"

"-

RESEARCH MEMORANDUM

PREPARATION AND PHYSICAL PROPERTIES

OF SOh/LE TRIALKYLBORANES

By Louis Rosenblum and Harrison Allen, Jr.

Lewis Flight Propulsion Laboratory Cleveland, Ohio

NACA RM E55E06 A

c NATIONAL ADVISORY C O W = FOR AERONAUTICS

By Louis Rosenblum and Harrison Allen, Jr.

SUMMARY

Triethylborane, tri-n-propylborane, and tri-n-butylborane were pre- pared from boron t r i f luoryde and the appropride Erignard reagent. The f i r s t two compounds were d i s t i l l e d to pur i t i e s of' 99.8 and 99.7 mole per- cent, respectively, but tri-n-butylborane decomposed during d i s t i l l a t i o n . The following physical prope&ies of triethylborane and tri-n-propylborane

ing point, (2) heat of fusion, (3) boiling point, (4) heat of combustion, (5) re f rac t ive index, (6) density, (7) d i e l ec t r i c constant, and (8) in- frared spectra. Experimental. va lues fo r t he r e f r ac t ive idex of tri- ethylborane, the heat of fusion of the tri-n-propylborane, and the di- electr ic constant of both compounds are reported here for the first time.

- 1 w e r e determined and comptxred with values from the literature: (1) freez-

*

INTRODUCTION

The lower-moleculax-weight trialkylboranes have heat-release prop- erties intermediate between those of the boron hydrides and the conven- t i o d hydrocarbon fuels. In order to evaluate the potential of these organoboron c ~ o u n d s as high-energy fuels and as ignition sources, and t o obtain fuel-system design Wormation, it i s necessary to have accu- rate values for several of their physical properties. A s w c h of the literature revealed values for only Borne of these properties (refs. 1 t o 6) , and among the values were several large discrepancies. In order t o obtain more accurate and consistent data, triethylborane, tri-n-propyl- borane, and tri-n-butylborane were synthesized and purified at-the NACA L e w i s laboratory-

There are several methods for the preparatian of triallrylboranes re- po r t ed i n t he l i t e r a tu re (refs. 1 t o 3 asd 7 t o 9 ) . Most of the methods are var ia t ions of the general reaction:

B X Q + ~ R M F R ~ B + ~ M X ether

2 ._ NACA RM E55EU6

where

X = halide, or OR

R = alkyl group

M = MgX, 1/2 @-, or- .1/2 C d - . ..

The reaction of boron t r i f luor ide etherate with the appopriarte G r i g n a r d reagen-b was reported to give good yields of . the desired products ( ref . 8) 5 and was selected for the present work. tr)

Triethylborane and tri-n-propylborane were obtained in pur i t i e s of 99.8 and 99.7 mole percent, reeq?ectively; tri-n-butylborane was not obtained i n a pure form since it decomposed &?in@; t h e d i s t i l l a t i o n process.

-

The physical properties determined f o r t h e two purified compounds are (1) freezing Point, ( 2 ) heat of fusion, (3) boiling point, ( 4 ) heat of combustion, (5) refractive index, (6) density, (7) d ie l ec t r i c conatant, and (8) infrared spectra.

SyNlCfIEsIS

General Procedure

The trialkylboranes w e r e prepared by the

EF3.Rt2O + 3 FM@r - R3B + Rh0 following general reaction:

3 M@rF

where

when R = ethyl, then R ' = butyl

but when

R = butyl or propyl, then R t = ethyl

A low-boiling product such a0 triethylborane could be isolated by d i r e c t d i s t i l l a t i o n from the reac t ion m i x t u r e (ref. 3). The higher- boiling trialkylboranes Could not be isolated in this manner because of decomposition at the higher distillation temperature i n t h e presence of copious amounts of s a l t s . The salts and other hydrolyzable components can be removed by the addition of-water upon ccmpletion of the reaction (ref. 8) . Separation of the ether and aqueous layers and dist i l la tLon of the ether yields -the triallrylborane; however, the amount of product - t h a t can be prepared i n this manner is- limited by the prac t tca l eize af the reaction vessel.

. . . 9

HACA RM E55Eo6 c

3

A modification of the existing procedure made possible the prepara- - t i o n of several batches of product without the usual dismantling, clean- ing, and reassenibling of equipment. When a l i t t l e less than an equivalent amount of boron t r i f luoride e thyl e therate had been added to the Grignard reagent, there was a rapid separation of t h e mixture into two layers. The top layer consisted mainly of t r i a w l b o r a n e and ether, and the bot- tom layer consist& of excess G r i g m r d reagent, salts and ether . The top layer could be siphoned from the react ion flask, more G r i g n d reagent added, and the operation repeated. Tri-n-propylborane and tri-n-butyl- borane w e r e prepared by t h f e modified procedure. -

Materials

Ethyl, n-propyl, and n-butyl bromide f r o m a commercial source were washed with Gater, dried over anhydrous sodium sulfate, and distilled. The refract ive indices of the materials were nzO 1.4242, 1.4341, and 1.4399, respectively.

a II

Boron trifluoride, 97-percent pure, was used as obtained from a comer cia1 sow ce .

I

Ethyl ether purchased commercially was dried over calcium hydride Etna was not purif ied further.

Butyl ether of commercial or igin was percolated through a column packed with activated alumina, dried over calcium hydride, and ueed with- out further treatment.

The preparations of tri-n-butylborane and triethylborane are de- scribed i n detail t o i l l u s t r a z e the general procedure for the synthesis of the high- and low-boiling trialkylboranes, respectively.

Preparation

Tri-n-butylborane. - The n-butyl magnesium bromide was prepared in a 5-gallon stainless steel reactor by the addition of 20 moles of n-butyl bromide Fn 3 liters of ethyl ether t o 20 moles of magnesium tum&gs i n 2 l i ters of e thyl e ther . The Gri@;na;rd solutfon was stored i n a 5-gallon tank under nitrogen until needed. A n ether solution of boron t r i f l uo r ide ethyl etherate was prepared by the slow addition with st irring of 6 moles of boron t r i f l u o r i d e t o 2 Uters of e thyl ether at loo C.

- -

4 NACA RM E55w6

The appaxatus f o r the preparation of tri-n-butylborane consisted of a graduated lZ-Uter three-neck flask equipped-with a stirrer, a large- - capacity stainless e t ee l condenser, and a graduated dropping funnel. The entire apparatus was enclosed i n a lmge dry box under a nitrogen atmosphere.

Three l f t e r s of Gri- solution were siphoned into the 12-liter flask from the storage tank. Then boron trif luoride etherate solution was added w i t h s t i r r i n g at such a rate (about 250 ml/br) as to promote vigorous refluxing. After approximately 650 milliliters of the boron t r l f l uo r ide etherate solution had been added, the reaction mixture sep- arated quite suddenly in to two layers, the bottom layer apparently con- sisted of Grignard solution and m a g n e s i u m salts. This layer, which was black with white salt-like solids sus-pended throughout it reacted vigor- ously w i t h water, evolving gas. The clem pale-yellow top layer sharea 110 reaction with water and was apparent-ly an ether solution of tri-n- butylborane. This top layer was siphoned in to a W g e separatory f k e l by means of a tube having a sintered-glass disk fused on one end t o a c t as a f i l ter . When the en t l re top layer had been removed, another J-U%er portion of Grignard reagent was added t o the reaction flask, and the . en t i r e procedure repeated. A total of 8300 millilitffs~~of Grignard SOU- t i o n and 5.9 mles of boron trif'luoride etherate was used. T h e r e wab a s-ht excess. of Grignard reagent remainin.aSter the Last of the boron t r i f l uo r ide had rezcted .

.

ro tr)

The top l a y e r s obtained from the above reaction were washed i n the sepsratory funnel with a saturated aqueous solution of ammonium chloride and then w i t h water. Finally the ether solution was transferred tu a distiUatrLon flask and the ether removed by vacuum d i s t i l l a t i o n . The l iquid residue which c0ntainei .a small amount of salts was filtered. The t o t a l weight of crude product was 989 grage; the yield was 9 1 percent based on- boron t r i f l uo r ide .

. . - .

Triethylborane. - Fi r s t e thy l magnesium bromide was prepared i n a 5-gallon reactor by the addition of a solution of 48 moles of e thyl bromide i n 3 liters of butyl ether t o a mixture of 48 moles of magnesium turnings i n 3 liters of butyl ether. The boron t r i f luorlde butyl e therate wa8 pre- pared by the low addition with s t i r r ing of E5 moles of boron t r i f l uo r ide t o 4 liters of butyl e ther a t loo C. This boron t r i f l uo r ide solution was added t o t h e Grignard so lu t ion in the reactor over a period of L$ hours. During the addition the outer jacket of the reactor was. cooled so as t o maintain a reaction temperature of 40' t o 45' C. when d l the boron tri- f luoride had been added, the reaction mixture was heated at- 55O C f o r - 2 hours. The triethylborane was then distilled out of the reactor direct ly into a steel cylinder. Through the entire synthesis there was a posit ive pressure of helium maintained i n t h e r e a c t o r t o exclude oxygen.

. .

I

NACA RM E55E06 5

Purif icat ion

The crude reaction products were each distilled through a 22- millimeter by 6-foot Podbielniak column. The entire dis t i l la t ion setup was enclosed i n a dry box f i l led with a hellum atmosphere ( f ig . 1) .

-

The distillate was received and stored i n 60-mil l i l i ter nazrow- mouth, 19/22 standard taper, outside-ground bottles with 19/22 st&a;rd- taper hollow caps. A strong steel spring secured the cap t o t h e b o t t l e . w

41 cn This type of bottle permitted the use of s i l icone grease t o aid i n sealing 0 v u l e preventing contamination of t he i n t e r io r of the bo t t le ( f ig . 2 ) .

Triethylborane was d i s t i l l e d at atumspheric pressure, while tri-n- propylborane and tri-n-butylborane w e r e distilled at pressures of 20 Z d 10 millimeters of merzury, respectively. Triethylborane and t r i -n- propylborane were successfully distilled, but tri-n-butylborane &corn- posed during d i s t i l l a t i o n . The decomposition p rd i i c t s adulterated the dist i l late besides causing the column t o flood continuously. Larering the pot temgerature about 20° C by reducing the d i s t i l l a t ion p ressure

d i s t i l l i n g t h e compound through an 18-inch glass column, the amunt of decomposition w a s reduced. However, the distillate s t i l l w a s no-t suffi- ciently pure to warrant determinations of physical properties.

4 did not noticeably decrease the decomposition or the flooding. By rapidly

- Measurement of Physical Properties

Samples of triethylborane and tri-n-propylborane of 99.8 and 99.7 mole-percent purity, respectively, were-used to obtain values ( table I) for the following physical properties :

Freezing point. - The freezing curves w e r e determined by m e a n s of a solenoid-stirred freezing-point apparatus (fig. 3) chazged with 10- milliliter samples. The apparatus w a s loaded and sealed with a stopper in a dry box inerted with helium; subsequent operations were conducted outside the box. The thermometric system consisted of a 25-ohm platinum resistance thermometer, a resistance bridge (Miieller type, L a #8069), a,nd a highly sensi t ive galvanometer. The freezing point and the freezing point a t zero impurity were obtaLned by analysis of the freezing curves (ref 8 . 10 and ll) . These data together wlth the values of the heat of fusion were used t o c a l c u h t e - t h e p u r i t y of the samples (ref. 11). Du- plicate determinations of the freezing point on 8 given sample differed by not m o r e than 0.002O C.

c

L

H e a t of fusion. - The heats of fusion were determined in a calibrated calorimeter similar t o the one described in reference 12. The calorimeter w a s loaded by using a 25-milliliter p ipe t te to m e t e r out the sample. A l l operations were. carried out i n a dry box inerted with helium. A low- temperature slush bath was used as EL uniform heat 60urce t o melt the

6 NACA RM ESSE06 . frozen sample, ra ther than the themoetated air bath used in reference 12. Analysis of the time-temperature curves of the sample and the heat I

source yielded a value for the heat af fusion. The estimated accuracy of t h e heat of fusion v a l u e s was s . 2 calorie per gram.

Boiling point. - The boiling points were determined i n an ebulli- ometer (ref. 13) equipped wlth a platinum resistance thermometer. The measurements were made i n a helium atmosphere with an estimated accuracy

"

of s.10 c. - Eeat of combustion. - Heats of combustion were measured i n a Parr

adiabatic calorimeter with an I l l l u m constant-volume bo& (ref. 14) . The calorimeter was modified by using an e lec t r i c heater to Con t roUhe jacket water temperature. This heater was controlled by a thermoswitch which w a s activated by temperature differences between the bucket and the Jacket. With this device It w a s possible to keep the average temper- ature variation between the bucket and Jacket to fo .05O C over the entire run. Even at the beg- of a run when temperature balance i e diffl- cult , the maximum deviation observed w&s only S.lo C.

a

Temperatures w e r e messured. with a platinum reaistance thermometer w h i c h could be read to. &O .OOlo C . The bozdb was c u b r a t e d by mean6 of 4

standard benzoic acid supplied by the Parr Instrument C o m p a n y . Because of their volatilL%y and suscept ibi l i ty to oxidation, the Bample8 of trialkylboranes were introduced into the bomb i n small glass bulbs. The couibustion procedure was the same as that used i n reference E. Complete- ness of combustion was determined by an analysis of the combustion prod- ucte, caxbon dioxide and boric acid. In general, combustion wae 97 t o 99 percent complete. Corrections *re m a d e f o r the unburned material on the assumption that it was exclusively eiemental amorphous carbon and boron. Additional corrections were made to f i t the colribuation t o the over-all process at 25' C:

The correctiona made for incompleteness of conibustion were satis- factory provided tbe conibu.8tion was at l ea s t .97 percent complete. In correcting for unburned carbon and boron, the unckta in ty an is approx- imately 0.2 percent. The uncertainty associated with the over-all raw heat of combustion frc i s 0.3 percent. For a aeries of four heat-of- conibustion determinations f o r triethylborane and for t r i -n-prqylborme, the standard deviation cr is 0.7 and 0.5 percent, respe&ively. For these determinations where a an + be, the corrections appear t o be good. When values for the heat of combustion .for runs with les s t h a n 97 percent combustion are included i n t h e data, the standard deviation

. .

NACA RM =SED6 7

% 4 0

increases so t ha t (5 >> 6, +- aC. This indicates that f o r conibustion less than 97 percent complete the assumption that elemental amorphous csrbon and boron were the only unburned material present was not v a l i d .

Refractive index. - The refract ive Fndices were measured with a Bausch and Lomb precision Abbe/ refractometer enclosed in a dry box with a helium atmosphere ( f i g . 4) . Measurements were made at 20 .O0& .lo C and 30 .Oo&Q. lo C and three wave lengths of l i g h t at each temperature. The spectral lines used were the sodium D (5893A), the mercury Hgg (54EO. 7A), and the mercury H (4358.3A). The estimated accuracy

of the measurements was M.0001. From the values f o r the refractive index and density, the m o l a r refract ion w a s calculated by means of the Lorentz-Loreaz equation.

@;b .v.

Density. - The densi t ies were measured with a m o d i f i e d Lipkin pycnometer. The modification consisted of fusFng a standard taper male joint onto the ends of the pycnometer ( f i g . 5) . These ends could then be capped with a female jo in t equipped with EL mall stopcock. Once the pycnometer w a s filled it could be sealed off without disturbing the l e v e l of the sample in the m s . The pycnometer was loaded and sealed in a dry box iner ted wtth helium. The densi t ies were measured outside the box at 20 .OOo@ .05O C and 30 .OOo_+O .05' C wfth an accuracy of kO.0002 gram per m i l l i l i t e r .

Dielectric constant. - The dielectr ic constants were measured i n a c e l l of 25-mil l i l i ter capaci ty ( f ig . 6 &nd ref. 16) . After being f i l l e d i n t h e dry box, t h e c e l l was sealed off and removed from the box fo r the capacity measurements. Measurements of the dielectric constant were made a t 20 . O 0 s .lo C and 30 .Oo+O .lo C with a probable accuracy of M -2 percent. The dielectric-constant apparatus has been previously described (ref. 17) .

Infrared spectra. - The infrared spectra were obtained with a a d doublebeam recording spectrophotometer. The precision of the instrument i s &l percent of the transmiesion value and 39.02 micron for ' the wave length. Undilutg samples were examined i n a 0.06-millimeter sodium chlor ide cel l and. samples diluted with carbon te t rachlor ide in a 0.10- mil l imeter cel l ( f ig . 7) .

RESuLT[j AND DISCUSSION

An examinatton of the values listed i n % l e I as compared with previous literature values (footnotes a t o j in table I) shows a s c r e p - ancies that cannot be explained in terms of experimental error. The cause of these discrepancies becomes evident when the trend i n these values i s examined. The refract ive index reported for tri-n-propylborane ( re f . 6) i s much lower and the dens i t ies for trFethySborase-and tri-n- propylborane ( r e f s . 2 and 6) are higher than the values obtaLned in %he

8 NACA RM E55M)6

present investigation. When trialkylboranes are oxidized, the product i s alkyldialkoxyborane, (R0)2HR, ( r e f . 18). Although no refract ive Wex and density data are avai lable for t h i s class of compounds, there m e data for the c lose ly related trialkoxyboranes, (R0)3B. The t r i a lky l - boranes have higher refractive indices and lower densit ies than the cor- responding trialkoxyboranes- (refs. 6, 19, and 20) . The allryldial-koxy- boranes would be expected t o have values between those of the trialkyl- boranes and the trialkoxyboranes . Therefore, contamination of the tri- alkyboranes by oxidation products would lower the refractive index and elevate the density values beyond those of the pure compound.

This explanation was further substantiate.d try an experiment run on a sample of tri-n-propylborane. The sample was placed in the refractm- e t e r and exposed t o air. The refractive index over a period of several secands was observed t o decrease rapidly as the sa~xple was oxidized.

CONCLUDING BEMARKS Same of the .charac te r i s t ics of the trialkylboranes prepared in t h i s

present investigation can now be examined i n r e l a t ion t o t he i r use as f u e l s i n an a i r c r a f t or missile.

As would be expected because af the .h i@ ra t io of hydrocarbon con- s t i t u e n t t o boron, the physical properties of these compounds axe In gen- eral very close .t;O thoee of the analogous hydrocarbons. In comparing JP-4 fuel with' triethylborane and tri-n-propylborane, the densi t ies of the boranes were found t o be about 10 &rcent lower and the heats of combustion about 10 percent higher. Examination of the freezing and boiling points of these compounds ind ica tes tha t the i r liquid range is adequate. The f a c t that d ie l ec t r i c propert-ies of the trialkylboranes are very similar t o thoee of hydrocarbon cmpounds should perinit u6e of the boranes with the same type of capacitance gages now used t o meter hydro- carbon f u e l s .

0

5

The aua i lab i l i ty of these trialkylboranes at the present time i B poor, since they can be prepared only i n laboratory quantities using Grignmd reactions. However, for large-scale prepmation other procedures may prove t o be feasible, such as the react ion of diborane with an o le f in (ref. 21) o r the reaction of an alkyl halide wlth boron t r i f l uo r ide i n the presence of aluminum (ref. 9) .

From the point of v i e w of se9ety in handUng these c ~ o u n d s , it must be noted tha t t he i r r eac t iv f ty with air makes them hazardous. In addition, they are most probably toxic. Individuals exposed t o small amounts of these trialkylboranes at the Lewis laboratory experienced head- - aches Etnd muaea . . .

The moet promising aspect of the lower-molecular-weight t r ia lkyl - L

boranes, beside8 their heats of conibustion, i s their e a ~ e of self-ignition. Liquid triethylbarane at room temperature will immediately ign i te spon- taneously when exposed t o air. - Even at par t ia l p ressures l ess than 1

NACA RM E55EO6 .0

9

millimeter of mercury at OO C, self-ignition occurs in oxygen (ref. 22) . This self- igni t ion phenomenon of triethylborane suggests that it could be used as a high-energy ignition source i n combustors. For bigher- molecular-weight boranes the spontaneous ign i t ion delay increaees, so that w h e n liquid tri-n-butylborane i s exposed to air there is an induc- t ion period of severai seconds before ignition occurs. Bevertheless, t h e i r high reactivity suggests that these campounds may serve as f'uels f o r a stable pilot f lame under severe operat- conditions.

-

CN (31 4 The albrylboranes, then, have physical characterist ics similar t o 0 those of hydrocarbons, but react ivi t ies with air that are much superior.

h w i s Fl ight Propulsion La;boratory National Advisory Comnittee f o r Aeronautics

Cleveland, Ohio, May 11, 1955

1. Frankland, E., und Duppa, B. F.: Vorldufige Rot iz iiber Borgthyl. Ann. - der Chem. und Pharmacie, Bd. 115, 1860, pp. 319-322.

2. Stock, Alfred, und Zeidler, Friedrich: Zur Kenntnis dea Bormethyls und Bor&hyls . Ber . D. Chem. Gesell., Abt . A, Jakcg. 54, N r . 3, bk. 1921, pp . 531-541.

3. mause, Erich, Una Nitsche, Rudolf: DazsteUung von organischen Bor- Verbindungen mi% H i l f e von Barf luorid. I. Baralkyle Una -1- borsguren. B e r . D. Chem. Gesell., Ab%. B, Jshrg. 54, Nr. 10, Nov. 1921, pp 2784-2791.

4. Laubengayer, A. W ., Ferguson, R. P .> and Newkirk, A . E. : The Densi- ties, Surface Tensions and Parachors of Diborme, Boron Triethyl and Boron Tribromide. The Atomic Parachor of Boron. Jour. Am. Chem. SOC., vol. 63, 110. 2, Feb. 1941, pp. 559-561.

L

5. Bamford, C. H., Levi, D. L., and N e w i t t , D. M.: Physical and Chemic& Properties of Organo-metallic Coqounda. P t I - The V a p o u r Pres- sures and Freezing Points of .Siyple Metal -1s of Groups II, In, ad. V. JOW. Ch" SOC., 1946, pp. 468471-

6. Pickard, P. L., and Patterson, M. K., Jr.: The Atomic Parachor, Refrac- tion and Dispersion o f Boron in Organo-Boron Comgaunds. Rep. Ka. CCC- 1024-933-45, Univ. Oklahoma, Sept. 29, 1954.

7. Frankland, E.: Ueber eine neue Reiche organischer Verbindungenwelche Bor enthalten. Ann. der Chem. und-Pharmacie, m. 124, 1862, pp. 129-157.

8. Johnson, John R., Snyder, E. R., and V a n Caqen, M. G., Jr. : Organo- boron Compounds. 111 - Reactions of Tri-n-butylborine. Jour. Am. Chem- SOC., vol. 60, no. 1,. Jm. 1938, pp. 13-121.

9 . Ruthruff, Robert F. : Trialkylborons. Chem. Aber., vol. 35, 110. 18, Sept. 20, 1943, p. 6269. (U. S . Patent 2,247,821.)

10 - Taylor, W i l l i a m J., and Rossini, Frederick D. : Theoretical Analy6iB of Certain Time-Temperature Freezing and Melting Curves aa Applied to Hydrocarbons. RP E85, Jour. Res. N a t . B u r . Standarb, vol. 32, no. 5, May 1944, pp. 197-213.

11. Glasgow, Augustus R ., Jr., St re i f f , Anton J., and. Rossini, Frederick D. : Determination O f the Puri ty of Eydrocarbons by Measurement of Freezing Point. RP 1676, Jour. Res . H a t . Bur . Standards, vol. 35, RO 5, NOV. 1945, pp - 355-373s

12. Rossini, Frederick D.: A Simple Calorimeter for Heats of Fusion. Data on the Fusion of Pseudocumene, Meitylene (a a d p ) , Hemimellitene, 0- and m-Xylene, and on Two Transitions of Hemi- mellitene. RP c07, J O ~ . ~ e s . Bat. 'BU. Standards, VOL LL, no. 4, O C t . 1933, pp 553-559

13. Gibbons, L. C ., et al. : The Purification and Physical Canstants of Aromatic Hydrocarbons. Jour. Am. Chem. SOC., vol. 68, no. 6, June 1946, pp . 1130-1131.

14. Anon.: Oxygen Bomb Calorimetry and Oxygen Bomb Coufbuetion Methode. Manual No. 120, P a r r Instr. co ., Mom (111.) , 1,948.

15. Tannenbaum, Stanley, Kaye, Samuel, and Lewenz, George F. : Synthesis and Properties of Some Aurysilanes . Jour. Am. (Bern. Soc . , vol. 75, 110. 3.5, A w - 5, 1953, p~ 3753-3757

16. Altshullm, Aubrey P ., asd Rosenblum, Louis: Dielectric Properties O f -SO= Alkyleilanes . Jour. Am. mem. Soc . , vol. 77, M . 2, Jan. 20, 1955, p. 272.

17. Altshuller, Aubrey P .: The Dielectr ic Constants, P o l ~ i z a t i ~ s , a d Dipole hments of Some AUrylbenzenes. Jour. Phys. Chem., vol . 58, no. 5, May 1954, pp. 392-395.

..

NACA RM E55ED6 11

20. Lewis, George L., and Smyth, Charles P. : The Dipole Moments and Structures of Esters of Some Fatty and Same Inorganic Acids. Jour. Am. Chem. Soc., vol. 62, no. 6 , Jmie 1940, pp. 1529-1533.

-

21. Hurd, Dallas T. : The Reactions of Diborane with Hydrocarbons. Jour. Am. Chem. SOC., vol. 70, no. 6 , June 1948, pp. 2053-2055.

22. Brokaw, Richard S . , Badin, Elmer J., and Pease, Robert H. : Wall Effects i n t h e Oxidation of Boron Triethyl Vagor. Ignition of n-Butane. Jour. Am. Chem. Soc ., vol. 70, no. 5, May 1948, pp. %

4 0

- 1921-1922

23. Schecter, William H., Jackson, C. B., and Ad-, Roy M. : Boron Hydrides and Related Compounds. Second ed., CaUery Chem. Co., May 1954, p. 185.

a-92.9a c (ref. 2 ) . f-65.50 c (nf. 5) .. b2 .4 .kceJ./mole (lief, 23) J g156° C (ref. 3); l64.5O C (mf. 5) . %sa c ( r d . 2 ) . . !ny, 1.4090 ( re f . 6 ) ~ ni2*5, 1.4155 (ref. 5 ) .

%0,1M) M.45 Bty/lb (data from W. H. Johnson and '45.3s (ref. 6). E. J. Proeen of NationeJ Bureau of Standards)

e 2s 30 'a:, 0.7639 ( r e f . 6); dC , 0.7204 (ref. 3). 24.7 d4 , 0.6951 (ref. 2); e , 0.6774 (ref. 4).

. f

.. . . .

c

Figure 1. - Distillation apparatus for air- and moisture-sensitive mterfals.

14 NACA RM E55E06

-.

Figure 2. - Bottles used to receive and store distillate.

XACA RM E55E06 A

. To vacuum

[/ Thermometer w e l l

-Stirrer

r Sample

Figure 3. - Freezing-point apparatus. -

16

c

.. Figure 4. - Physical-constante dry box. -

NACA RM E5B06

c

NACA RM E55E06

.

-1

Figure 5. - Modified Lipkin pycnometer.

17

18 NACA RM E55E06

!I 0

i. Capacitcr, four concentric nickel c yUnder 13

Figure 6. - Cell used to mea8ur8 dielectric constants. -

1 A

. . ..

, 4

a f r’

- -sa. P (b) !ki-gprogylborane (2-Pr3B).

Figure 7. - Infrared spectra.

- .. . . . . . . . ..