge, J. Chem. Soc, 3912 (1963). 5. N. N. Greenwood and A. Storr, J. Chem. Soc, 3426 (1965). 6. H. Bretsacher and B. Sifigel. JACS, 87, 4255 (1967). 7. A. E. Finholt, A. C. Bond, and H. I. Schlesinger, JACS, 69, 1199 (1947) 8. E. V/iberg and H. Noth, F. Naturforsch, 63 (1957) 9. E. V/iberg and H. Heule, ibid, 126, 576 (1952). 10. N. N. Greenwood, New Pathways in Inorganic Chemistry, Camb. U. Press, 1968, P37. 11. -N. N. Greenwood, A. Storr and M. G. H. Vallbridge, Inorg. Chem., 2, 1036 (1963) 12. N. N. Greenwood, E. J. F. Ross, and A. Storr, J. Chem.; Soc.-(A), 706 (1966). 13. A. Storr, J. Chem. Soc.(A), 2605 (1968) 14. A. B. Burg and C. D-. Good, J. Inorg. Nucl. Chem., 2, 237 (1956) 15. S. Akerfeldt et. al., Acta Chem. Scand., 23, 115 (1969) 16. H. Ringertz, Acta Chem. Scand., 23, 137 (1969) •17. S. Akerfeldt and TI. Hellstrom, Acta Chem. .Scand., 20, 1418 (1966) 18. R. L. Williams, Acta Chem. Scand., 23, 149 (1969) 19. W. Marconi et. al., Gazz. Chim.Ital., 92, 1062 (1962) 20. R. Ehrlich et. al., Inorg. Chem., 628 (1964) 21. E.J. F. Ross.,Ph. D. Thesis, 1965, University of Newcastle-T.Tpon-Tyne 22. "Handbook of Chemistry and Physics" 49 th ed. Chemical Rubber Co., Cleveland, 1968. 23. N. N. Greenwood and K. Wade., J. Chem. Soc, 1527 (1956) 24. H. Wenker, JACS, 57, 2328 (1935) 25. P. Leighton et. al., JACS, 69, 1540 (1947) 26. W. Reves et. al.,JACS 73, 3522 (1951) 27. P. Leighton et. al., JACS, 75, 2505.(1953) •' • -^.iW \\-:.-\. ;:,Mfe 28. J). G. Gains and L. Schaeffer, JACS, 85; 395 (1963) 29. J. K. Ruff and K. F. hawthorne, JACS, 82, 2141 (i960) 30. F. G. A. Stone, Chem. Rev., 58, 101 (1958) 31. B. Allred and F. Rochow, J. Inorg. Nucl. Chem., j>, 264 (1958) 32. 0. T. Beachley and G. E. Coates, JCS, 591 (1965) 33. M. P. Brown, R. V/. Heseltine and L. H. Sutcliffe, J. Chem. Soc (A),6l2 (1968) 34. J. L. Atwood and G. D. Stucky, JACS, 92, 285 (1970) 35. G. E. Coates and K. V/ade Organometallic Compounds: Methuen, London 1967 vol. 1., p 307 36. E. Campaigne, II. F. Chamberlain, and B. E. Edwards, JOC, 27, 135 (1962) 37. A. Segre and.J. Musher, JACS, 89, 706 (1962) 38. J. L. Jungnickel and C. A. Rielly, Nuclear Magnetic Resonance in Chemistry, ed. B. Pesce, Acad. Press, 1965, P 83 39. N. Ti. Greenwood, E. J. F. Ross and A. Storr, JCS (A), 706 (1966) '40. N. N. Greenwood, E. J. F. Ross and A. Storr, JCS, 1400 (1965). 41. T. N. Marfulis and M. Fisher, JACS, 89, 223 (196?) •4.2.. E. Adman and T. N. Margulis,. JACS, 90, 4.517 (I968) 43. M. P. Brown, R. V/. Heseltine, P. A. Smith, and P, J. Walker, JCS (A), 410 (1970 44. P. L. Corio, J. Chem. Ed., 46, 345 (1969) 45. W. Harrison, A. Storr and J. Trotter, Chem. Comm., submitted for publication. 


Citation Scheme:


Usage Statistics

Country Views Downloads
Japan 6 0
United States 3 0
China 3 11
City Views Downloads
Tokyo 6 0
Beijing 3 0
Ashburn 2 0
Sunnyvale 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items