University of Southern California Dissertations and Theses

A study of the structures and properties of triaryl compounds of boron

(USC Thesis Other)

A study of the structures and properties of triaryl compounds of boron

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A STUDY OP THE STRUCTURES AND PROPERTIES OP TRIARYL COMPOUNDS OP BORON A Dissertation Presented to the Faculty of the Graduate School The University of Southern California In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy by Carl W. Moeller, Jr. ; August.1954 UMI Number: DP21773 All rights reserved INFO RM ATIO N TO ALL U SER S The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation Publishing UMI D P21773 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 -1 3 4 6 T^O C ' 5 S * /V)<W This dissertation, w ritte n by Carl tf. Moellerf Jr. .... 'P v \? under the direction ofU.i,3.Guidance Com m ittee, and approved by a ll its members, has been p re sented to and accepted by the F a cu lty of the Graduate School, in p a rtia l fu lfillm e n t of re quirements fo r the degree of D O C T O R O F P H I L O S O P H Y Date. .......... C U & . Dean % 0 .. G uidance C om m ittee _ Chairman . R . 3 . ........... ...... .... 3 ^ ffjt A 3 Q 4 > d 4^6 ACKNOWLEDGEMENT To Professor W. K. Wilmarth for his encourage- 1 raent and stimulating suggestions during the course of this research. TABLE OF CONTENTS CHAPTER PAGE I. INTRODUCTION TO THE PROBLEM ........ 1 Statement of the problem . ............. 1 Discussion of previous research In the field . . ........... 3 II. EXPERIMENTAL PROCEDURES .......... 6 Preparation of the compounds............ 6 The triarylboron compounds ....... 6 The monosodium salts ........ 7 The disodium salts .......... 11 The monosodium salt of TNB in tetrahydrofuran ............ 13f Analyses ........ . 13 Magnetic susceptibility studies ...... 15 Absorption spectra in ether and tetrahydrofuran . 20 Molecular weight measurements .......... 22" Ortho-parahydrogen conversion ...... 29 Nitric oxide absorption ......... 30 III. OBSERVATIONS ........................ 31 Magnetic susceptibility measurements . . . 31 Absorption spectra ............ 33 Molecular weight measurements ...... 46 ±v CHAPTER PAGE Ortho-parahydrogen conversion ......... 48 IV. DISCUSSION OP THE RESULTS ........ 52 Structure of the dimer ........ 52 Steric effects..................... . 56 Resonance effects . ........... 57 Solvent effect...... ................... 60 The two step reduction of the triarylboron compounds .............. 62 Comparison of the spectra of the isoelectronic compounds ....... 65 Summary . .. .. .. .. . . . . . . . 66 BIBLIOGRAPHY . ................. 70 APPENDIX ................. * • 75 TABLE I. II. III. LIST OP TABLES PAGE Magnetic Susceptibility D a t a ............ 32 Molecular Weights of the Triarylboron Compounds . ......................... ^9 Parahydrogen Conversion by the Sodium Salts of Triphenylboron ........ 5© LIST OP FIGURES FIGURE PAGE 1. Dilution Apparatus for Absorption Spectra, with Preparation Vessels ......... 9 2. Apparatus for Filling Magnet Tubes with Solid - ...... ~ - . ' . . . ... . . . 18 3- a. Molecular Weight Apparatus ; (Front View) . 25 b. Molecular Weight Apparatus (Side View) . 26 4. Spectra of Triphenylboron in Ether, A, and the Same Solution after Three Weeks Exposure to Air, B ........ 3^ 5. Spectra of TNB In Ether, A, and the Same - Solution after Three Weeks Exposure to Air, B ............. 35 6. Absorption Spectra of (NaTFB)2 in Ether (Curve A) and (NaTPB)g in THF (Curve B) . . 36 7. Absorption Spectra of (NaTPB)2 in THF Prepared from Amalgams........ ... 37 8. Absorption Curves for (NaTNB)2 in Ether at Several Concentrations ......... 38 9. Absorption Spectra of NagTNB in Ether (Curve A) and of Na2TNB in Tetrahydrofuran (Curve B) ........................... 39 3 vii | FIGURE PAGE 10. Absorption Spectra of NaTNB in Tetrahydrofuran, Curve A, and of Sodiura-Naphthalene in Tetrahydrofuran, Curve B ........ 40 11. Absorption Spectrum of (NaTFB)g with Excess Sodium Boron Concentration About 1 x 10"2 * M. ........... 41 12. The Change of Optical Absorption with Change in Concentration for (NaTNB)g in Ether, Measured at 42© mu. Data from Figure 7 • 42 13. Calibration of Molecular Weight Apparatus with Benzil ........ ............ 4? CHAPTER I INTRODUCTION TO THE PROBLEM Statement of the Problem Although the, triarylmethyl radicals have been ex tensively studied to measure the extent of their associa tion into dimers, 2 Ar^C* Ar^CCAr^ (1), and to determine the effect of sterlc strain In the dimer and resonance stabilization of the radical on the associa tion constant,1 there has been little correlative research on the neighboring isoelectronic molecules. Of particular interest is the replacement of the central carbon atom by boron to form Ar^B” or by nitrogen to form Ar^N+ and the comparison of these compounds with the arylmethyl radicals. The only study reported on the triarylnitrogen com pounds Is the work of Rumpf and Trombe who found by mag netic susceptibility measurements that (jo-tolylJ^NCIO^ was completely dissociated into radicals In the solid state.2 . i 1 G. W. Wheland, Advanced Organic Chemistry (New York: J. Wiley and Sons], IMf3; W. A. Waters, The Chem istry of Free Radicals (London: Oxford Press, i9#8). 2 P. Rumpf and F. Trombe, Compt. rend., 206, 671 (1938). This may be compared with the tri-j>-tolylraethyl radical which is 84 per cent associated to dimer in 0.1 M. benzene solution.^ Since even the simple tritelyInitrogen cation shows no association and the triphenylnitrogen cation can not be isolated due to the reactivity of the para-hydrogen atoms, it seems unlikely that any study of the effects of substituents on these compounds will succeed. Therefore, attention has centered on the arylboron compounds and, in order to observe the greatest changes in sterlc and resonance effects, the studies have concentrated upon trlphenylboron, TFB, and tri-oc -naphthylboron, TNB, and their monosodium and disodium salts. The monosodium salts have been found to exhibit variable dissociation into free radicals. The physical measurements used in this stuty Included magnetic susceptibility measurements on a Gouy- type apparatus to detect any radical present at high con centration of solute, absorption spectra of solutions at low concentrations and molecular weight determinations. Because of the extreme reactivity of these compounds with air and water, special vacuum techniques were devised to ! prepare the compounds and to Isolate them for the measure- i i ments. ^ C. S. Marvel, J. F. Kaplan and C. M. Himel, J. Am. Chem. Soc., 6&, 1892 (1941). Discussion of Previous Research in the Field As early as 1880 Mlchaelis staled the reaction of boron trichloride with diphenyl mercury and obtained the 4 monophenylboron and diphenylboron compounds; however, It was not until 1922 that Krause and Nitsche bubbled BF^ into an ether solution of excess CgH^MgBr and, after re- fluxing, isolated B(CgH^)^ by a distillation process.-* Modifications of this Grignard type reaction have led to the successful preparation of several other triarylboron compounds including B( oc-naphthyl)^, B(benzyl)^, Bftolyl)^. and B(anlSJrl)3. Since these aromatic compounds of boron are unstable in air or moisture, they are handled in vacuum systems or under dry nitrogen. The reactions of the alkali metals with triaryl boron compounds was first described by Krause who showed by analyses that the one electron reduction product could be obtained:^ Na + BAr3 NaBAr^ (2). The reversibility of this reaction and the conductance of the sodium salts in ether were investigated by Bent and * A. Mlchaelis, Ber., 27, 244 (1894); A. Mlchaelis, Ann., 315, 19 (19©1). ^ E. Krause and R. Nitsche, Ber., 55B, 1261 (1922). 6 E. Krause, Ber., 57# 216 (1924); E. Krause and H. Polack, ibid., 597777 US . . 2 _ Dorfman who obtained an equilibrium constant of about 10 for £) and found that the salts behaved like weak electrolytes in ether.7a,b A complication is introduced in the preparation and the study of these salts by their ability to undergo further reduction to . The salts are all colored and react rapidly with loss of color when exposed to air. Krause and his coworkers studied the acid-base reactions of the triarylboron compounds* e.g.* BAr3 4 - NH^ — ^ H3NBAr3 (3) and found them to be considerably less reactive with water and air in these compounds where the boron octet has 8 been completed. They investigated the molecular weights of these substances and of the triarylboron compounds. They found them to be simple monomers.^ Prior to the present study no molecular weight studies had been made on the salts* and there were no spectral or magnetic data available. Therefore* after ' a. H. E. Bent and M. Dorfman* J. Am. Chem. Soc.* 57* 1259 (1935); b. H. E. Bent and M. Dirfi5h,“lMd.7~57, 1524 (1935). 8 E. Krause* Ber.* 57* 813 (1924); E. Krause and H. Polack* ibid., 61, 271 (193EDj E. Krause and P. Nobbe* Ibid., 62T934 TT930). 9 s. Krause and P. Dittmar, Ber., 63, 2347 (1930). investigations of some anomalies in the preparative chemistry of the alkali metal salts had been completed, the major portion of the research was devoted to these physical measurements. CHAPTER II EXPERIMENTAL PROCEDURES I i i Preparation of the Compounds I i The triarylboron compounds. Initially triphenyl- | r 6 1 boron was prepared by the method of Krause. * Later It I ! was found convenient to make large amounts of trimethyl- ammonium tetraphenylboron, (CH^J^NHBfCgH^)^, and to prepare the triphenylboron from this very stable, easily stored 10 compound. The method used was that of Wittig. The (CH3)3NHB(C6H5)4, which was prepared from (CH^NHCl and LiBlCgH^)^, was thermally decomposed to (CH^^N, benzene, and BtCgH^)^. The amine and benzene were trapped in liquid nitrogen, and the triphenylboron which remained behind as a solid was sublimed under partial vacuum, 12 mm. of Hg. After being recrystallized three times from ether under anhydrous nitrogen gas, the B(CgH^was stored in sealed bulbs. The yields obtained by using the first direct method were low and usually about 20 to 25 per cent of the anticipated recrystallized product was obtained. The second method gave over-all yields of the recrystallized product which were 60 to 70 per cent of that anticipated. The melting points which were obtained in evacuated 10 G. Wittig and P. Raff, Ann., 573, 195 (1951). __J . _ r capillary tubes were in the range 142-144° (Reported: 142°). The tri-oc-naphthylboron was prepared according to the method of Brown and Sujishl and recrystallized from 11 benzene. The final recrystallized product was obtained in 40 per cent yield. The melting points (evacuated capillaries) were in the range 204-206°. (Reported: 206- 207° by Brown and Sujishij11 203° by Krause and Knobbe.^) Brown and Sujishi discuss the possibility of obtaining two Isomeric structures of this compound and, using their terminology, it is considered possible that the unsym- metrical molecule was obtained. This is concluded from the melting points and the observation that the product crystallizes with one mole of benzene per molecule of tri- o( -naphthylboron. This is discussed in more detail in the last chapter. The preparation of tri -£-b ipheny lb or on, trlxenylborcn , by a lithium interconversion method is described in Ap pendix A. The monosodium salts. In these preparations the complete vacuum apparatus used depended upon the particular physical measurements to be made, but in general the 11 H. C. Brown and S. Sujishi, J_. Am. Chem. Soc., JO, 2800 (1948). 8 ! I apparatus consisted of two flasks plus the magnetic sus- ! eeptlbility tubes, spectrophotometer cells, molecular i ; weight apparatus or any other apparatus needed for the ! physical measurements. These were attached to the vacuum j line and the apparatus was dried by sparking and flaming i | the glass while evacuating. Then Ng gas, which had been I dried by passing through drlerite and glass beads coated with P205, was admitted and the process of evacuation and flaming was repeated. With nitrogen gas streaming through the apparatus, a sealed capsule containing a weighed | amount of triarylboron compound was put into flask B; see | I Figure 1. The capsules were prepared by filling the small ' glass bulbs with the triarylboron compound in a nitrogen- filled drybox. The capsules, which had been weighed before filling, were reweighed, evacuated and sealed. The solvent i and excess triphenylboron (hereafter referred to as TFB) i were added. The solvent had been dried over and then distilled from sodium metal or LiAlH*; the middle fraction 12 was used. Finally, sodium metal was added through the j neck of B. The sodium metal was either pressed in under 1 N2 or capsules, which had been filled by breaking a thin i < glass bulb of an evacuated tube under molten sodium, were , - i | Because of the presence of peroxide inhibitors ! in the tetrahydrofuran, it was found necessary first to j • treat this solvent with NaOH, then to distill and store j : it over LiAlH^. j Front View View FIGURE 1 DILUTION APPARATUS ABSORPTION WITH PREPARATION added to the side arm e. The vessel was evacuated through v and, after the evacuation was completed, the entire ap- | paratus was sealed from the vacuum system. The sodium was \ j melted and forced through the pinholes into B. When the ! solution had warmed to room temperature, the apparatus was : placed on a~mechanlcal shaker to speed up the heterogeneous i reaction. The solution became golden around the sodium upon contact and after shaking it soon became deep yellow throughout. This solution was used as a wash solution and, after all of the glass surfaces had been washed, it was poured into flask C. The solvent was distilled from C several times to rinse the apparatus, and finally all of the solvent was distilled to B and frozen in liquid nitrogen. Flask C which contained the residue of the wash solution ' was sealed off and removed. The necessity for this care- ? ful washing and rinsing is discussed In a later section. The dimensions of B depended on the concentration desired. In some cases tube B contained a round flask Of known j volume as shown. ■ The capsule of TFB in flask B was broken by gentle I • i I shaking of the apparatus or, in those cases where shaking j | was undesirable, a glass-enclosed bar magnet was sealed I to the capsule. This was then raised by another magnet j 1 __outside__of_the_apparatus_and_allowed to drop through .the_J j the neck of flask B. The TPB from the capsule reacted * with the shining, silver-colored sodium. The product was i i either studied as the yellow-orange solution or it was ' i I crystallized from ether (the solubility is 0.8 g. per liter I • j of ether at 20° C.). Both solids and solutions react | rapidly in air with considerable evolution of heat. ! I 1 ! The preparation of the monosodium salt of tri- °<- naphthylboron (later referred to as TUB) was similar to that described above except for some difficulties due to the easily formed NagTNB as described below. The disodium salts. Bent and Dorfman had shown that the black sodium salt of TNB actually contained two lii equivalents of sodium per boron atom. No one since then seems to have Isolated a monosodium tri- oC-naphthylboron compound or to have prepared the disodium salts of other t i triarylboron compounds. It was found that, if an ether solution of the black NagTNB reacted with an equivalent amount of TNB, the red-orange monosodium salt was formed | ^ little difficulty was encountered in breaking ; these capsules which were prepared by pulling 4 mm. pyrex tubing to a point, blowing a bubble at the tip, and then | collapsing the bubble slightly so that it was no longer \ spherical. A couple of crossed scratches were then made i on the surface with a porcelain chip. I ! H. E. Bent and M. Dorfman, J. Am. Chem. Soe., 54,! : 2133 (1932). I ; . and could be crystallized from ether.^ NagTNB + TNB - — *• (NaTNB)g (4) black colorless red-orange The disodium salt, which forms rapidly in ether, tetra hydrofuran, or dimethoxyethane, does not form In benzene. In this nonpolar solvent only the monosodium salt is ob tained even after months of standing in contact with sodium. With this solvent effect as a hint, the preparation of NagTFB was attempted in the very polar solvent tetra hydrofuran (referred to hereafter as THF). Previous work ers had reported a green color appearing after long shaking of TPB with sodium amalgams and they Interpreted this as impurities.^ The effect was found to be more pronounced with tetrahydrofuran solutions, and when sodium metal was I used the solutions became deep blue after two or three days of shaking. Within about ten days the solutions be came red even when Isolated from the sodium and kept in the dark. The analyses for sodium were always lower than anticipated as discussed below. Attempts to crystallize NagTFB by removal of the tetrahydrofuran from the blue ^ The formula (NaTNB)2 is used since the investi gation has disclosed that the salt dimerlzes in ether. This is discussed in ddtail below. solutions led to decomposition and appearance of thin films. ' of sodium. ; i • i * i The monosodium salt of TNB In tetrahydrofuran. It ^ was found that the reaction of sodium with TNB In THF did . 1 ■ i ! i not produce the Initial red-orange solution which was ob- j 4 , talned in ether, but instead a green solution which \ i analyzed one mole of sodium per mole of boron was formed. j On further reaction a black solution containing the 2:1 ratio of sodium to boron formed. This black solution gave an absorption spectrum similar to that found for NagTNB In ether, except for shifts in the maxima apparently due to solvent effect. The green monosodium tri--naphthylboron solution was obtained either by reaction of a solution of NagTNB with an equivalent amount of TNB solution (using break-off seals in the vacuum apparatus) or by dissolving I the red-orange (NaTNB)g, crystallized from ether, into THF. The reaction between NagTNB and TNB was rapid. Upon re moval of THF the solid residue was red-orange in color and when dissolved in ether the solution gave the original i 1 spectra of (NaTNB)g. i ! Analyses. The purity of the TEB and the TNB were ‘ checked by melting point determinations and boron analyses* I ! : The melting points were studied in evacuated capillaries to_avoid._reac_tions_with,air. Some samples were analyzed j j for carbon and hydrogen by Joseph Pirie of the University I of Southern California mieroanalytieal laboratory. The i following is a typical analysis. Calculated for TPB: C, j i . | | 89.29; H, 6.24. Pound: C, 89.03; H, 6.57. The alkali ! metal salts were analyzed for metal and boron only. The i reaction vessel was opened to the air, and the solvent was removed by using a vacuum pump and trap. The white residue was dissolved in 3 to 5 ml. of concentrated HgSO^ by warming gently to form a red solution. After cooling, this was poured into a small distilling flask and the reaction vessel was rinsed with cold methanol. Then 50 ml. of methanol were added to the distilling flask, and the methyl borate was distilled until the residue no longer gave a positive boron flame test. For the usual sample size, less than 1 per cent of the boron could have 1 been detected in the residue. The methyl borate in i methanol was hydrolyzed in the receiver with GOg free water and titrated with standard base. Mannltol was added and the phenolphthalein endpoint used. The residue in the distilling flask was rinsed into a platinum crucible;, warmed gently to distill any remaining methanol, then more i | strongly to fume the HgSO^, and finally heated to redness, j | The amount of sodium was calculated from the weight of j | NagSOj^ obtained. In a typical analysis of NagTNB, 0.442 j Lmillimole._o£_TNB_reacted_with_ sodium, and the analysis_____ j j showed 0.437 millimole of boron and 0.880 millimole of 1 I sodium in the product3 cumulative errors in the last two ! ^ ' 1 numbers are about 1 per cent. Analyses for sodium using 1 t ! the flame photometer attachment on a Beckman model DU I ■ : spectrophotometer agreed well with those made by the method! » above, and this method was used for analyzing the smaller j 1 quantities from the spectra studies. No samples of the blue solutions obtained by the reaction of NaTEB with excess sodium in THF gave Na:B ratios as large as 2:1; most of the analyses indicate a ratio of about 1.30 to 1 with a spread of 5 per cent. 1 Magnetic Susceptibility Studies Magnetic susceptibility measurements using the Gouy 1 1 method, with field strengths variable up to 15*000 oersted, were performed on an apparatus similar to that described - 1 ^ , by Selwood. Apparent changes in sample weight with the field off and on were measured to .01 mg. on an Ainsworth balance (l^pe TCX). The magnetic susceptibility sample tubes which were used for the solution studies had an internal diameter of 0.80 cm. and the sample height was usually 11 cm. For the solid samples, tubes of 0.30 cm. i I ! and 0.45 cm. internal diameter were used. The lower por- | —— — '" i j 16 / I F. W. Selwood, Magnetochemistry (New York: r Interscience Publishers, 1943). i j tions of the tubes were glass rod to compensate partially for the diamagnetism of the tube and to prevent swaying in ! the magnetic field. I j The magnetic susceptibility depends on the amount j of matter, and in general either the susceptibility per j gram, X or the susceptibility per mole, Xjfl* is de termined. The molar susceptibility of a nonferromagnetic i substance can be expressed: X H - A + N2pS/32/3HP (5) | where N is Avogadro’s number, p is the magnetic moment per molecule, ft is a factor which converts the moment to Bohr magnetons, R is the gas constant and T is the abso- lute temperature. A is the diamagnetic term and it can be calculated as the sum of the atomic susceptibilities of all the atoms in the molecule. However, correction terms I must be applied for multiple bonds, aromatic ring atoms and other structural factors. Since the values for gram —6 atomic susceptibilities range around -10 x 10“ to -30 x 10“ esu per gram-atom, the values of A for the ! z T ! triarylboron compounds are several hundred times -10” i j esu per mole. j i Using the expression j I p * / n(n 4- (6), ! i ! which is derived from quantum mechanics, one can insert ! _ “ IT the value one for n and solve for ji. Here n Is the number of unpaired electrons per molecule. The value of M obtained can then be substituted In the second term of equation (5), and one can obtain the magnitude of this term at any temperature for a molecule containing one un paired electron. At 20° 0. the value Is 1260 x 10”^ esu per mole. The values of Xg or /(jvj are obtained by weighing the sample with the field off and then observing the ap parent change In weight, z* w, when the field is on. The 16 expression -used Is: y _ 2 * B . x A w x i (?) s w x H / where g is the acceleration of gravity, w is the sample weight, £ is the sample length and H is the field strength. The field strengths, which could be varied by changing the amount of current passing through the magnet coll, were calibrated by using degassed water in sealed sr i 16 6 tubes. The value -.720 x 10“ esu per gram was used for the magnetic susceptibility of water.' The solutions were prepared in the manner described above. They were poured into the magnet tube, frozen in liquid nitrogen and sealed off. The solid sodium salt was prepared in the glass vacuum apparatus shown in Figure 2. FIGURE 2 APPARATUS FOR FILLING MAGNET TUBES WITH SOLID • ® Flasks C and B are 200 ml. round-bottom flasks. After a I saturated solution of the compound had been prepared in B, | It was poured into A. Ether was partly removed from the j i solution by cooling B in an ice bath and the salt began to ; i precipitate. Finally A was cooled to 0° C. and the liquid \ i was poured back to B to dissolve more of the salt. This ! procedure was repeated until all of the salt had been recrystallized in A. The last traces of solvent were re moved from the crystals in A by cooling B in liquid nitrogen and finally warming A to 100° C. The flask B was removed at a and the crystals were put into the magnet tube. They were added a few at a time by gentle tapping and then they were packed into the tube by tamping with the glass rod, which was sealed to a glass-enclosed bar magnet as shown. This could be manipulated with a magnet 1 outside of the apparatus. After the tube had been filled i and the sample packed down, the tube was sealed and re moved from the rest of the apparatus at d. The internal diameter of the tube was 3 mm.; the rod was 2.5 mm. in diameter. The temperature studies were made in a special i j Dewar flask which was fitted between the poles of the magnet. A dry nitrogen atmosphere was maintained in the Dewar. The magnet tube was suspended in the center of an . I I Laluminum.block which_ was injfche Dewar. The block could___ I ! be cooled to a desired temperature by controlling the rate ^ of flow of cold nitrogen gas (from liquid nitrogen), which | circulated through coils in the block. The higher tempera- I tures were obtained by electrically heating the block. ! i ! ! Over the temperature range studied, the relay system, which1 I I was used to control the nitrogen flow or the electric cur- , rent, kept the temperature constant to * 0.1° C. The temperatures were measured on a Leeds and Northrup potentiometer using copper-constantan thermocouples which were located in the Dewar. I Absorption Spectra in Ether and Tetrahydrofuran The spectra were obtained by using a Beckman model DU spectrophotometer with the samples sealed in corex cells. The sodium salts were found to be partly decom posed by the water driven from the glass during the ! sealing. The samples, which were frozen in liquid nitro gen while being sealed, remained colored as the cells were sealed off, but upon warming to room temperature the color faded. Over a period of days the maxima in the | absorption spectra slowly dropped in value, and this ef- i | feet was independent of whether the sample was stored in ( the dark or light. Similar observations were made by j j Bent and his coworkers during their work with NaefCgH^)^ j 1 ' ' 21 ■ and the alkali metal ketyls.1^ Therefore, to avoid the sealing of Individual cells, a sealed dilution apparatus was used; see Figure 1* By using this apparatus larger ; volumes of more concentrated solutions could be sealed ! i off, and less than 10 per cent of the salt was destroyed | i by the water which escapes from the glass when It Is fused. [ i This estimate is based on calculations resulting from a study of the loss of optical density during the sealing of individual cells containing different concentrations of solution. The entire apparatus was first washed with a solu tion of the sodium salt from which the solvent was later distilled as described above. After the sodium salt, which was to be studied, had been prepared from a weighed amount of boron compound in a measured amount of solvent, 1 it was poured into A. The solution was cooled in liquid i nitrogen and the dilution apparatus was sealed and removed from the rest of the vacuum apparatus at a. At room tem perature the solution was then poured into the cell and readings were made on the spectrophotometer. The entire dilution apparatus was sealed from external light by a black cloth. After the solution had been studied at the ^ H. E. Bent and G. J. Lesnich, J. Am. Chem. Soc., I 1 57, 1246 (1935); H. E. Bent and H. H. Irwln7~Jr77~Tbi^T7 | j g, 2072 (1936). ; initial concentration, the vessel was tipped and the solu- I { tion was poured back to A through the capillary. A fixed, j predetermined amount was retained in b around the capillary. | This solution was then poured into the cell, and the | solvent was distilled to a calibrated mark on b. The ab- i sorption spectrum was studied at the new concentration. j The dilutions were repeated four to six times with spectral readings taken at each concentration. Tubing £ made it possible to wash and rinse the cell initially; also the most dilute solution could be poured completely back to A, and the spectrum of the concentrated solution was re measured to check for decomposition which might have oc curred during the measurements. The sldearm d contained a break-off seal, which was used in some cases to allow another solution to react with that in A. Dilutions with i known solutions, as well as repeated runs with unknowns, showed that the concentrations could be duplicated to ± 1 per cent. Molecular Weight Measurements The initial attempts to study the molecular weights 18 ■ by isothermal distillation techniques proved unsatls- 1 factory due to decomposition during the lengthy time i | l , - - - r T ^ r - , nr„ r i i r j „ | 18 R. Signer, Ann., 478, 246 (1930). | | ' " ' " ' 23 j required to reach equilibrium in the systems. Therefore, i a thermoelectric molecular weight apparatus incorporating i i some of the features of those previously described1^ was designed and built. The temperature sensing elements were ! two Western Electric type 14A thermistors, which had re- t sistances of 99>800 ohms and 100,300 ohms. These were j i used in two arms of a Wheatstone bridge system and balancedj against two 100,000 ohm resistors and a variable resistance box. The eleetrical system included a 5 volt, 25 cycle, AC input signal, which after amplification was detected on a Dumont model 241 oscilloscope. A null method was used; the bridge was first balanced with the two thermistors dipping into the solvent and then, with one thermistor still in solvent, the other was placed in the solution being studied. Some of the measurements were made by 1 using a DC bridge system with the potential furnished by a three volt battery and detection by a sensitive galvan ometer, sensitivity 6 x 10“^ amps./ram. With either bridge system, balanee could be detected to about * 10 ohms. It was found necessary to activate the thermistors for | about two hours to obtain steady readings. i I Due to the reactivity of the solutions it was j necessary to design a completely sealed, evacuated vessel I , \ I ....... ......* | 19 S. B. Kulkami, Nature, 171, 219 (1953); R. H. —Muller—and-H.- J-.—Stolten,-Inal. Chern.,_ 25, 1103—(1953). - J for preparing the solutions and for measuring the molecular weights of the solutes. As shown in Figure 3, the apparatus consisted of two 200 ml. round-bottom flasks joined through their necks. The thermistor beads rested in small cups which could be filled with solvent or solution and which could be emptied by rotating about the a,b axis. Tube < c led to a calibrated tube (not shown),- which contained the solution when the thermistors were being balanced with solvent in A and B. The calibrated tube was connected to the rest of the vacuum system,where the solutions were pre pared as described above. The thermistor leads were spot- welded through nickel to tungsten leads, which were sealed in glass. During the measurements the apparatus was sub merged in a water bath at 25.00° C. with fluctuations of i .003° G. Of course, one of the advantages of this method of molecular weight measurements is the possibil ity of making studies in different solvents and at various temperatures in each solvent. The two thermistors will be at the same temperature when both dip in the solvent, but if the thermistor in B, Tg, dips in the solvent and thermistor TA dips in the solution, then Tgwwlll cool due to evaporation and TA will warm due to condensation. These effeets will depend on the rate of evaporation, which in turn will depend on the _cone.entration_of_the—solut±on_as_well_as_other_faotorsJ__ 25 Glass Joint sealed with vamo Glass Tubes shield leads a QD CD Front View FIGURE 3a MOLECULAR WEIGHT APPARATUS a Side View FIGURE 3b MOLECULAR WEIGHT APPARATUS The thermistors have a negative temperature coefficient of F resistance, and the resistance of Tg increases, while that i of T. decreases. By measuring the change in the bridge j I resistance at the null point, AR, one can calculate the temperature difference of the two thermistors. After calibration of the apparatus this temperature difference, or the resistance change, can be used to calculate the concentrations of solutions. For the measurement of A R, the apparatus was placed in the bath at such an angle that the thermistor cups touched the main body of solvent in A and B. When the thermistors had reached the bath temperature, the bridge ( was balanced and the resistance was recorded. Then the solvent in A was put into the calibrated tube, and the capsule of solute was broken. After the solution was pre- 1 pared, it was poured into A. When the apparatus had reached i temperature equilibrium with the bath, the bulbs A and B were rotated about the a,b axis so that the cups were no longer in contact with the liquid. It was necessary to control the amount of liquid in A and B in order that when | the thermistors were vertical the liquid did not touch the I cups. Then the bridge was again balanced and the re- i | sistance change of recorded. Readings were taken about | every 5 minutes for one hour:;. 1_______The apparatus was^calibrated with benzil as the __ | solute In ether and THF.20 The calibrations were checked I j with azobenzene and tri- c<-naphthylboron. In the calibra- i j tions it was found that the results could be expressed as i | AR = kN, where AR is the resistance ehange when Tf l is I • dipped into solution instead of into solvent, N is the ■ 1 ! mole fraction, and k is a constant which varies with i solvent— and probably the dimensions of the apparatus. For ether k is 23,500 ohms and for THF k is 25,700 ohms. -2 At concentrations above H a 2.5 x 10 , the calibration curve showed considerable deviation from linearity. This may be due to aggregation of the solute or due to con siderable distillation occurring during the time required to reach a steady state. However, no unknown solutions were studied at these relatively high concentrations. It was found that a steady state was reached in i about 30 to 60 minutes. The over-all reproducibility was about ± 20 ohms. This was due in part to the lack of sensitivity in the detecting instruments and in part to slow changes in the solvent-solvent balance point, which i could not be as readily checked as in the apparatus of 19 more conventional design. , . * i ; 20 i j This has been demonstrated to form ideal solu- j : tions by boiling point elevation studies at various con- \ i centrations. E. Beckmann, Z. physik. Chem., 63, 197 1 ( 1908) . ! i . Oirtho-parahyflrogen Conversion i ! The apparatus described by Wilraarth and Baes was used to measure the rate of parahydrogen conversion by l pi the arylboron compounds. The techniques and theory are i | discussed by them. One refinement was the addition of a I side a m to the neck of the reaction vessel between the stopcock and the flask; to this was added the flask in which the sodium and TEB reacted. The solvent used was THF. When the reaction was completed, the solution was poured into the ortho-parahydrogen reaction vessel and the j i side a m was sealed. The vessel was attached to the vacuumj line and the solution was flushed with normal hydrogen be- < fore the parahydrogen enriched gas was pumped in. The solubility of hydrogen in THF, which had been dried over LiAlH^ and distilled at 66° C. (middle fraction)] was measured in the jacketed reaction flask. The vessel was evacuated and flushed with normal hydrogen; then the hydrogen absorption was determined by simultaneously open ing the evacuated vessel and a gas manometer to one atmos- : phere of normal hydrogen. The initial manometer reading was obtained by following the rate of absorption and f extrapolating back to zero time. No stirring was done 21 W. K. Wilmarth and C. F. Baes, Jr., J. Chem. Phys., 20, 116 (1952). 30 during this period of 4 to 5 minutes while the rate was followed. Then the solution was stirred at 400 rpm until equilibrium between the solution and the gas was attained. It was found that 3*50 cc. of hydrogen dissolved in 50.0 ml. ©f TUP at 25° 0. with the final pressure 76.0 cm. of mercury. For a solution of 0.37 g. of TEB in 50.0 ml. of THF (0.031 molar), the solubility of hydrogen was 2.80 cc. at 25° 0. Nitric Oxide Absorption Exploratory studies of the reactions of the com pounds in THF solution with NO were made. The rate of NO absorption, even without stirring, was too rapid to fol low^ att25° G. and one atmosphere NO pressure, but one mole of NO was absorbed per mole of NaTFB. However, TEB Itself absorbs NO in a more complicated manner. A slow initial rate was observed and this was followed by a very rapid absorption. The total NO absorbed was almost four moles per mole of TEB. The solubility of NO in THF was j 13.7 cc. in 50 ml. of THF at 25° C. and one atmosphere pressure of NO. The rate data and stoichiometry indicate that further studies at lower temperatures and varying pressures may be worthwhile. CHAPTER III OBSERVATIONS Magnetic Susceptibility Measurements The data from the magnetic susceptibility studies are tabulated in Table I. The low solubility of (NaTFB)g and (NaTNB)g in ether made it difficult to obtain magnetic data on these solutions since the large corrections needed for the diamagnetism of the solvent and the solute could disguise a small paramagnetic effect. The spectral studies indicate that the monosodium salt of TEB is dlmerlzed in ether as well as in THF. NaTNB in THF is definitely paramagnetic as shown in Table I. The concen tration effect in the magnetic data and the molecular weight measurements indicate that the substance is probably completely dissociated to radicals. The value for the paramagnetic susceptibility is low, but this may be due to the diamagnetic correction as discussed by Selwood for the trixenylraethyl radicals. The studies of TNB and (NaTNB)g in the solid state, as well as NagTNB in solution, serve as a cheek on the diamagnetic corrections calculated from Pascal's constants. 22 P. W. Selwood and R. M. Dobres, J. Am. Chem. See., J2, 3860 (1950). TABLE I MAGNETIC SUSCEPTIBILITY DATA Solvent Substance i Concentration X a x © X M * 106 A x 106 (XM - A) x 3 L06 Ether Na2TNB .027 M -.63 -247 -286 39 (NaTPB)g .155 M GO • 1 -117 -165 48 NaTNB ’ .026 M .76 298 -440* 738 NaTNB .072 M .73 286 -584* 870 * * NaTNB .115 M 1.05 420 -480* 900 Na2TNB .022 M -.57 230 -286 56 Solid TNB -.610 -239 -268 29 ► (NaTNB)2 j~. . " . . . - -.660 -544 -554 10 * ' Excess TNB present in these measurements. u> ro They show fair agreement as evidenced by the magnetic sus- ceptibility of TNB which was found to be *239 x 10 esu per -6 mole compared to -268 x 10“ esu per mole according to Pascal’s constants. Similar agreement is found for the other two substances as shown in the table. However, the diamagnetism of the radical NaTNB in THF is probably not the same as that calculated from Pascal’s constants.22* The temperature studies on the solids TNB and NaTNB showed the magnetic susceptibilities to be constant to i 1 per cent from 250° K. to 363° K. The solutions were studied only at 20° * 2° C. Absorption Spectra The absorption spectra of the solutions studied are shown in Figures 4 through 11. Figures 4 and 5 show the absorption curves for TPB and TNB in ether solutions and the decomposition products formed upon their exposure to air. The absorption maxima around 220 and 280 mp seem to be prevalent in all sub- 24 stituted naphthalene compounds, while the higher wave length absorption at 350 up is associated with TNB and l. Pauling, J. Chem. Phys., 4, 673 (1936). 24 R. A. Friedel and M. Orchin, Ultraviolet Spectra of Aromatic Compounds (New York: J. Wiley and Sons, I95l). 220 SPECTRA OP TRIPHENYLBORON IN E' SOLUTION AFTER THREE WEEKS ' 35 4.0 2.0 1.0 SAME SOLUTION , AIR, B________ SPECTRA OF IN ETHER, AFTER THREE WEEKS 280 320 380 400 440 48a 520 FIGURE 6 ABSORPTION SPECTRA OF (NaTEB)p IN ETHER (CURVE A) AND (NaTEB)c IN THF (CURVE B) d A - from 40# amalgam B - from 1# amalgam 0.8 0.2 pO 300 300 400 420 440 rajT FIGURE 7 ABSORPTION SPECTRA OF (NaTPB)2 IN THF PREPARED FROM AMALGAMS 280 320 360 400 440 480 FIGURE 8 ABSORPTION CURVES FOR (NaTNB )2 IN ETHER AT SEVERAL CONCENTRATIONS i _______ 520 rnp. log IVI A (conc. 6.3 x 0.2 720 r a j * ABSORPTION SPECTRA OP Na0TNB IN LO f a O r t 0.8 FIGURE 10 ABSORPTION SPECTRA OF NaTNB IN TETRAHYDROFURAN, CURVE A, AND SODIUM-NAPHTHALENE IN TETRAHYDROFURAN, CURVE B 4 = - d 700 800 900 1000 mp FIGURE 11 ^ ABSORPTION SPECTRUM OF (NaTPB)o WITH EXCESS SODIUM BORON CONCENTRATION ABOUT 1 X 10“4 M. log I Vi 0.8 0.6 0.2 t Concentration x FIGURE 12 OF OPTICAL ABSORPTION WITH CHAN IN ETHER, MEASURED AT 4 THE DATA FIGURE 7 disappears after long exposure of the ether solutions to t air. When THF is used for the solvent, a similar absorp- i i tlon curve is obtained for these substances with no i observable shift in maxima. | The absorption curves of (NaTEB)2 in ether and THF { are shown in Figure 6; the absorption is similar in the j i two solvents except for a shift to higher wavelengths in THF. Ghu in his study of the monosodium salt of tri- phenylboron in THF obtained not only the absorption maximum at 420 mp, but also a peak at about 340 mp as OK shown in Figure 7. This does not correspond to an ab sorption band of TPB or its decomposition products. It was found that this peak, which does not appear on re action of sodium metal with TPB, could be obtained by using 3odium amalgam. One curve obtained with a 1 per 1 cent amalgam (Ghu used 40 per cent amalgam) is shown in I Figure 7. The peak disappeared upon exposure of the solution to air. The concentrations were not known with sufficient accuracy to make quantitative comparisons. Figure 8 shows the nest of curves obtained in a study of the change with dilution of the optical absorp tion of (NaTNB)2 in ether. The concentrations are caleu- j lated from the known initial concentrations and the i — . ..— | 25 T. L. Chu, J. Am. Chem. Soc., 75, 1730 j i dilution factor of the apparatus. The adherence to Beer’s 1 law Is demonstrated by the linear relationship found when • the concentration is plotted against log Ic/l, Figure 12. The solutions of NaTNB in THF* as well as (NaTFB)g in ether and THF, were found to obey Beer’s -law in the con- _2i _c centration range 10 to 10 ^ molar. <? „ In Figure >1 it is seen that the shape of the ab sorption curves of the disodium salt of TNB in ether or THF appear similar in the visible region, but they exhibit a fairly large solvent shift with the maxima appearing about 40 mp to the red in THF. This large positive bathochromlc effect is not unknown for polar solvents as discussed recently by Hunig, et al., who found shifts of 26 over 100 mp in their studies of substituted quinones. The TNB peak at 350 mp does not show any displacement in 1 THF, Figure 5. Beyond the upper wavelength region shown i in Figure 11 the absorption curve of Na2TNB in THF begins to rise and was still rising at 1000 mp. Two features which distinguish the spectrum of NaTNE | in THF, Figure 10, from that of (NaTNB)2 in ether, | Figure 8, are the fairly sharp double maxima at 445 mp and 472 mp and the slight absorption at 640 rap. This substance j -------------- I I og ! S. Hunig, H. Schweeberg and H. Schwarz, Ann., j 2 (1954); G. N. Lewis and M. Calvin, Chem. Rev., \ (1939) | also exhibits absorption in the near infrared with the ! j curve still rising at 1000 mjx. This infrared absorption ofj i NaTNB and of NagTNB in THF is not found in ether solutions i i of any of the compounds or in THF solutions of (NaTPB)g or j sodium-naphthalene. Even the THF solution of TNB does not \ exhibit absorption in the infrared and, as mentioned above, | the solution does not show the bathochromic effect in THF that was found for the sodium salts of the compound. Recent magnetic susceptibility measurementsby Chu * and Yu indicate that the naphthalene anion is paramagnetic in THF solution.2^ The absorption spectrum of sodium- naphthalene, which was obtained by the reaction of sodium metal and naphthalene in THF, was identical with that ob tained by the reaction of one of the volatile decomposi tion products of NaTNB with sodium in THF. This was ac- 1 complished by heating the THF solution of NaTNB for some f time, and then distilling the THF over onto clean sodium metal in another flask of the vacuum system. The result ing green solution was poured into the dilution apparatus and sealed off. It is concluded that naphthalene is one of the decomposition products of NaTNB. The examination ■ of the spectrum of NaTNB in THF shows that the presence of J I | : ^ T. L. Chu and S. C. Yu, J. Am. Chem. Soc.* 76, ! i 3367 (1954). : ---------- ; --- 46“ small amounts of sodium-naphthalene cannot account for the I large paramagnetism of these solutions. When (NaTPB)2 In THF reacted with excess sodium, a blue solution formed. The absorption spectrum of this solution is shown in Figure 11. The analyses of these solutions showed that if Na2TPB was formed, it was not formed quantitatively. Therefore, at present, the spectrum is not identified with a particular substance. Molecular Weight Measurements - Since boron is known to exhibit such unusual bond ing ability in its bridged compounds, it was decided to examine the molecular weights of some of these arylboron compounds. The thermoelectric method offered the ad vantages that the measurements could be made relatively rapidly and that measurements could be made in different solvents at the same temperature. Using the calibration curve, Figure 13* the unknown molecular weight was determined by measuring AR and look ing up N on the curve. The molecular weight was calculated by using the equations M. W. * w(l - N)Nms (9) Here w represents the weight of solute, m_ represents the s moles of solvent, and N is the mole fraction of solute ob- __tained_frora_the. measurements. In._o.necase the resistance THF solutions Ether solutions Mole Fraction, N x 10' FIGURE 13 MOLECULAR WEIGHT APPARATUS WITH BENZIL CALIBRATION change was measured first for the triarylboron compound, i then for the monosodium salt, and finally for the disodium ! salt; thus a check on the calibration curve was obtained. j The results for the solutions are shown in Table II.j i The mono sodium salt of TNB was found to be monomeric in | i THF, thus confirming the paramagnetic susceptibility. The j diamagnetic monosodium salts of TNB in ether and TEB in THF were found to be dimers. The low solubility of (NaTPB)2 in ether made it impossible to determine the molecular weight. The high value found for TNB in THF is not con sidered significant and was due to the experimental dif ficulty Involved in getting a representative sample in the cup while trying to avoid reaction of the TNB with the sodium metal, which was to be used later to form the sodium salts. i Ortho-parahydrogen Conversion The results of these runs are listed in Table III. Wfcien the blue solutions of NaTFB with excess Na in THF j are allowed to stand for some time, they become red ap- ! parently due to reaction with the solvent as discussed in j the next chapter. The red solutions are unstable in the i > air, turning colorless, and Hun 1' shows the catalytic ef- I feet on the conversion rate by the red solution resulting j i I i from the solution used in Run 1. Run 4 is for the solution' MOLECULAR TABLE II WEIGHTS ©P THE TRIARYLBORON COMPOUNDS _ Solute b, R Coneentration (N) Molecular Weight Pound Molecular Weight Calculated Ether Solutions: k = 23,500 Benzll TO .0034 Standard 170 .0072 Standard 270 .0114 Standard 1 310 .0132 Standard Azobenzene 150 .0065 160 4 10 182 130 .0056 180 4 15 182 1 Tri- CX-naphthyl- 1 boron 40 .0017 350 ±90 392 (NaTNB)2 60 .0024 860 ± 140 830 Tetrahydrofuran Solutions: k = 25,700 Benzil 310 .0127 Standard 185 .0072 Standard 80 .0033 Standard (NaTPB)2 110 .0040 540 4 50 53© TNB 140 .0058 470 ± 40 392 NaTNB 150 10060 460 4 30 415 NagTNB 150 .0060 450 ± 30 438 TABLE III PARAHYDROGEN CONVERSION BY THE SODIUM SALTS OF TRIPHENYLBORON Run Solute k/rain Total Hg k* (25°) No. (Cone.) Hg in soln. 1 Na2TPB* (0.31M.) .00145 10.14 .0147 1‘ NapTPB* (red soln. from l) .00071 10.14 .0072 2 NaTPB (.042M.) .00023 15.2 .0035 3 NaTPB (.29 M.) .00024 13.6 .0033 4 LiAlH^ (.5 M.) .00033 9.22 .0031 *These are solutions of (NaTPB )2 with excess sodium. ............. 1 -------------- 51~ of in THF, which should show neither base catalysis nor paramagnetic catalysis. This served as a check ©n the 28 rate due to solvent and salt effects. The data indicate that the rate of conversion due to the catalytic effect of the monosodium salt is negli gible; especially noticeable is the result that no in crease in rate was observed with almost a ten-fold in crease in concentration. The blue solution, on the other hand, shows a large rate increase. This may be the result of base chtalysis. The solutions showed no evidence of exchange with deuterium when this gas was stirred with then. »! I after the hydrogen had been removed. Run 4 was made by Mr. J. Flournoy. CHAPTER IV DISCUSSION OF THE RESULTS The magnetic susceptibility measurements in con junction with the spectra and the molecular weight data Indicate that the monosodium salt of TPB in ether or THF and the monosodium salt of TNB in ether are dimerlzed. They should therefore be represented as (NaTPB)2 and (NaTNB)2. This conclusion is supported for (NaTPB )2 in THF by the results of the ortho-parahydrogen conversion measurements. In the solid state, the two substances are i diamagnetic as shown in Table I for (NaTNB )2 and as in dicated for (NaTPB)2 by the paramagnetic resonance absorp tion study of Chu.2^ Gn the other hand, NaTNB in THF is found to be paramagnetic and monomeric. Two of the re sults which seem worthy of consideration are the small dissociations of the monosodium salts to radicals in ether and the important effect of solvent when THF is used. Structure of the Dimer Before one can make comparisons between the iso- electronic boron and carbon compounds, one should know something about the bonding involved in the boron compounds and the bond energy available for this bonding. Recently Skinner and his coworkers determined the ' average energy of the boron-earbon bond to be 73*9 kcal. t per mole In Although part of this energy may j result from some double bond character, the B-C single j. bond must be quite strong. A boron-boron bond energy can j be calculated from the heat of sublimation of boron; however this heat of sublimation is In doubt-* and the bonding In boron Is complex. Also, while the data accumulated show that the average B-M bond is stronger than the correspond ing C-M bond, where M may be fluorine, chlorine, bromine, oxygen, or nitrogen, 29-31 results from TT bonding to boron, and It cannot be used to estimate the B-B bond energy. One can only say that such B-B bonding is energetically favorable. Various possible structures can be drawn for the dimer, but the ethane-like structure with boron-boron bonding would seem to be the most plausible. This form allows all of the aromatic groups to retain full Kekule resonance, and an examination of the Fischer-Hirschfelder models for TNB and TEB shows that such bonding is not pro hibited by the steric requirements of the aromatic groups. j Chem. Soe., 1952 j J 3378. T. Cham ley, H. A. Skinner and N. B. Smith, J. , 2288. ~ H. A. Skinner and T. F. S. Tees, ibid., 1953. H. A. Skinner and N. B. Smith, ibid., 1953. 54 n Since Brown and Sujishi found that even in the solid state the amine complexes of the unsymmetrieal form of TNB slowly converted to the symmetrical form, it is probable that such conversions are rapid in solution and that the symmetrical structure is present. Possible interactions between para-carbon atoms would Involve the loss of some Kekule resonance energy, but this might be balanced by the charge separation. (a) • The evidence that this form of carbon-carbon bonding does not appear in the isoelectronic Ar^CCAr^ -is an argument against such interaction. A study of the state of aggre gation of NaBCp-tolyl)^ would be of value, since here the para-substltuents might effectively block the bonding. The formation of a molecule with boron bonded to four phenyl groups cannot be disregarded by a steric argument, since boron is known to form such compounds as LiB(CgH^)^.^2 The carbon-boron bond is not the same in that compound, however, as it would be for the molecule depicted in formula (b) where the para-hydrogen atom ^2 G. Wittig, G. Keicher, A. Huckert and P. Raff, Ann., 563, 110 (1949). 55' would Interfere with the bonding. The L1B( «-naphthyl is unreported. A consideration of possible structures, which use aromatic groups to form bridges between the boron atoms, shows that the case is not analogous to the boron hydrides, in which hydrogen bridges serve to bond the borine groups. J First it should be observed that the B&r^ molecules are monomeric; therefore, bridging through the orbitals of the aromatic rings does not occur in those molecules. When one adds an electron to each of two BAr^ molecules, several possible bridge structures can be drawn. Since the ethane-like structure is not electron-deficient, these forms seem unlikely; also they require loss of Kekule resonance energy. At present, it is felt that the most plausible structure is the one involving boron-boron interactions. As discussed below, most of the observations are explicable * J ^ H. Ci. Longuet-Higgins and R. P. Bell, £. Chem. Soc., 1943, 250; K. Hedberg and V. Schomaker, J. Am'. Chem. Soc., 7§7"i482 (1951); K. Hedberg, M. E. Jones~anH“Y. ScEomaker, ibid., 73, 3538 (19517; and J. S. Kasper, C. M. Lucht and D. larker, ibid., 70, 80I (1948). : 56-, with this structure for the dimer. Steric Effects The weakening of the bonding in the dimer due to steric repulsions will be considered first by comparison of the dissociation constants of the isoelectronic groups across the periodic table. Prom the adherence of the absorption spectra to Beer's law, one can calculate that the equilibrium constant for the dissociation of (NaTPB)2 to monomer is less than 8 o 10 , while for hexaphenylethane*3 the dissociation constant is about 5 x 10 and the one nitrogen compound studied was completely dissociated. Although C( cx-naphthyl)^ has not been prepared, the fact that diphenyl- cx-naphthyl- 84 —2 methylJ has a dissociation constant of 4 x 10 would in dicate that the trisubstituted compound would be greatly dissociated. Yet NaTNB is undissociated in ether. It seems probable that steric effects play a big role in these dissociations. Certainly in going from CAr^ to NAr^+ one would predict that the central atom will become smaller and the steric effect more important in weakening the bond energy. So, as observed, the boron compounds are least dissociated to free radicals. The ^ C. S. Marvel, J. W. Shackelton, C. M. Himel and J. Whitson, ibid., 64, 1824 (1942). 57" trend may, however, merely reflect the variation in bond energies; this is particularly true for the nitrogen case since the nitrogen-nitrogen bond is known to be weak. s ' In THF where dissociation of the boron dimer is possible, the steric effect becomes important and the trinaphthyl compound is dissociated while (NaTPB )2 remains dimerized. Resonance Effects The question of the importance of resonance in stabilizing the triarylmethyl radicals has been discussed i in some detail by Wheland. The study of the effect of substituents On the associations of these radicals seems to indicate that resonance is of somewhat less importance than steric strain. In Considering the resonance structures of the mono sodium salts of the triarylboron compounds and the iso- electronic compounds of carbon and nitrogen, one can see that the principal contribution will arise from forms in which the impaired electron has been shifted from the central atom to one of the ortho- or para-carbon atoms of the aryl group. An example of one of these resonance forms is: * > Q «— ► J > 0 * (10) 58 | I Here E may represent boron, carbon or nitrogen with suit able charge on the radical. In these structures, no mat ter which of the elements we are considering, all of the atoms in the entire unit retain the same formal charge in the various resonance structures. It is further evident that the,quinoidal form in all cases is much higher in energy than the first form where the Kekule resonance in the aromatic rings has not been destroyed. In spite of their unfavorable structures, the quinoidal forms must contribute a considerable resonance energy to the system. However, one can see no special reason for a large de crease in stability for the boron case although there may be a continuous trend in this direction in going from nitrogen to boron. The rings must approach coplanarity for appreciable resonance or overlapping of the 7T orbitals of the aromatic carbon atoms with that of the boron atom to occur. This may be slightly more favorable for BAr^~ than for CAr^, because of the larger bond radius of the neutral boron atom (0.80 2 units) compared to that of carbon (0.77 & units). The addition of the extra electron may cause some further expansion of the boron radius in BAr^~. Examina tion of models of TUB shows that complete planarity of atoms is impossible; however, the structure is surprising- -ly -flat.-— fteissraan,_et_al.-,_have„recently shown.by_____ 59 1 paramagnetic resonance absorption that sodium trimesityl- j boron in THF is paramagnetic. ^ A consideration of models i i of this compound shows that the steric effects of the ortho? 1 i methyl groups will be very effective in preventing a planar; t I configuration of the aromatic rings. Therefore, it seems ! I unlikely that resonance could play much of a role in the ; | stabilization of this radical; however, the boron atom is so deeply buried in this bulky molecule that probably steric effects are sufficient to prevent dimerlzation. Information on the extent of dissociation is not yet avail able, and the isoelectronic C(mesltyl)^ has not been reported. It should be noted that the concept of a resonating free electron may have to be modified soon. Recently Welssman and Sowden described a study of the paramagnetic [ ' resonance absorption of triphenylmethyl radical which contained in the methyl position.They calculate from the results that the average distance between the methyl carbon atom and the free electron is 0.7 $ units, which may indicate that little resonance occurs. However, | they caution against making interpretations until further g, Weissman, J. Townsend, D. E. Paul, G. E. | Pake, J. Chera. Fhys., 21, 2227 (1953)* j o g S. I. Weissman and J. C. Sowden, J. Am. Chem. i soc., ZS, 503 (1953). „ 60~: i studies have been coinpleted. j i Solvent Effect j I ■ The fact that the dimer dissociated appreciably in | THP and not in ether could result from a stabilization of ! the monomer in THP compared to ether or from a stabiliza- j i tion of the dimer in ether, me evidence shows that solu- | tions of NaTNB form when THP is brought into eontact with solid (NaTNB)2 and that no NaTNB is discernible in saturated ether solutions. Therefore, relative to the solid, the monomer is more stable in THP than in ether. By measuring the amount of dimer present in THP when the solution is saturated, one could determine whether the dimer is also stabilized in THP. At present this cannot be calculated since none of the physical measurements were made on THP solutions of the NaTNB in equilibrium with I i solid. The same trend may be present in the case of NaTPB in THP or ether; here, however, not enough monomer to be detected is formed in either solvent. The increased solu bility of the dimer (NATPB)2 shows that it is stabilized in THP compared to ether, i The stabilization of NaTNB in THP might result from I | an acid-base reaction between the triarylboron compound ! I * ' | and THP. This would prevent reduction occurring on the j I boron atom, and instead the sodium would react on one of ! the naphthyl groups. The acid-base studies of Brown and Adams indicate that, for the acid BFg, THF is a stronger base than ether by about 2.6 kcal. per mole.The a F j for the reaction of ether and BF^ is -2.85 kcal. per mole ' and for the reaction of THF and BF^ a f is -5.45 kcal. I per mole; these values of the free energy changes at 20° C.j were calculated from the data of Brown and Adams. Data j are not available for the reaction of TFB and TNB with the two solvents, but some reaction will probably occur. If the THF fills the boron orbital by acid-base reaction, | then on reduction there will be n + 1 electrons for n i orbitals and the odd electron will be delocalized; this would make bonding less favorable. Increased solvation of the cation in THF might also account for the monomer stabilization. One would predict i that solvation of the aromatic groups would not vary much i as the solvent changes from ether to THF. Changing the alkali metal might give some clue to the importance of this effect. Lithium ions might show a larger solvent effect, while the potassium salts would indicate the | steric effect of the alkali metal. f ! ^ H. C. Brown and R. M. Adams, J. Am. Chem. Soc,, i : -64, 2557-63 (1942). “ “ “ - : , 62“ ] Two-step Reduction of the Triarylboron Compounds Although from the view of classical electrostatics it should be difficult to reduce the arylboron anions fur ther, one finds that the second step in the reduction occurs almost as easily as the first. The reduction of the triphenylmethyl radical to form Ar^CT requires 1.9 volts in the gas state according to Bent.38 The reduction of Ar^B to Ar^B” requires 2.1 volts according to the data of 7a —2 Bent and Dorfman and the further reduction to Ar^B occurs almost as readily. An explanation of this can be given in terms of resonance. The resonance structures for the ions containing one more electron, while formally similar, differ in a fundamental way from those in Equation (10). This may be demonstrated by considering 2- two structures for an Ar^B ion: (11) Here in going from nitrogen to boron the negative charge on the central atom is progressively increased, and at boron the first form may be as unstable as the quinoidal form. This would permit an increase in resonance. It 38 H. E. Bent, J. Am. Chem. Soc., 52, 1499 (1930); E. Swift, Jr., ibid., F0,“T403 (193#). j may be noted that this two-step reduction occurs most i ! readily with TNB for which more resonance is possible. i I Expressed another way, the "smearing" of the negative j charge over the entire ion stabilizes the dinegative ion 1 1 * i compared to the form where both charges would be localized j i on the same atom. This effect may be aided by the lower ! electronegativity of boron. The formation of NagTPB, which may be indicated by the appearance of the blue solutions when (NaTPB)2 reacts with excess sodium, has not been proved by the analyses which are consistently low in sodium. When attempts were made to isolate Na2TPB by removal of solvent, sodium metal deposited on the walls of the vessel; upon complete removal of solvent, the compound was irreversibly changed and the white polymeric residue did not react with sodium. The 1 spectrum indicates that there is not an equilibrium in- i volved between (NaTPB)2 and the disodium salt. The results may be due to competing reactions which prevent quanti- * tative formation of NagTPB; this is indicated by the com- j plete change of color within about two weeks to a red ' solution. The effect is Independent of whether the solu- ' j j tion is in contact with sodium or not. It also occurs in j the dark. This is possibly a reaction with ’ IBP similar I ! to that found by Wittig and Buckert with mixtures of I ; 1 NaCfC^H,-)^ and TPB in THF. from which they isolated the | 64 5>5,5* triphenylpentanol. The red solutions which are | obtained in the sealed reaction vessels are decolorized on i ! exposure to air. Further studies of the dependence of the | | reactions on concentrations, shaking speed and solvent, as I i well as identification of the products are needed. ^ t I The disodium salt of TNB appears black and is j destroyed rapidly upon exposure to air. This may be com- i pared to the products obtained upon reaction of TNB with ammonia or amines. These are very stable and apparently by filling the boron orbital the bases shield it from further attack; they are colorless. The reactivity of i the dlsodlura salt no doubt results from the large reduction potential. It was found that the solutions of the sodium salts would even attack Teflon and within a month destroy about a mm. layer. The absorption in the visible indi- 1 cates considerable resonance for the dlsodlum salt compared 1 26 to the amine complex. One use made of the spectra was the study of the possible reactions i B(C6h5)3 + NaB(C6H5)3 ^ (CgHj^BNaBtCgH^ (12) t ! proposed by Bent to explain anomalous behavior in his j equilibrium studies.*^a Neither titrations of NaBCCgH^)^ j 1 solutions with solutions nor dilutions of solu- 1 tions containing both NaTPB and TPB gave any spectral i ! j indication of such an interaction. The titrations were I made by using the breakoff seal on the apparatus in Figure j 1. A fixed amount of the solution of NaTPB in ether was I retained in the cell, and the TPB solution was poured in j slowly to the mark on B. The TPB solution was prepared | 1 in the vacuum system and the precautions usually observed for preparing the sodium salts were taken. The maximum in the absorption curve was lowered only by the amount anti cipated for the dilution effect. Repeated titrations con firmed this result. Also solutions which contained both TPB and NaTPB gave no deviations from Beer’s law to i 1 per cent when diluted by a factor of two, rediluted by two and finally diluted a third time. If reaction (8) occurs, its equilibrium constant must be much lower than indicated by Bent’s data, which indicate that the equilibrium con- 4 i stant must be about 10 . Other studies in this program Indicate that the reaction of the sodium salts with water, which during the months required to reach equilibrium slowly diffused from the glass, ean account for his ob servations. I i | | i Comparison of the Spectra of the Isoelectronlc Compounds j 1 40 l The absorption spectra of sodium triphenylmethyl, j i ___________________ I J | /V I L. C. Anderson, J. Am. Chem. Soc., 57, 1673 l ; (1935). 66n ill iip tripheny lmethyl radical, and triphenylmethylchloride i have been published, and an attempt was made to correlate these with the spectra of the boron compounds: TPB, (NaTPB)2, and the solution of (NaTPB)2 with excess sodium. | No similarities were observable either in the positions of i the maxima or in the general shapes of the curves. Sodium j triphenylmethyl in ether at room temperature, for example, shows four peaks at about the same height between 250 up and 500 mp, while TPB shows the narrow band and shoulder, Figure 5. Attempts to explain these differences must await a more complete theory for the spectra of such com plex molecules suid more information about the different configurations involved. Summary To gain a better understanding of the, factors in- i ' fluencing the formation of free radicals, a study has been made of some of the boron compounds which are isoelectronic to the triarylmethyl compounds. Magnetic susceptibility measurements, spectra and molecular weight studies of solutions of triphenylboron, 1 j - - - - - - - - - - - - - - - - - - - - - ' j | 1 G. N. Lewis and D. Lipkin, ibid., 64, 2801 (1942) i G. N. Lewis, D. Lipkin and T. T. Magel, ibid., 66, 1579 ! (1944). j I 42 T. L. Chu and S. I. Weissman, J. Chem. Phys., ! I — 21,_22_.(.1953-)____________________________ ~ ~ ____________I 67 tri- a-naphthylboron and their sodium salts have been in terpreted in terms of their state of dissociation to,free radicals. Although (NaTEB)2 in THF or ether and (NaTHB)g in ether are not dissociated to radicals, the free radical NaTNB exists in THF. This effect of solvent is inter preted in terms of increased steric repulsion for NaTNB when solvated, or possibly greater solvation energy for the monomer than for the dimer. Considerable solvation in THF is indicated in the absorption spectra. Although the molecular weight studies show the com pounds to be dimers, there is as yet no structural evi dence to-indieate the configuration of the dimer. The small dissociation of the boron compounds in ether com pared to the isoelectronic carbon compounds may be inter preted as a steric effect due to differences in the relative sizes of the central atoms, or it may be related-to the type of bonding in the boron compounds. Attempts to obtain evidence of a possible inter action between TPB and (NaTEB)2 as proposed by Bent and Dorfman gave negative results; it is possible that their anomalous results are due to reaction of the sodium salt with water which slowly diffuses from the glass. Extreme care in the treatment of the glass was found necessary before preparing the sodium salts. -------The - ab ility-of—the_arylboron_c orapounds.. to_undergo__ — — 68 “ two step reduction is described and an explanation for the relative ease of this reduction is given in terns of resonance structures. i BIBLIOGRAPHY A. BOOKS Friedel, R. A., and M. Orchin, Ultraviolet Spectra of Aromatic Compounds. Hew York: J . Wiley and Sons, Gilman, H., and R. B. Jones, Organic Reactions, Vol. V. New York: J. Wiley and Sons, 195TI “ Selwood, P. W., Magnetochemistry. New York: Interscience Publishers, 1943. Waters, W. A., The Chemistry of Free Radicals. London: Oxford PresiT^f^. ~-------- Wheland, G. W. Advanced Organic Chemistry. New York: J. Wiley and Sons, 1949? B. JOURNALS Anderson, L. C., J. Am. Chem. Soc., 57, 1673 (1935)• Beckman, E., Z. physik. Chem., 63, 197 (1908). Bent, H. E., J. Am. Chem. Soc., ^2, 1499 (1930) • Bent, H. E,, and M. Dorfman, J* (1932). Am. Chem. Soc., 2133 Bent, H. E., and G. J. Lesnich, 1246 (1935). J. Am. Chem. Soc., 51* Bent, H. E., and M. Dorfman, J. (1935). Am. Chem. t s Soc., 57, 1259 Bent, H. E., and M. Dorfman, J. (1935). Am. Chem. Soe., .51* 1924 Bent, H. E., and H. M. Irwin, Jr., J. Am. 2072 (1936). ~ Chem. Soc 58. Brown, H. C., and R. M. Adams, J. Am* Chem, Soc., 64, 2557 | (19#S). Brown, H. C., and S. Sujishi, J. Am. Chem. Soc., 70, 2800 (1948). Chamley, T., H. A. Skinner, and N. B. Smith, J. Chem. Soc., 1952, 2288. ” Chu, T. L., and S. I. Weissman, J. Chem. Phys., 21, 22 (1953). Chu, T. L., J. Am. Chem. Soc., 75, 1730 (1953). Chu, T. L. and S. C. Yu, J. Am. Chem. Soc., J6, 33^7 (1954). Hedberg, K. and V. Schomaker, J. Am. Chem. Soc., 73, 1482 (1951). Hedberg, K., fo E. Jones, and V. Schomaker, J. Am. Chem. Soc., 73, 3538 (1951). Huni^, S^ H. Sehweeberg, and H. Schwarz, Ann., 587, 132 Kasper, J. S., C. M. Lucht, and D. Harker, J. Am. Chem. Soc., JO, 881 (1948). Krause, E., and R. Nitsehe, Ber., 55B, 1261 (1922). Krause, E., Ber., 57, 216 (1924). Krause, E., Ber., 57, 813 (1924). Krause, E., and H. Polaek, Ber., 59, 777 (1926). Krause, E., and H. Polack, Ber., 6l, 271 (1928). Krause, E., and P. Nobbe, Ber., 63, 934 (1930). Krause, E., and P. Dittmar, Ber., 63, 2347 (1930). Kulkarni, S. B., Mature, 171, 219 (1953). Lewis, G. N., and M. Calvin, Chem. Rev., 2£, 273 (1939). Lewis, G. N., D. LLpkin, J. Am. Chem. Soc., 64, 2801 (1942). ----------------------------: ---------- 72H I Lewis, G. N., D. Lipkin, and T. T. Magel, J. Am, Chem, Soc,I, 66, 1579 (1944). --------- ------ Longuet-Higgins, H. C., and R. P. Bell, J. Chem. Soc., 1943j 250. Marvel, C. S., J. P. Kaplan, and C. M. RLmel, J. Am. Chem. Soc., 6^, 1892 (1941). Marvel, C. S., J. W. Shackelton, C. M. Himel. and J. Whitson, J. Am. Chem. Soc., 64, 1824 (1942). Michaelis, A., Ber., 2£, 244 (1894). Michaelis, A., Ann., 315, 19 &1901). Muller, R. H., and H. J. Stolten, Anal. Chem., 25, 1103 (1953). Pauling, L., J. Chem. Phys., 4, 673 (1936). Rumpf, P., and P. Trornbe, Compte. rend., 206, 671 (1938). Selwood, P. W., and R. M. Dobres, J. Am. Chem. Soc., J2, 3860 (1950). ~ Signer, R., Ann., 478, 246 (1930). Skinner, H. A., and T. P. S. Tees, J. Chem. Soc., 1953» 3378. Skinner, H. A., and N. B. Smith, J. Chem. Soc., 1953* 4025. Swift, E., Jr., J. Am. Chem. Soc., 60, 1403 (1938). Weissman, S. I., and J. G. Sowden, J. Am. Chem. Soc., 75» 503 (1953). - Weissman, S. I., J. Townsend, D. E. Paul, and G. E. Pake, J. Chem. Phys., 21, 2227 (1953). Wilmarth, W. K., and C. F. Baes, Jr.> J. Chem. Phys., 20, 116 (1952). Wittig, G., G. Keicher, A. Rnckert and P. Raff, Ann., 563j 110 (1949). Wittig, G., and A. Ruckert, Ann., 566, 101 (1950). Wittig, G., and P. Raff, Ann., 573, 195 (1951). APPENDIX APPENDIX A PREPARATION OF Bfe-XENYL)^ Initial attempts to prepare tri-£-biphenylboron, TXB, by the reaction of biphenyl bromide with magnesium, in the presence of iodine, to form the Grignard reagent proved unsuccessful, although the other arylboron compounds were obtained by this method. The preparation was next attempted using a method found suitable by Wi&tig for the preparation of Bfphenyl)^;^2 here lithium phenyl reacted with BF^-etherate. It was found that some lithium xenyl formed when xenyl bromide refluxed with lithium metal in anhydrous ether under nitrogen, but for best yields of the lithium compound the lithium interconversion method was adopted.^ To 200 ml. of dry ether in a one liter 3-neck flask there were added 4.3 g. (0.6 gram-atoms) of lithium ribbon cut into small pieces under a stream of nitrogen gas. Witfc a stirrer turning at moderate speed, 30 drops of a solution of 34.2 g. (0.25 moles) of n-butyl bromide in 50 ml. of ether were added. The reaction was cooled to -10° In a Dry Iee-acetone bath which was kept at -30° to -40°. The 4-a J H. Gilman and H. B. Jones, Organic Reactions (New York: Wiley and Sons, 1951), Vol. VI, p. 339- ‘ 76 lithium became shining in the solution but black on the surfaee where it contacted Ng, and the solution soon clouded and became purple colored. The remainder of the butyl bromide was dropped in slowly over a period of 30 minutes, and when the addition was completed the reaction mixture was warmed in an ice bath at 0° to 10° for one and a half hours. To separate the solution from the lithium, it was then poured through a bent glass tube containing glass woolinto another one liter 3-neck flask. There was nitro gen in both flasks during this manipulation. A solution of 40 g. (0.17 moles) of xenyl bromide in 200 ml. of ether was added dropwise while stirring; this took about one *\ hour. The reaction was kept at 0° in an iee bath during the addition. The color of the reaction mixture changed from grey to light-brown to turbid white. About 20 ml. benzene were added to keep the solution homogeneous. The ice bath was removed and the reaction kept at 25° for 30 minutes, then 5.7 g* (.04 moles) of BE^-etherate in 25 ml. of ether were added during a 15 minute period. The solution became faintly yellow, and it was then refluxed for three and a half hour's. After cooling, the light yellow solution was hydrolyzed with chipped ice and the ether layer was dried over CaClg. A white solid separated during,the_hydrolysis; it had the melting point of p.p1- 77 dixenyl (296-298°, uncorr.) and contained no boron. The B(xenyl)^ was precipitated by concentrating the ether layer. This was then heated while evacuating; when * ^ } it was kept at 160-165° for a half hour a white solid with a melting range of 87-88° (uncorr.) was obtained in a cold trap. Xenyl brpmide has a melting point range from 89-91°, whereas that of biphenyl is 69-70°. The final product, recrystallized from benzene, was a faintly yellow powder of melting point range 210-212°, and although this was apparently still solvated attempts to purify by subliming at higher temperatures resulted in decomposition. The product obtained reacted with Na in ether in the apparatus previously described. The solution became ©range, then blue-black, but it reacted very slowly apparently because of the low solubility of the TXB in ether. rf2JeesNr«r»irv o f t t o n t t o n C s f i M a b

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An experimental investigation of amino compounds of boron hydrides

PDF

A study of the reaction of tetramethyldiborane with sodium in liquid ammonia and the products derived from this reaction

PDF

A study of the activation of hydrogen by cuprous salts in quinoline / by Max K. Barsh

PDF

An experimental investigation of the interchange of carbon monoxide and cyanide ion in zerovalent nickel compounds

PDF

An experimental investigation of amine addition compounds of trimeric and tetrameric phosphonitrilic chloride

PDF

A study of the kinetics of the reaction of sulfenyl halides with olefins

PDF

Application of the Hammet equation to the dissociation of the substituted 1,1,4,4-tetraphenyl-2,3-dibenzoyltetrazanes.

PDF

A study of the reaction of 2,4-dinitrobenzene-sulfenyl halides with ketones

PDF

Addition compounds of trimethylamine oxide

PDF

Addition compounds of trimethylamine with boron fluoride and its methyl derivatives.

PDF

Addition compounds of boron fluoride with certain sulfoxy-amines

PDF

A preliminary investigation of the formation of molecular addition compounds by phosphorus chloronitride

PDF

Alpha nitro sulfides and related studies

PDF

Cryoscopic and conductometric investigations of manganese (VII) and chromium (VI) compounds in sulfuric acid

PDF

Complex compounds of trimethylamine oxide

PDF

A study of the interactions of polar gases with solid proteins and some simple organic compounds

PDF

Chemical properties of the phenoxyacetic acids: Reactivity with Fenton's reagent

PDF

A study of the polargraphic behavior of decaammine-[mu]-peroxodicobalt (III,IV) perchlorate

PDF

A study of the reaction of 2,4-dinitrobenzenesulfenyl chloride with the cis- and trans-stilbenes

PDF

A determination of ionic diffusion coefficients

Asset Metadata

Creator Moeller, Carl W., Jr (author)

Core Title A study of the structures and properties of triaryl compounds of boron

Degree Doctor of Philosophy

Degree Program Chemistry

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemistry, organic,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Wilmarth, W.K. (committee chair), [illegible] (committee member), Brown, Ronald J. (committee member), Burg, Anton B. (committee member), Copeland, C.S. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-619948

Unique identifier UC11354987

Identifier DP21773.pdf (filename),usctheses-c17-619948 (legacy record id)

Legacy Identifier DP21773.pdf

Dmrecord 619948

Document Type Dissertation

Rights Moeller, Carl W., Jr.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA