<%BANNER%>
HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Materials and procedures
 Syntheses and reactions
 Discussion
 Bibliography
 Biographical sketch


UFIR








F" gdre Page

20. Tlehyiene region of the H nmr spectrum of (CH3CII NUEHBHC 62
22. M of ,

21. Infrared spectrum of (CH3CH2 )INHBHC12 . ....... 63

22. Methylene region of the -H nmr spectrum of (CCi 2NHBCI3. 64

23. Ifrarc-d spectrum of (CI3CH2) 2IBC3 . .. ...... . 65

24. Methylene region of the 'H nmr spectrum of (CB- CI 2) NHBH~ r 67

25. Methylene region of the 1 nmr spectrum of (CF3CH2) 'NHHBlr2 6S

26. Methvyene region of the H nmr spectrum of (CH- CH2) lHBBr,. 70

27. Infrared spectrum of (CH3CH 2NHBBr3 .. . . . . .. 71

28. Methylene region of the 1H rinr snactrun. o (CH CH NHBH.J 72

29. Infrared spectrum of (CH3CH2)0NIBt{2 I .......... .. 7.1

30. Methvlene region of the 11H nmr spectrum of (CH3CH2) NHBBII2 .. 75
1
31. Methylene region of the H nmr spectrum of ,'CH C l2) 2 F . 3 77

32. Chemical shift of methyl protons in haloborane adducts of
tCH3)'NII and (CH,),N as a function of extent cf halogen
substitution on boron ... . . . . . . . .1

33. Chemical shift of nmtehyl and methylene protons in haloborar.e
adducis of (CH3211i2)2NI and (CH3CH2)3N as a 3functiun of c:tent
of h-irgene substitution on boron .. . . ..... .... 113

3-. Sketch of approximate bond lengths, bond angles, and van 6ter
Wiaale' lrdii for an H-C-N-B--X s stem (X = Cl, Br T) showing
the potential for severe overcrowding ............ 121

35. Chenmical shift cf non equivalent methylene protons in haloborane
adducts of (CH3C 2)9NH as a functici of extent of halogen
substitution on boron . . . . . ... 12

3G. Nowman projections si.coinag simi lrity between C3H CH2CXYZ
SCHC2
arc GB C








Figure P1, ge

37. Newman projections showing rotameric isomers of
(CH3CH )2 NHBXY2 (with torsional preferences) . . .... . 128

38. Newman projections of two rctameric forms of
CH CHi2OC(O)CH2N(-)(CH ).BCI3. . . . . . . 131

39. Newman projections showing possible non-equivaIlcnce of methyl
groups in trimethylamine-haloboranes produced by hindered
rotation at low temperature . . . . . . . . 134










Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy

TRENDS IN THE PROTON NMR SPECTRA OF SOME
AAUNE-HALOBORANES: STERIC EFFECTS

By

William Howard Myers

March, 1972

Chairman: Professor G.E. Ryschkewitsch
Major Department: Chemistry


Boranc adducts of triethylamine, dimethylamine, and diethylamine were

halogenated using free halogens or hydrogen halides, and the proton nrmr spectra

of these amine-haloborane adducts were obtained. The resonances of these

aducats and of the haloborane adducts of trimethylamine showed a shift to

lower field with increased size of halogen or with increased number of halogens

on boron. This shift to lower field had been previously attributed to inductive

effccts, biut in this work it was shown that the shift to lfwer fie1' was duo to

steric interiac.on betvcen halogens on boron and alkyl groups on nitrogen.

Proto, nmnr spectra for diethylamine-baloboranes vere complex and

showed pftt'nis attributable to non-equivslent methylene prators. Computer

analyses of the spectra allowed assignments consistent wiCh prtfe: re-.d .r-lational

configurations.

The reactions between the secondary ninen, dirnethylamaine, and three

bcron trihalidecs were examined in detail. DimethyrtAnnie and bo-op triflr'oride








reacted to produce the 1:1 adduct, dimethylamine-boron trifluoride.

Dimethylamine reacted with boron trichloride or boron tribromide to produce

mixtures of both the 1:1 adduct and bis-(dimethylamine)-dihaloboronium

tetrahaloborate saks. In order to obtain pure 1:1 adduct, it was necessary to

take advantage of the low solubility of the boroniun salts in inethylene chloride

or to directly halogenate dimethylamine-borane.

During the reaction of N-deuterodimethylamine-borane with chlorine,

considerable exchange of hydrogen for deuterium on nitrogen occurred. An

extensive investigation of the halogenation reactions of N-deuterodimethylamine--

borane indicated that the exchange process occured'only via the loss of

deuterium chloride from a molecule activated as a result of halogenation.








CHAPTER I

INTRODUCTION


In the last decade, a variety of research techniques have been used to

examine the relative acid character of various boranes and boron halides. On

the basis of electronegativity and relative steric requirements of the halogens,

the accepted order of acid strength1 had been BF,,>BC13 >BBr3, but the evidence
3

presented in these later articles indicates the reverse order to be correct.

Laubcngyer and Sears2 were the first to suggest that BF is not the
3
strongest acid of the boron trihalide series, as the result of a calorimetric

study of the formation of the BF and BCI adducts of acetonitrile. In a lter
3 3
article, Miller and Onyszchuk3 added to this work with a calormetric examination

of the formation of the BBr3 adduct of acetonitrile. In a similar study, Brown and

HeIlmes made calorimetric measurements on the formation of pyridine-boron

trihalide adducts, which indicated an order of acid strength of BF 3 .. 3- 3
Stimulated by this new evidencE,Cottirn and Leto1 performed -olecular orbital

calculations of the reorganization energies involved in the formation of dcnor-

acceptor molecules by boron trihalides. The theoretical evidence showed that

the major iinrdrance to adduct formation for a boror trihalide involved the break-

up of rF-bc're,... between boron and the halogens, and thus that the order of acid

ot:-ength was d,-iermined by relative extent of 7r-bonding rather than by rolativa

e-ectrona.'atilty or relative steric bulk. That is, F3, is --he poorest acid of the

series because it has the mrost -bonding. to break up.








Additional support came from the work of Bax, Katritzky, and Sutton, 6

who determined the dipole moments of boron trihalide and borane (11 ) adducts

of trimethylamine and pyridine.

In two related articles,7, 8 Onyszchuk and others reported on a study of

the gas-phase displacement reactions of boron trihalides from their trimethyl-

amine adducts. The order of acid strength was shown, unambiguously, to be

BBr3>BC13>BF3>1/2(B 2H6) Coyle showed that this displacement can occur

with or without B-N bond cleavage.

In the recent literature, two articles have appeared giving structural

evidence which confirms this order of acid strengths. Ibers10 reported the

X-ray structure of CHCN-'BF3 and CH CN-BC13 The B-N bond length shortens

by 0. 07 A going from the BF to the ECI adduct, implying a stronger B-N
3 3
bond in the BC1I adduct. Bryan and Kuczkowski11 reported on the B-N bond

length in (CH ) NBF as determined from the microwave spectrum of the compound,

and the complete crystal structures of the other (CH3 3NBX3 adducts (X = Cl, Br,

I) are to be published by Taylor.12 In addition Bryan and Kuezkowskii reported

a B-N bond length for (CH3)3NBH3, as determined by Schirdewahn13 from the

microwave spectrum of the compound. These bond lengths are given below:

Compound Bond length (

Ct 3(NBF3 1. 630 ( 0. 004)

CH3CNBCI3 1. 562 ( 0. 008)



CH.)NBH3 1.65 ( 0, 2)

(CH )JNBF, 1.636 ;:0.004)









Compound Bond length (A)

(C 3)3NBC13 1. 610 ( 0. 006)

(CH3)3NBBr3 1,603 (- 0.02)

(CH3)3NBI3 1. 584 (+ 0. 025)

The B-N bond length is seen to become shorter as one goes from BF3 to

BI3 as the coordinating acid. This is consistent with the strength of interaction

(i. e,, acid strength of BX3 towards a given donor) increasing in the order

BF3< BC3
would expect the B-N bond to be shorter for a given (CH3) 3NBX3 adduct than

for a corresponding CH3CNBX adduct. The data above show the B-N bond

length to be about the same for the two BF3 adducts, and significantly shorter

for the CH CNBC13 complex compared to the (CH3)3NBCl complex. One

possible rationalization for this anomaly lies in the fact that nitrogen in CH,,CN

uses an sp hybrid orbital to bond to boron while nitrogen in (CI ,) IN uses an sp

hybrid orbital. A consideration of some simple organic molecules predicts a

shortening of 0. 05 to 0. 08 A for a change from sp3 to sp hybridizatlon for one

carbon in a C-C single bond. Thus, CH CNF3 and (CH'.,lNl':3 ,iuld have

similar B-N bond strengths only if the B-N bond in C-C,r'NBFT wevre at least

0.05A shorter. (Valence force constants were determined for di- -N bond in

(CHf)S.NBF by Clippard14 and in CH-CNBF. by Sw'anson and Shriver,15 The


3 3
values are ?;.53 mdynes/A [(CH,.,)NBY,,j and 2.5 mcynes/A (CTI3CNF3).

demonstrating thai the inteirction in :C.FIs 3N1T? is inrdecd much stronger.)

One wou'd also ha-,e expected the c-,ange frcm UPF to EC!3 to have affected the

two adducis niiilarly, yet the B.-N oond in the CHE CNECI3 adduct is shortened
u J









by 0.04 A more than the B-N bond in (CH3 )3NBC3. One might conclude that

ateric interaction (F-strain) in (CH3)3NBCI3 precludes the B-N bond from

shortening as much as expected. One should note that. within experimental

error, the B-N bond length is the same in the (CHII) N adducts with BC13, BBr3,

and BI although bond shortening would have been expected. This is also

consistent with steric interaction (F-strain) interfering with B-N bonding.
12
Taylor12 comments that these three adducts are sterically quite crowded. In

fact, a comparison of the intramolecular non-bonded distances with the sum of

the van der Waals radii shows that the X-CH3 distance is shortened by about

0. 63 A; this implies a considerable interaction.

Having determined with good confidence that the order of :cidily in the

boron trihalides was BF3< BC13
to correlate spectral properties of donor-acceptor complexes with acidity. The

basic tenets for three major areas of work* are set dcwa below, with references

to pertinent articles:
14-20
1. Infrared Spectral Shifts. These correlations are based on the effect

of the acceptor group on electrcn deniji., in, the donor group. For example,

coordination of an c.id group to a carbonl. gou:p removes electron density from


fr'0th type of spectroscopic i., e.tigc'aic. o donor-acceptor comipoulds
has recently been reported by L;it'.le ard \ii.c.r.. Th' -'y exaninod the mass
spectra c'f trimethyJamiae adeucts o tlie' bocun i-tlhaiddes and some mixed
trihalides and found that the results cold be texok.ined in terns of relative
acceptol strength of the boi on triha idtis ani' borsii'e towards trinmth'ylaminie.
The re oulti are consistent with tlhe LeW'"s acidity increasing with size of the
ialogein, ,nd evidence is present which indicates decreasing boron-halogen
bond rlrcrgtih wi'l increasing size cf halogen. This i a trend is rdicussed in
terms of "' -sidcal ,,-bonding" between boron and hcalce n i; the am'ine-borane
nioleccale. (For a full discussion of this ;ast point, see Tl-rown, ri.oqr, aod
Bolles. 7 )










the oxygen, thus polarizing the C-O bond. The electrons in the n--systenm ;.aourl

be more localized on oxygen, and the'C-O bond strength would decrease.

Another way to express this idea is to say that there is an increased contribution

of a resonance form which has the r-bonding electrons localized on oxygen, and

the formal positive charge localized on carbon. The lowering of the C-O bond

strength would cause the C-O stretch to shift to lower frequency, and the

magnaitu'dLs of the shift is a measure of the strength of interaction, i.e., a

measure of the acid strength of the acceptor group. Of course, the real

variable one is attempting to evaluate in this manner is the force constant of

the bond, and one must be careful to establish that the bond whose force

constant is to be qualitatively examined by this method has its stretching

motions isolated from other molecular vibrations. If fte stretching motion

is stroingi enupled to other motions ti the molecule, problems in interpretation

nma arijs:, A case in point is the B-N bond i ; CH CN3BX adducts, which is

highly coupled to other motions in the molecule. This system has been examined

Dv Eeattic and Gilson.21 Purcell and Drago,22 and Swanson an.d Shriver.15

An interesting correlation is reported by Coyle ad Stone23 for donor

-trenetb i: adducts containing B-H: bonds. This corcilai;o- is obsed on :Cotoen

a-nd Letu's aricmelin of reouraunizaLnon. ener-.,' and u-.ljgeets that the more

*.ibl a: .;' .i ti't i-, the move tet-aihc-'ra! t:e g-oups aro ,u:i the donor and

cr.cet:i.r atoms should he. Thus, since the s character cf i-HI b-.dms in such a

vst'r:.: v;i b? less ii a tetral-edral case (s ) than iin a r-ase where the

2
cue.Ctus ,mnoi2,- approaches plan:.r ty (sp ), their B-H boid should be veakoer

in the :more strongly bonded adduct; that is, the B-H stretch should shift to









lower frequency. The data presented by Coyle and Stone are, for the most

part, con.sis'tern with this picture. However, a more general point of view

would be to lok at the system as a B+3 ion coordinated to three donor groups

(X, Y. :;ad Z) and hydride (H ). Changes in the donor groups will affect the

stretching frequency of the -IH bond in a predictable manner, namely, the

more electron-releasing the donor groups become, the more electron density

will be pushed onto the hydrogen, and the more the B-H stretch will shift to

iower frequency. Conversely, the more electron-withdrawing the donor groups

become, the more electron density will be pulled away from the hydrogen, and

the more the B-H stretch will shift to higher frequency. In other words, the

B-IH stretching frequency should be a measure of the "hydridic character" of

the boron-attached hydrogen. Thus, the B-IH stretching frequiircy in a series

of (CH ) NBIX adducts (X = Cl, Br, and I) may be explained on tlus oasis.

The values are as follows:
-1
Compound v H(cm ) Reference


(CHi NBIC12 2480 24
3 2

(CH ) NBH.Br. 2510 25

(CH 3)NBHI2 2525 25

One z:ayv co-.clude that the act wi.':dratnal of clct.o. denslt. by : in (CiHI)3NBIIX2

is in 'the ord.; C
odtr.eigh elcefr'negati'iLy effects, even in tetrlahcd.ual boron. (For :; full
17
iJscust-;n of thi.s lst point, see Brouwi, Drago. and Beoles. )

'. ? i. 1MR1Data. There bhae been two ma jo' attempt. to correlate
li 23 11 a
B nomr 'ndata with acrdity. Mutties s the f o g t that lu -
nndt. i I t'. nH "ste tL'01









coupling constants, in compounds containing B-H bonds, might be a measure of

donor-acceptor interaction, on the basis that the value of the coupling constant

is related to the amount of s-character in the boron orbital used in bonding to

hydrogen. (The idea originated in the study of 1C nmr data; see, for example,

Gutowsky et al. ) One can rationalize the idea as follows. The greater the

acidity of a given borane, the more easily the borane "reorganizes" to

tetrahedral geometry. Since the more tetrahedral the borane, the less s-

3 2 11 1
character in B-H bonds (sp vs. sp ), it follows that if the B- H coupling

constant is directly proportional to the amount of s-character in the B-I bond,

the coupling should decrease with increasing acidity. Conversely for a given

horane, the greater the base strength of the donor group, the more tctraliodral

the borane, thus the less the s-character and the less the coupling constant.

In practice, this concept works only for limited situations, that is, for

systems in which the donor group is varied, or in which the borane gioup is

varied within limits. A case in point where the correlation does not hold is

12 1
in the case of (CH) N coordinated to BH3, BHI2C1, and BIC1 The B- H

24
couphng constants for these three adducts are 101, 12, and 132 iHz, respectively.2

The acidity of these groups surely increases in the order PlII <32 C1
but the J, values indicate the reverse order,
3- H
A raioonaiization of sorts is obtained by con:sijdeo;ing tii; various factors

which affect the value of -TI Grant anad '',,chmai pointl out that 1J
B- -i C- I
values are not only a fr.nction of ire!ati.e s-c-ha;a;ctu bat -aso of effective

n,:!'ear ]b-arge at the carbon:


T r" (i-Ft'ivc' v --1 racier' relativee, effect ,e nuclear charge Id in CH.).
ole' 4w '"'" '









Thus, the correlation of J values with relative s-character works only when the

relative effective nuclear charge is constant. When a system is considered in

which great changes occur in effective nuclear charge at carbon, one cannot

use J values as indicators of relative s-character.

In the system described above [(CH ) N coordinated to BH3, B2 C1, and

BIIC1 1 the changes in effective nuclear charge at boron as a result of chloride

substitution have apparently swamped out any effect ot relative s-character.

That is, increasing the effective nuclear charge at boron will increase the

value of J no matter what the effect on relative s-character of the boron
11 1
B- IH
orbitals used to bond to hydrogen.

The change in B-I! stretching frequency in these adducts supports this

33
argument. The frequencies33 (symmetrical stretch) are 2260-2290 (BHI3 adduct),
-1
2340 (B1 2Cl adduct), and 2480 (BHCi2 adduct) cm- The trend in frequencies

implies a lowering of hydridic character in the order BIl,>BH C1>BHC12 which

would be expected if the acidity increases in the order BH 3
It should be noted that a rough correlation of structure to J, can be
11 1
B- H
made for tetracoordinate boron as shown below:

S ect cr, Approximate J in 11z
~~~"--- ------------11 1
B- H
BH, 80

X,,IZ XBF 3 100 K = N,O donor

borai-,ies, XYBII7 XFXYBH 120-140 Y Cl, Br, I

oLoxwines, XY BlH 10.-170

'The predominant factor woulj appear to ',be eci-...egtivity, or, in other words,

for a wide variety ot compounds, J is a I'incuion of relative effective

i'-jci.e;: r Chc"r-c at boron.









The second major attempt to correlate 1B mnr data with strength of
97a-a
donor-acceptor interaction was based on Mooney's idea that the change in
11
SB chemical shift from uncomplexed to completed borane is a good measure
28
of the strength of the donor-acceptor bond. Heitsch directly contradicted this
11
.idea, pointing out that the order of basicity as determined by B chemical

shifts for a series of amines coordinated to BH3, BF, and B(CH3)3 cannot be

rationalized on the basis of any known or assumed order of complex stability.
29
Mooney, in a review article, suggested that the data IHeitsch compiled might

be suspect because of the solvents he used -1, 4-dioxane and acetonitrile. In

addition, Mooney sidestepped the issue by suggesting that the orders of basicity

which Heitsch's data imply might actually be the true orders of base strength

towards the reference acid in question.
27a 11
Mooney has given B chemical shift data for triethyiamine adducts
28 11
of BX,,(X = F, Cl, Br, I). Heitsch gave a value for the B chemical shift
2,
26 11
of triethylamine-boron trimethyl, and Muetterties2 gave a value for the B

chemical shift of uncomplexed trimethylboron. These data are given below

(all 6 relative to external Et20 BF ):

Borane i5 of free borane 6 of Et3N adduct A5 (after Mooney)
-E --- -B-

BF ----- 0.2 + 9.5




BBr3 + + 45. 1

B3 ---- +59.8 + 53. 1

n(CH3)3 -86.4 +13. G +100.0









Thus, according to Mooney's criterion, B(CH3)3 would be a stronger acid than

any BX3 (X -' halogen) toward Et3N. This perhaps conclusively demonstrates

that Mooney's criterion is not universally applicable, and thus that care should be

;iken in its use.
30
Although the work of Spielvogel was not directly concerned with the

question of strength of donor-acceptor interaction, it is nevertheless of such

significance that it should be briefly discussed. Essentially, Spielvogel attempted

to correlate 1B chemical shifts with the nature of the substituents on boron.
13
Such a correlation has been relatively successful in the study of 1C shifts, and

the idea that the same type of correlation exists for B is not surprising.

Briefly, the correlation works as follows. So-called ''"p-ir wise anditiity

parameters' are determined for possible substituent combinations. Four

substituents may be grouped into six groups of two substituents each. For each

group, a contribution to the shift is assumed, and the sum of the six contributions

is a good measure of the 1B chemical shift of the molecule. As an example,

consider Et NBHBr The four substituents are Et N (iactu:.l11. y i for an
3 2 3

approximate calculation), , r, and Br. The six groups, are then: [rEt N, ,

[EtN, Br), [Et3N, Br], [H, Brl, [H, Er], and [Br, Pr1. The c.aribiutions for

these groups are (in ppm fromn Et 0. BlT):

Group COpt:ribmUtn to 1, (pM

[Et3N, H] -1. 6

[Et3N, Br] -4. 57

([E Br] + 4.52

MBr, Br] +fi 27










The chemical shift is then calculated to be:

6 =S[X, Y] = (-1. 63) + 2(-4. 57) 2(4. 52) -(6. 27) = +4.54 or 4. 5 ppm.

In the present work, the observed value for the 1B chemical shift of this

compound was 2.1 ppm referred to the same standard.

In this manner Spielvogel also calculated 6(Et 3NBHI2) = 39.4 ppm, while

the observed value in the present work was 6 = 24.3 ppm. The agreement in

this second case is not so good; however in most cases the agreement between
11
calculated and observed values for the 1B chemical shifts was very good,

typically with less than 1.0 ppm error.
1 D 34-38 34
3. H NAR Data. Coyle and Stone were the first to attempt to

correlate H nmr shifts to donor-acceptor properties and adduct stabilitis.

They examined a series of donor molecules (which contained -CH CH3 groups)

coordinated to BI13, BF and B(CH3)3, to see if any correlation existed between

the stabilities of the adducts formed (as determined by other means) and the

"internal shift" of the ethyl group in the ligand. Although in ail ive of the cases

studied (EtN, EtlMe2N, Et2S, Et20, and EtMeO) no such correlation was found,

the idea, that a correlation might exist between donor-accepto;: properties and

proion chemical shift, encouraged other research groups to examine other systems.

Hohnesi and Carter35 examined the reaceion of a series of dimethylamino-

ethylphosphiines with triethyiborane. The relative e se strengths were determined

by a displacement study, and the proton nmr data were examined for correlations.

Tt was fo.md that the change in chemical shift of the boron--attached ethyl


The term internal shift" refers to the sepiiation of the- alsorptioLn duo
lo the methylene and mothyi protoe-s of an othsy grouc, and is taken as positive
vwhen tha methyl absorption is nt I'.i.er fie'l,










group correlated well with adduct stability:

order of base strength as determined by displacement 6 BE(ppm)*
reactions

(strongest) (CH ) P + 0. 35 (farthest upfield)
3 3

>(CH3 3N + 0.31

(CH PN(3 CH3)2 + 0. 29

> (CH3)3P[N(CH) 2 1 + 0.21

(weakest) > P[N(C3)2j3 0. 11 (least downfield)

This correlation is explained in that coordination of B(Et)3 to a basic center

should increase the electron density on the ethyl groups attached to boron,

causing an upfield shift of 6 B t. and that the more basic the ligand attached to
B-

the borane is, the greater that upfield shift should be.
36 1
Miller and Onyszchuk reported the H nmr shifts of the (CliH)3.; iducts

of boron trihalides and borane and showed that the chemical sliifts correlated

wih ihe previously determined order of acceptor power: EBr3 > BC]. > BF >
3 3

1/2 (B2 ). In a related study, Miller and Onyszchuk reported37 the 1H nmr

s]l fts of the CH3CN adducts of boron trihalides and showed that these chemical

233
shifts also correlated with the order of acceptor power. 2,3 Ali those values

are shown below:

2 in BX, 6 of (CII3) N adduct t 5 of CII3CN adduct**


I- 159.5
3. 39
0 i- determined by the miomicot m-;theo; referenced to cyclohcxane.

.L6 =6 A positive value ir'.c;tes an unfield shift.
.-Et addwat free acid
t Values inr Hz, downfield front internal '.IS, in CHCI3- solu;ilcn.
** Values in THz, dowi:field from internal TMS, in CH N.:O solui.oin.
11 -










X in BX3 0 of (CH3) N adduct 5 of CH3CN adduct


F 157.3 161.8

CI 180.2 182.8

Br 190.8 189.9
37
I 201.6

The idea in this correlation is that coordinating an acid to a base causes a net

removal of electron density from the donor, thus deshielding protons in alkyl

groups in the donor moiety. The greater the acceptor power of the acid, the

greater should be the downfield shift of the alkyl protons in the donor group.

Other H inmr data on donor-acceptor complexes are given in papers by
23 38 40
Heitsch, Hartman and Miller, and Derek, Clagueand Danti. Other papers

which discuss donor-acceptor properties in terms of 1H nmr shifts include

those by Young, McAchran, and Shore,41 Massey and Park,42 Cowiev and Mi ;s, '

and Clippard and Taylor.44

The correlation of relative acid strength of boron acids with changes in

the 1II ninr shii't of protons in alkyl groups in the donor portion of donor-acceptor

complexes formed between the acids and a reference base .3 perhaps a difficult

idea to accept. The basis of the argument is that the effect (a do\wn eld shift

vwhen stronger acids are coordinated) iE d dShieldiug due to L:It indjctite with-

drewiog power c: :he acid. This argument clearly applies in those cases where

there are onl5 siimpie cr-bonding interaeiions betscen donor and acceptor.

FHvever, the etfect of coordinating BX ,groups !X = 1, C'. Br, I) it oioosite to

Iuit expect-' on. the hbsis of the trend in electronegativity of the halogen substituents

that is, the n- t deshielding occu:.s when Y -= 1, and the least w-het X -= ). As









stated earlier, the effect has been attributed to the fact that the acid strength

increases in the order:

BF< BC13 < BBr3 < BI

and thus that the inductive withdrawal effect of the BX3 group on coordinated

door groups should be in the same order. To evaluate this argument let us

consider in detail the reason that the order of acid strength is BF BC1 3 3
Bi3. One can distinguish three major factors:

1) The order of electronegativity of the halogens is F > Cl > Br > I.

On the basis of electronegativity alone, one would thus predict an

order of BF3 >> BC1 > Br3 > B3. This order is opposite to

the order observed.

2) The ability of the halogens to r-bond to boron (in the free acid)

increases in the order I < Br < C1 << F. If Tr-bonding in the free

acid must be destroyed to form the adduct, the strongest acid would

be the one which has the least T-bonding. Thus, on the basis of

extent of r,-bonding, one would predict an order of BF3 << EC3 <
3 3
.BBr < E13. This order is the same as the order observed.

3) The rCeati e storic bulk of the halogens increases in the order

< CI < Br < I. Front strain in the adduct is a destabilizing

influence, and one would product that the acid strength, if

dete'miied by staic bu

BkC1a3 Br, > BI This order is opposite to the order observed.

IF the -:ctuai order of acid strength is determined by these three factors eoe must

cnclud:? 'hat the predominant offoct is that of factor 2), :nc. o.nly factor 2)









correctly predicts the order of acid strength. That is, the energy required to

break up nr-bonding in the free acid determines the trend in the relative acid

strength of the boron trihalides.

It should be obvious then that if all of the i-bonding in the borne is

destroyed in forming the adduct, the inductive withdrawing power of the

coordinated borane would be determined by the other two factors, principally

by the electronegativity of the halogens. Thus, the inductive withdrawing power

should be in the order:

BF3 > BC13 > BBr3 > BI3

To the extent that there is residual n-bonding or "memory of the free acid" in

the acceptor portion of the adduct, this order may be modified somewhat, but it

Is difficult to believe that there could be enough residual 7-bonding in the adduct

to cause the order of inductive withdrawal power of the coordinated borane to be

the reverse of that predicted by electron gativity considerations.* Cue must

conclude therefore that some other factors besides inductive effect are at v ork

in determining the chemical shifts of these adduets. It is the purpose of this


It must be pointed out that this argument is not without conflicting
evidence. A'rgvn'ents in favor of residual n-bonding in these donor-acceptor
molecules have become common in the recent literature, and certainly
experimental evidence exists supporting the idea of such residual 7-bonding
being a factor in thicss. systems. The argu ,ent for residual r-bonding is best
put forward by Browil, Drago, and Bolles' additional experimental evidence is
presnentd by S afnsoh, Shriver, and Ibersl Clippaid, Hanson, and Taylor,12
and Lanthier and Miller16 Nev'ertheless, even if residual n-bonding does play a
role in the bondring of thC se adducts, one should expect the opposing effects to
cancel oit in all the. hRl.noranca e:'cept perhaps BF The fcet that large changes
i.i chemical shift occur between BCil3 can BBr3 and especially btwcee" BBr3 and
.13 coordinated to a reference base would seemu to, clearly indicate ihat the
ex.l"aiioa n of tlhAe trend in chemical shifts in teris of inductive withdrawal effects
is at ieat. incoimpiete, if nort ai ogethier *,'foing.




16





work to present evidence which supports one such factor steric interaction

between the halogens on boron and protons on alkyl groups in bases coordinated

to boron.












CHAPTER II

MATERIALS AND PROCEDURES


Materials

Trime hylamine-borane, dimethylamine-borane, and triethylarnine-borane

were obtained from Callery Chemical Company and were used withouL further

purification. Other chemicals and solvents were reagent grade and, except

where noted were used without further purification, but were dried over

Molecular Sieve 3A where appropriate.

Apparatus and Methods

Infrared spectra were obtained using a Beckman IR-10 spectrophotometer.

Solid samples were prepared as KBr pellets; liquid samples '.were ':r.ep ed as

liquid films between KBr plates.

Melting points were recorded on a Thomas-Hoover apparatvs and were riot

corrected.

Ele~iental analyses were performn- by Peninsular Chnc.-lie.-arch. Irc.

Gainesville, Florida.

Exceap heree ioted, prcto'n ; amr apectra of lte amine--boran'es indI horlboranes

,e.re record je. t amioi.er't tei .rp.Cc:a ;re,.: ci HCi sol0 :noa with :._U' teth".iiys. iane

as an intter ,! ?- a).!iiuard. C oncea- ir.' t<-s woire typically ":.,t the range 0, 25 to 0.65 tM,

altho;.-y sC.ote ,cf th& dieihy.aiine-hLaloborane spectra wv.Cer cobt0iuc ,n more

coce';!-a;cn :ourio)'ns. The spectra were obtained! at .i V1H., U'ao Ver,',a

Model A60--A E specfi .oa ter. The c'hem i shif> rep.' r d in Pi:., lw,,fiedF

from ilteraal tetraniethl.iiia.ie. in to i.ases, to : e n -tii'u d h',te-. sW trt









were also obtained at 100 1MHz, using a Varian Model XL-100 spectrometer.

Calibration of the spectra of the diethylamine-haloboranes was accomplished

by means of sidebands generated from thie peak due to internal totramethylsilane.

Computer Analysis of 1H NMYR Spectra cf Diethr-lmine-!laloboranes

The -H nmr spectra of the diethylamnine-baloborayre adducts were analyzed

by means of the program LAMP2, a version of LAME ,45 written for an IBM

360-65 computer with a Calcomp plotter, and kindly supplied by Dr. R.W, King

of the University of Florida Chemistry Department. LAME is a modification of

LAOCOON36, 47 to include magnetic equivalence tactoring, which allows a

group of magnetically equivalent nuclei to be treated as one nucleus of spin equal

to the total spin of the group, as regards interaction with th'r cur.lei, but which

weights the energy level populations for the g o up according to the bin,,mmial

expansion. This modification greatly simplifies tl:h calculation in cases where

such groups of magnetically equivalent nuclei exist.

The program JAMIP2 is made up of three different subroutines: non-

iterative, iterative, and plot,

The non--iterative subroutine has the capability of causing the computer to

generate, from an arbitrarily chosen set of chemical shifts and coupling constants

for a system of two to seven gr-oups of magnetically euiivao!ent nuclei oC spin-1/2,

a tabic of ft'requeaies and inrens;ties of the lines expected in the nrr spectrum.

'ihe program uses the input data to former and solve tbh. l.amiltonian, matrLx for

the stationary state energies of the systein, follovi'ig the pl.'ceeduren, outlined iy

Pople, Sch.ieidcr, and. Bernistein.48 Thll frequencies of ell.owed transitions

correspond to nernigy differences of the station r"y stafnt giving rise t: th- fralr.'tiponi










If a spectrum thus calculated bears a recognizable resemblance to an

observed spectrum, the iterative subroutine can cause the computer to perform

an iterative calculation by means of which the calculated freouencies of assigned

lines are brought as close as possible (by the least-squares criterion) to the

corresponding observed lines. The chemical shifts and coupling constants

which yield this best fit are then printed out along with information about the

expected error in them and along with a table of observed and calculated line

frequencies, calculated intensities, and errors in fitting the frequencies of

observed lines.

The procedure involved in the iterative subroutine is as follows. The

partial derivatives 6v./6o. are fund in the region of the input parameters,
1 1
'hcre 6v'.,/ip is the rate of change of the i-th line position with respect to the
1 3
j-th parameter. For small, finite changes, one assumes that Avi = (6q/6p)Ap..
1 1 J J

During each cycle of the iteration the program seeks to correct each parameter

by t-he amount required to produce agreement with experiment. This produces

equations of the form, T (6vi/6.)Ap. = (v b- v ) fcr each selected line
j i 1 obs calc k
:-ncitio:n :.. In matrix form this group of equations becomes DA=N, where D is

the (itij) matrLb of partial derivatives, A is a row matiix; of corrections to the j

paat.etprs, andr N is the column matrix of i line position errors. As long as

i > j: the equations are overdetermined,and if the ini'ipl guesses are good, the

icogram is strongly convergent. A best least-sr;uares fit to observed line positions

is sought, teathcr than exact agreemem.

it should be emphasized that iterative programs typically seek convergence

either on energy levels or on transition frequencies (i. e differences between.









energy levels). The program used in this work iterates to convergence on

transition frequencies.

The plot subroutine can cause the computer, usiug the Calcomp plotter, to

generate a calculated spectrum, based on Castellano's hybrid line shape

function.49 whichh includes a histogram showing exact line positions.

The spectra of these diethylainine-haioborane adducts were analyzed as

ar ABCD3 spin system (A,B = ron-equivalent methylene protons; C = nitrogen

proton; D = methyl protons) since the full eleven-proton system was too large

for the program as currently dimensioned. Little error should be introduced

by this approximation, since any cross-coupling between the ethyl groups

should be exceedingly small, and since the nitrogen-attachoe proton is only

weakly coupled to all ethers (J/A 6 does not exceed 0.1). The system

approximates to ABMX3.

The 100 MIIz spectra of the dibromo- and monochloroborane adducts of

diethylamine were obtained, and analysis of these, made simpler by the spreading

out of the patHcrn at the higher field strength, allowed subsequent assignments

in the GO MHz spectra to be made with more certainty. The initial assignment

of parameters was made as follows. Values of 6 for the various protons were

estimalcre by visual approximation. The centers of mass of the two halves of

the met: ylene pattern were used as initial estimates of 2(A) and 6 B).
C,22
The tv.o J 's were assumed to be appro:dmatel\ equal. since,in every
CIi -CH
2 3
case, the methyl resonances were 1:2:1 triplets; the value used wv:;s the spacing

of the triplct. ',Ilues for J weree assume to be either zero, by visual

observation, or -,bout 5Hz. by analogy to the rlimPthylamine-halobo'ane result;.










Values for JB were estimated visually on the basis of spacings in the methylene
A B

pattern. Signs for the coupling constants were assumed to be positive for the

3-bond constants (J and J ) and negative for the 2-bond constant
CH CH NH-CH
2 3 2
(JAB) by analogy to carbon systems. After one spectrum had been fally analyzed,
AD

assignment of parameters in other spectra were made by a straightforward

comparison of patterns and/or by visual estimations.

The spectra of diethylamine-trichloro- and tribromoborane were poorly

resolved, probably due to the added feature of coupling to boron, and as a result

it as not possible to carry out full analyses on these two spectra. Approximate

values for the parameters were obtained by simulating the spectra. The

accuracy of these results is obviously somewhat lower than those for the six

mono- and dihaloborane adducts, which results NN ere obtained by rigorous

analysis.

No fine structure was observed in any of the NH resonances, probably on

accountt of quadrupolar broadening by coupling to nitrogen and/or boron.

Accordingly, approximate chemical shifts were assigned to these resonances, and

i:hese shifts were not refined.

On the average,20 of 48 lines (and in no ease iess than i7) were assigned

in the nmethylene region of a spectrum and were given equal weight. In each case

16 of 32 lires were assigned in the methyl region. The root-nean-square error

or observed v. calculated transition frequencies did not exceed 0.25 Hz and

average 0. 16 z:. Geod agreement was found between observed 60 MHz spectra

and calculated spectra as plotted by the computer. These observed and

cr lculatod spectra are shown in the expieimental section.










CHAPTER III

SYNTHESES AND REACTIONS


The syntheses of amine-haloborane adducts examined in this study were

accomplished by following methods established in previous work in this laboratory,

which involved the preparation of mono-, di-, and trililoborane adducts of

trimethylamine and pyridine via halogenation of the respective amine-borane.2425'50-52

In that work it was determined that pyridine-borane was more reactive than

trirnethyanmine-boranc towards various halogenating agents. The reactions

of interest to the work discussed here are shown in the equations which follow


(TMA = (CH3 3N, py C5H5N):

(1) T:\Bil3 + HX

(2) TLMA1I3 + 1/2 X -

(3) TMABH I3 3/2X2 -

(4) TMABIH + 5/2X2 -

(5) TMA + BX
3
(13) pyBHi, HX -+
0 H

(7" PyB.r11 + 1/2X2 -

,(8) pyIi. + 2HX

(9) pyB1.3 X1 2 -

(10) py.BI3 + 2 X2
2J*


T?'IABtl2X

TMABIIX


TMABH-X2

TMABX3

TMABX3

pyBH 2 X
2
pyBJ-I X

pyBHX2

pyBHX2

pv!3x.'


+ 112

+ 1/2112

+ 1/2H2

+ 1/2H2


t{2

1/2 H2

2H9.


12

'V


(X = Cl. Br)*

(X I)

+ HX (X=Br, I)

2HX (X = Cl, Br, I)**

iX F, Cl, Br, 1)

(X = C1. Br)

(X = I)

(X = Cl, Br)t

(X = )

I IHX (X = C '. Dr) M


* This reaction proceeds no farther, even with excess HX, unl,.ss the
solution is heated to about 100
Reflux i: required for X=I. Reaction prodlices a miixturc d) product..- forX=Cl.
** Rractlo stoichiornelry is assumed for X=CI. -Relflu is required for X"1.
l The reaction proceeds no farther, even with excess HX and heating.
ft Reaction scthiomctrv is assIumd for X C'.

22










In some preliminary experiments, it was established that triethylamine-

borane reacted in a fashion similar to trimethylamine-borane, and that dimethyl--

and diethylamine-borane reacted in a fashion similar to pyridine-borane. If, as
5, 52
has been suggested in some of the work previously cited,25 these reactions

are affected by steric requirements in the reacting molecule, it is reasonable

that tertiary amine-boranes should all react similarly and that secondary amine-

boranes should react more easily, hence the resemblance to pyridine-borane

which also presents less steric bulk.

Preparation of Triethvlamine-Haloboranes

Reaction of triethvlamine-borane and chlorine. -- The reaction of

triethylamine-borane and chlorine was examined by monitoring the change in the

nmr spectrum of a solution of (CH3CH ) NBH3 in CH9Cl as a function of the

amount of C12 added. The original pattern for (CH 3Ci 2)3'BH3 ( a triplet in the

methyl region and a quartet in the methylene region) was observed to diminish

in intensity as a new triplet and quartet grew downfield from the original triplet

and quartet. This new set of peaks grew to a maximum then began to diminish

as a third set of peaks (again a triplet and a quartet) grew donnfield from the

second set. As this third set of peaks grew to a maximum and began to

diminish, a fourth set of peaks (a triplet and a complex multiple) emerged

do'wnfieid from the third set. This fourth set grew to a maximum as the other

three sets disappeared, and did not diminish on further addition of Cl,. These

obervation i re explained on the basis of stepwise chlorination of

CHt G, I. NBH3, producing successively (CH. C )3NBH2C1, (CH3CH)3 NBHCi2,

ond finally (CH3CIi2)3NBCl The reaction was such as to produce a ternary
3 AO 3









system at most points in the reaction; that is. before all the starting material

had been monochlorinated, some of the product monochloroborane adduct had

been dichlorinated to give dichloroborane adduct. The nmr spectrum observed

for (CH3CH2) NBC]3 is the same as that reported by Miller and Onyszchuk,36

who prepared (CH3CH ) NBCi by direct combination of (CH3CH ) N and BCI3.
3 2 3 3 323 3
The spectrum reported here gave better resolution of lines so that the tentative

assignment of MTiller and Onyszchuk is confirmed. The complex multiple in the

methylene region of the spectrum of (CH3CH ) NBC1 was produced by coupling

of the CH2 protons to boron as well as to the CII protons. The pattern (Figure

1) actually appeared as a 13-peak multiple because of overlap of some of the

peaks in the expected 16-peak pattern.

The chemical shifts and coupling constants for these adducts are given in

Table Ill.

Reaction of triethylamine-borane and bromine. -- The reaction of

triethylamine-borane and bromine was also examined by monitoring the change

in the nmr spectrum of a solution of (CH3CH2)3NBH3 in CH Cl2 as a function of

the amount of Br2 added. As in the reaction of (CH3CH2)3NBH.3 Ind Cl2, the

peaks due to starting material were replaced by two sets of peaks (each set

consisting cf a triplet and a quartet) downfield from the original set, which

successivcly increased and then decreased in intensity as more bromine \vas

added. These two sets of triplets and quartets were due to (CH CH 3 NEIIlBr

and (CHI3CH I NBHBr respeclively. A triplet and a complex multiple appeared
3 2'S 2
dovwfield of the triplet and quartet of (CH CH) 3NBHBr2 as more Br was added,

and this set of penks grew to a maxiaumin and did nol diminish on further addition





















ET3NBCI3
3 3


20 Hz
c--- --- -------!


Figure 1. Methylene region of the II rmr
spectrum of (CHTI3CH) NNBCi .










of Br2. This fourth set of peaks was due to (CH3CH,) NBBr3. The mu!tiplet
of r 3 33 Th u

in the CH2 region (Figure 2) showed 15 peaks of the expected 16--peak pattern,
13
one peak being masked by a C satellite of the solvent, CH C1,.

The chemical shifts and coupling constants for these adducts are given

in Table III.

The 1B nmr spectrum of (CII3CH2)3NBHBr2 showed a doublet (JBH

11 Hz) centered at 20.3 ppm upfield from trimethylborate.

Reaction of triethylamine-borane and iodine. -- The reaction of triethylamine-

borane and iodine was likewise examined by monitoring the change in the nmr

spectrum of a solution of (CH3CH2)3NBH3 in CI2C12 as a function of the
au2 of3 I 2dded. Triethvantna-b
amount of I added. Triethylamine-borane (2.22 g. 19.3 mmlol) was dissolved

in 25 ml of CH 2C1 2. Solid iodine (2.44 g, 9.62 mmol) was added in portions

with stirring. The nmr spectrum of this solution showed a triplet in the methyl

region and a quartet in the methylene region, each downfield from the triplet and

quartet of the starting material. This pattern is due, therefore, to (CH3C i2)3N-


BH2.

To the solution of (CII3CH2 ) 3NBH in CH2Cl2 prepared above, more

iodine (3. 21 g, 12. 0 mmol) was added with stirring and reflux. The rmr

spectrum of this solution, taken several times over a period of about 20 hours,

showed the trip',at and quartet due to (CH3CH I) NBII 1 slowly disappearing,
3 23 2

while a new triplet and quartet appeare-. downfield. After 22 hours of reflux,

the ne. triplet and quartet were the only peaks in the spectrum other than

solvenlt peaks. These peaks thus correspond to (CH Cii ) NBII2.
3 23 2
























ET3NBBr
3 LJ3


20 Hz 1
,--------------------- ^'" \t








Figure 2. I.l~1;lene region of the K inmr
spectrum of (CH C(H ) NBBr3
S3 21-1 3









11
The B umr spectrum of (CH3CIH2)3NBHI showed a doublet (J =

155 Hz) centered at 412. ppm upfield from trimcthylborate.

To the solution of (CH3CH2)3NBHI2 in CHCl1 prepared above, more

icdine (4. 89 g, 19. 3 mmol) was added with stirring and reflux. The nmr

spectrum of this solution still showed only the triplet and quartet due to

(CH3CH2)3NBHI2 after a short (1/2 hour) period of reflux, but the spectrum

taken after about 4 hours of reflux showed a new set of peaks, the peaks for

(CH3CH2)3NBHI2 having almost disappeared. This new set of peaks was a

triplet in the methyl region and a doublet of quartets in the methylene region.

On refiu;ing for an additional 24 hours, the new pattern remained unchanged.

This pattern is not consistent with (CII3CH2 )NB3 but is consistent with
+
3 2 3
(CI!3CH2 3NH .

H--
Accordingly, a sample of (CH CH2 ) NI I was prepared and its mnr

spectrum in CII2 C2 obtained. To this solution, some iodine was added, so as
+ -
to give (Ci3CII2)3NII 13 and the nmr spectrum again obtained. Both spectra

were very similar in pattern and position to the spectrum obtained above,

namely, a doublet of quartets in the methylene region and a triplet in the methyl

region. As a final cheek, the two solutions [(CH3CH2)3NH I3 and the unknown

soluNrio above] were mixed and the nmr spectrum taken. The spectrum showed

c;ily o:e set of peaks, the pattern being unchanged from the spectra of the

solution. (iken separately. Therefore, the p-ioduct of the exhaustive iodination

+ -
of (ClCi-21 )NBIH is not (CHi3,CI2 3N13 but is instead (CII CH ) NH A,
2" 2 2 23 3 3 2 3

where A 1- (or scme other anion).

The olhem;ical shilts and coupling constants for these adducts are given in

Tn'ale IIIn.











Reaction of triethylanline and boron trifluoride-etherate. -- A solution of

triethylamine in ether was added dropwise to an ether solution of boron

trifluoride-etherate. Evaporation of the volatiles left low-melting, colorless

crystals of triethylamine-trifluoroborane. On standing in air these crystals

slowly decomposed to a brown oil. An nmr spectrum of this material in CH Cl

solution showed a triplet in the methyl region and a broadened quartet in the

methylene region. The chemical shifts and coupling constants are given in

Table II.

Preparation of Dimethvlamine-Haloboranes

Synthesis of dimethvlamine-nmono- and dichloroborane. -- (CH ) NHBH2 C1
53
and (CII ) NHBHCI2 were prepared by V.R, Miller" bh the reaction of HCl and

(CII )2NHB, These procedures were confirmed in this present work, and the
32 3

mnir data for these two compounds, quoted from Miller's work,are given in

Table II.

Synthesis of dimethylamine-trichloroborane.54-- (CHI)NHBH3 (1.90 g,

22.2 nrol) ',as dissolved in 25 ml of CH C1 Gaseous CI2 was bubbled into
2 2

:his eciiion at a moderate rate, with stirring, until the solution turned yellow.

fl. trEis tu.ne a white solid was present, and the flask was warm from the reaction.

Al.e !.:-" vc-ri:es woere stripped off by pumping on a rotary e- apcportor, lea'drg a

. it. yelio. solid- The prtodrc- was stirred with 2 n ml of fresh C21 C2 and Lhen

20 *i petrolaum ether was added to precipitate 'he portion of the product that

wns in so:;.ti.on The white solid thus obtained was filtered, washed with two

5 mI portions of petroleum ether, and dried by purmping. Tlhe yield was 4.56 g

(28 1 rinmo1'- 7. (i0, based cn (;i 'H) NHPHT ), mp = 125-7 Anal. Called. for

Ci NN C!, 14t -T, 4.25; N, .. . Found: .. J 1 P';: ii, 4. 52: N, 8. 27.
7i









The infr-iard spectrum of (CH3)2NIIBC13 (Figure 3) is characterized by
-I
a sharp band nl 3130 cm assigned to N-H stretch, and a broad band at
-1
700-780 cm assig-ed to coordinated BCI, vibrations. The sharp band at

000 cm vwhic Gerrard6 assigned to BC2 is probably due to C-N stretch.55

The entire inft rced spectrum is given in Table I.

The proton nmr spectrum (Figure 4) showed a 1:1::2:2:1: sextet in the

aliphatic i.gion. This pattern arises from coupling of the methyl protons with

bo:h the nitrogen-attached proton and boron, giving a doublet of quartets: the

coupling constants are such as to cause overlapping of two sets of tv.'o peaks.

The chemical -hift and coupling constants are given in Table 1I.

The :- nri'i; spectrum showed a single peak at 9.4 'pnm u)ffield from

trimi' thylimjt ,: .

Reaction of cdimeihvlamine-borane and bromine. --- Dimcthylamine-boran,

(0. 51 g, 11.0 n3irol) wvas dissolved in 30 ml of CH Cl Bromine (0.885 g,

. 54 mnmoil)dissolved in 5 nil of CH2C!i, was added dropwins ,xith stirring.

The namr -:r.. t"run of this, solution consisted of a stiorg do:','et do'wnfieid from

:,: dJoub,lt wi .-: to trst.rtng' material. There was a small inount of itan rating

n:aoria! c,-o S:Ie :.Le dcoublet due tuo starting material had not .ire-ly disaro?'alJ.

Therei..r \-is3 '.:so, ev)rence for a third product since a sFmall 'oe : wa.s ortse-rved

d 'ofiel'd ;-l '. t!i t';ro ,; d, biet. iT'- observati. s, ai. e .-itn' ',;ih 1 no

cf I'; (CU N)3 ieH h h ,,i ',een crnvi-, ted e o t (C l .) NH . t-:, ins o little

(CI rl U H

T'. )r U',-,."C.*,!: ol>-e i' ](";')2NM ',3Br in C.17H ;.C '-. re- .. o e b :i'' .:

(0. 4 g!4, '7.2 '"m ol), dc-isolved in 3 ml (.f CHI CL- I.d ..,,,xlv _tit.
,"-2












*7.t! I
F-
I1,~-h I' ~~Ii_

:,i I~~..-i-i-,..2l~ I..
Iii



Kr .2 i- iii-jj7


L.-. -_I- Z4--t-i--..4-- __ I
"0 0'6,)--- 0- 0 200206


WAVEl.rt GH IN W.C tCNOC
35 6 65 7 75 9 10 11


I -

---






1 F00
O0 1800


Ji j i



I -
T'2







-t


1600I 14
1000 I1


II 7




La- -K

17 2P


00 1230 10O0
WA0VENUMlit C6M


12 14 16 6 1 l 20 ti 30




S'I !
o





'ii

.i..... .I i
.O...0 0 1-.00 IO 16 ,o
--- LtJ 10 1`
30





0oo 600 600 503 411 500


Fl',;rCu 5.Infrared spectrmuil of (CB,)2NN1Br.3
.1 2 3 *I


" :--i-- '----c------_ .
























ME NHBCLI
2- ~


10 tHz


I
FgUre 4. H fln1r spectrum of ICi-! ) N1;,BCI
g~3 2









'the nmr spectrum of this solution showed the doublet assigned to (CH-3)2NHBH2Br

to have nearly disappeared and the double assigned to (CH 3)2NHBHBr2 to have

become very strong. There was a shoulder on the downfield side of the

(CH3) N2 BIIBr2 double which vas assigned to a small amount of (CH 3)NHBBr3.

To the solution of (CH3)2NHBHBr2 in CH2Cl2 prepared above, more

biomine (0.986 g, G. 17mmol),dissolved in CII2C12, was added slowly with

stirring. The nmr spectrum of this solution showed the doublet previously

assigned to (CH3)2NHBHBr2 reduced somewhat in intensity, along with a

complex pattern of similar intensity at the position of the shoulder previously

assigned to (CI)2 NHB3Br3. More bromine was added to this solution until a

permanent coior was produced. The nmr spectrum of tis solution showed

only the conmple pattern assigned to (CH3)2NHBB 3. Chemical shifts and

coupling constants for these adducts are given in Table II.
54
SNnthesis of dimethvlamine-tribromoborane. -- (CI3) 2NHBH3 (2.95 g,
It). inn, Br( 05 g, 100. 4 ll

tO.0 mrmol) was dissolved in 40 ml of dry CH2C12. Br2 (16.05 g, 100.4 mmol

was dissolved in 40 ml of dry CH2C 2. The Br2 solution was added dropvise,

...ih v;gc'riu;s stirring. to the (CH3)2NBH3 solution. As the :ast of the Br2

soluion wac added, a wluhe precipitate formed, and the solution, which was

previously ciloriess. turned slightly yellow. All the volatiles were stripped off

b :nTirn'. on 2 oar' ,-apo rator. TLhe whitee sold which remained ,was laker,

:.ii ir tihe i':k. n1 a oairbox whcre [he product was transferred to 9 tared

bootie. *'e' y'. as 13. 0 4 (44.0 mmol: 88. 0%. ba'ed el (CHI3)2NHBE3),

= _50.35-_.. ._. Ca!cd. for C2H 7 NBBr: C, S.13: Si, 2.39; 4.74,

','., !... r, fo.d: C, 8. 1'4: H. 2. 3,; N, 4. 64: Br: 81.21.









The infra red spectrum of (CII3)2NHBBr3 (Figure 5) showed a strong,
--1
sharp band at 3150cm ,m assigned to N-H stretch, and a broad band at 650-'
-1 55
700 cm assigned to coordinated BBr vibrations.5 The entire infrared

spectrum is given in Table I.

The proton nmr spectrum (Figure 6) showed a multiplet in the aliphatic

region, which resembles a sextet of 1:1:2:2:1:1 intensity, but which shows eight

peaks under high resolution. The pattern is a doublet of quartets due to coupling

of the methyl protons with both the nitrogen-attached proton and born. The

chemical shift and coupling constants are given in Table II.

The 11B nmr spectrum showed a single peak at 24.8 ppm upfield from

trimethylborate.

Reaction of dimethylamine-borane and iodine. -- To dimethylamino-borane

(0. 534 g, 9.06 minol),dissolved in 20 ml of CI Ci2 2was added solid iodine (1. 1 g,

4.49 mmol) piecewise with stirring. The nmr spectrum of the resulting solution

showed only a double downfield from the position of the doublet due to starting

material. This double was thus assigned to (CII3)NHBH2I.
11
The B 3 nmr spectrum of (CH3)2NHBlIi showed a triplet (JBl = 143 Hz)

oeitered at 311.0 ppm upfield from trimethylborate.

To dimethyih!'aine-bor:nc (1.05 g, 17. rnmeol) in 25 ml Cli2Cl2 was added

icrine (4. 52 g, 1"7. a ol), piecewise with sti'ring. The nnir spectrum of the

i'.;ulting s'lutio;, Taken 45 mir.ites after all the 12 had been added, showed a

strong dou'blet downfield from t.iie pnsitioo of the double prcrv usly assignerd t

(Ct) N[H I2. T'i'hi' doblet was assignc-d to (C13)2,'iBII There wasi in the

sp;ac;.u a oum;i]a ealk corresponding to tie right-hand peak of the doublet due to
























ME2NHBBr3


10 Hz
t--------1--


1
Figure 6. iH nmr spectrum of (CHI)2jT fBr3
S)2 3









(CH3)2NHBH21. The left-hand peak would be masked by the double assigned

to (CH,,)rN1INBHIn The spectrum taken after 5 hours sholwved only the

double assigned to (CH 3)2NHB-HI2

The choernical shifts and coupling constants for these adducts are gven in

Table D.

The lB nmr spectrum of (CI3,) I2HBII2 showed a doublet (JBH = j58 Hz)

centered at 43. 5 ppm upfield from trimethylborate.

Synthesis of dimethvlamiie-triicdoborane. -- (CII )2NHBH (2. 5 g, 42 mmol)

asis dissolved in 40 ml of dry CIl2Ci2 and 12 was added piece by piece with

stirring untti a permanent color was given to the solution (about 21 mmol of

io was required). A large excess of 1I ,n.as then added (at least 10'i emmol), and

the solution -.was refluxed for about 40 hours. After refluxrng, io volic :ies were

rerova.l on a rotary evaporator. A black. fuming soliedr -nained in the flask.

Fresh :C'I, C! (40 ml) was added to this solid. To this solution about 3 mi of

merieury was adiedt,and the mixture wvs vigorously stirred for 1.5 minutes. At

t1is- irre the s'oution -as doeo--'. cnd solid HgT, was obscu \ed in the flask.

'i he co'u;er.n- o" -he flr.sl '-.'erLe fi '4reCd.* i ,Aci e was sidded uc he- filtrate,

5 :c .' ,o-.f- iu,, ;-. '.. :, --':p fl. "..-, s- I. T e w i. r .r,- oV. C -st .h "i.V'h tv,0








The infrai ed spectrum of (CH3)2NHIBl1 (Figure 7) is characterized by a

broad band at 3300-2950 cm- (with peaks at 3210 and 3110 cm1) due to N-H

stretch, and a double at 615 and 600 cm probably due to B-I stretch.

The entire infrared spectrum is given in Table I.

The proton nmr spectrum (Figure 8) showed a multiple in the aliphatic

region, which resembles a quintet of 1:2:2:2:1 intensity, but which shows eight

peaks under high resolution. The pattern is a double of quartets due to coupling

of the methyl protons with both the nitrogen-attached proton and horon. The

chemical shift and coupling constants are given in Table iI.

Synthesis of dimethvrimine-trifluoroborane. -- [This is the only dimethyl-

amine-trihaloboriane which can be prepared pure by direct addition of amine and

bornTe. Two different methods exist for the preparation of dimethylamine-

rrifluoroborane. The adduct may be prepared by direct addition of dimethylanlnime

to boron trifluoride5 (in the gas phase or in an inert solvent), or the adduct

may be prepared via a displacement reaction, typically the reaction of dimethyl-

ami;ne with boron trifluorido-etherate. ]

Both ;vethodn-; ,were 'ased in this study, ann the best results were obtained

v.-Iii- ti-e i.o gases were bubbled together in on inert solvJt (CI2 Cl2). Care

should h: takcn to kcp r F 3 in slight excess. The product of this reaction,

isolated by pl m.p.n off [he volatiles or, a ro, ry evaprorator. wq'a a white

cry.staliac solid, ip = -40-44 Anal. Calcd, for C H N1F : C, 21.28; H, 6.25;
2- 3
N, 12., i. Fo.ud: C, 20. 0; 5, 5.7 0; N, 12. 16.

The inrlarred spectrum of (CHI) NHBF (FJgur. 91 was broad and poorly

resolvCed, probably :icciuse of the low ILiting point ui the crystal,. It is

















ME2NHBI3


/ /






Figure 8. Ii rmr spect.uriu of (CH ) NIBIf,
3 2









-1
characterized by a band at 3260 cm assigned to N-H stretch, and a set of

-1
three broad, ill-defined bands at 1210-1150, 1120-1040, and 980-900 cm ,

probably due respectively to B-F stretch. CII3 wag, and a combination of B-F
55 -1
stretch and C-N stretch.55 The sharp peak at 705 crm is probably due to B-N
40
stretch.40 The entire infrared spectrum is given in Table I.

The proton nmr spectrum (Figure 10) showed a 1:1:1:2:1:1:1 septet in the

aliphatic region. This pattern arises from coupling of the methyl protons to

40
both the nitrogen-attached proton and boron;0 the coupling constants are such

as to cause the overlapping of two of the expected 8 peaks. The chemical shift

and coupling constants are given in Table II.

Reaction of Dimethylamine and Boron Trihalides

Reaction of dinethvlamine and boron trichloride. --- Boron trichloride

(1.50 ml, 13.3 mmol) .vas dissolved in 50 ml of dry oenzene on a vacuum line (the

pressure of the solution was 120 mm). Dimethylamine (1.20 ml, 18.3 mmol) was

trapped out rn another part of the line and allowed to warm up. Whn:t: trhe pressure

ef dimeWhy)amine reached 500 mlm, the stopcock to the BC13-benzene solution

wa.:; ol'peIedr, a'lo,.-ing hoe dimrlthlylamine to enter at such a rate as to hold the

r01 sst'.r1 of din:tthv'h-:nine in the external system steady ati 500C mm; a x.hite

solid -r. t..T d" rri s l this procedure. The addition was facilitated bi sti'cinf.

Aftcr a'! the Ji-:nelt : lniim c ]hadl b en f tr-,,j sfirrel into iie ECl -benzcne t iUluion,

the vo;atiies w' y.. -'.moved by evacuation. iJavir.ng vite solid in the flask.

After tr ts far to a laredJ viai. the prodJcte vi.ih.- 2. 553 g (SG yieldd.

A small c.it 'unt aboutt 0. 2 g) of this prodl;. ct ":.'s s;-irr.d ;ih 20 m) of

CTHI C for fl' .'r-e Jh-urs. 'FThe so!'iti n '.c. fi '"v: :-; 3a '-bite sel (A). Th
.2, 2

























ME2NHBF3


10 Hz.
i---------------- -I-


Figure 10. H nmr spectrum of (CIiI NBFP ,.
312 "










solvent was removed from the filtrate on a rotary evaporator,Icaving a second

white solid (B). Each fraction consisted of about 0. 1 g of material.

An nmr spectrum was taken in CH2Cl2 solution of the CH2C1-soluble

material (B), which showed a seven-peak multiple in the aliphatic region.

This mulliplet was the result of an overlapping of two resonances: 1) a 1:1:2:2:1:1

sextet, centered at 172 Hz, due to the 1:1 adduct, (CH )2NHBC!3, and

2) a 1:2:1 triplet, centered at 167. 5 Hz, due to dimethylammonium ion,
+
(C3) NH2 NCH = 5. 7 Hz). The relative proportions of these two species

was determined (by estimating visually) to be about 5:1, the adduct in excess.

The infrared spectrum of this fraction (B) was similar to that of authentic

(CH)2 NHBCI 3 but there were some small differences. A small peak was
-1 59
observed at 650 cm wich was probably due to BC14 ion. A broad band
4
-1 -1
at 3050-2920 cm two sharp peaks at 2680 and 2600 cm ,and two sharp
-1
peaks at 2690 and 2G00 cm- were probably caused by [(CH3)2NH]2BC12 The

absence cf strong! peaks at 2440 and 1025 cm ^ indicates the absence of

:C} ,'1,',, J i': >"'In (/B). The ammonium ion observed in the nmr spectrum

of f-ac.Uion (B) : us3t then have come from the hydrolysis of [(CH)2 Nl 2BC1

3ince cniihdroivizr-: [(CH3l) 2NH] BC1+ ion would not ha\ been distinguishable

in the nrr (5 = 1'7 Hz In C)H Cl about 15 Hz Mond), one cannot judge the

rej! Pt'e r o!;Cor!ions of addilici and b?roniurn salt frno' ti c ratio of a"ck.dct to

,a m,.oniu, io: a a dIetern-'in.-,i by nmr integration. I'ov er, the rjo.atite

i.:t: i:ie *i: '.c ro' eak,-' in Ih' in ilared sp-octruri imply a ratio on the order of

3:i. [i T'ht i., '!he tI'aclion is 60', aiduct and 10/, boronitrm sa1 ., r, weight.]










The infrared spectrum of the CH2C12 insoluble material (A) (Figure 11)

differed from that of authentic (CH)2NHECi3 in several ways. There was a

-1 59
peal: of mcdir:m intensity at 650 nc probably due to BC1 ion, and

-1
another peak of medium intensity at 540 cm probably due to amine,BC1
-1
The pattern of peaks between 700-850 cm was markedly different from the

pattern in the spectrum of (CI3 )2NHBC13, probably reflecting both the absence
4-
of coordinated BC1 groups and the presence of amine2 EBC groups. There
3 *2 2 g
-!
was a marked difference in the 2500-3300 cm region also. The spectrum

-1 -1
showed two strong,broad bands centered at 3200 cm and 2960 enc and

two ship bands at 2780 cm- and 2680 e-m These bands are probably due
+
to the amine vibrations (N-H and C-H stretch) in aminc2BC12 The rimn

spectrum of this fraction(A) in CD CN showed a broad. fcature.l-s- n,~'at.,n
S3

ir. the aliphatic region (157 Hiz). [Exposure to atmosphere during Sample

preparation resulted in the appearance of a 1.2:1 triplet in th'j spectrum
-I-
(6 160 Hz, J = 5. 5 Hz). It would be expected that [(Ct3)2,! i2BCl2 ion should

be cqioc :'.s3!ivec to hydrulysis. j

As n-e-ui-tive ,,roo.f oCf thesee assignments, an authentic sample of

(CH.21;i '-K 2C1 vas prepared and conm pae wli thi if t'ron, (A).

1 '.T:r!i! s: f h i-ij2im t O\- i', :iOe)-- 'o 'ao:niu chl.hrido. -- [fC:}i ..I ],] E1!, Cl

'. a!- ,rea- ;d*. lowl;: s;i.h i rl' in ta t Tcscrihed y Miller and ,;ys h-;, -. '

(."* the- :; at ^, I'oiC;:41 I C C/ .,iSf 1. 3 20, (t1 M o,
*- .7 N, 1 H7 a I,!. O o li C
6 -


:^ .:,:no." -,w-;A a-Jd ed slo ly vith ozrrring. Th'e vo'alti'cs %ere,?: r .,










pumping, and 30 ml of fresh benzene was added to the white solid. A 1. 71-N

solution of (CIHi3)NH in C6I6 (12. 0 ml, 20. 5 mmol) was added with stirring. The

volatiles were removed by pumping, and the white solid which remained was
+-
transferred to E tared vial. Yield oC [(CH 3)2 NH 2BI CI was 2.62 g (94.7%,

based on (CH3) N1HH3H), ip = 161-3 The infrared (Figure 12) and nmr spectra

were nearly identical to those observed by Miller and Rys dkewitsch60 for
--
[(CH ),NH]2Bt2 I Anal. Caled. for C HI1 N2BC1: C, 34. 70; H, 11.65;
2 2 2 2 -- 16 2

N, 20.23. Found: C, 33.50;H, 11.70; N, 19.73.

Synthesis of lis-(di rethyvamine)-dichloroborol:ium chloride. -- [(CHI ) JNIH -
.9-2 -2
+-
BII, Cl (0.47 g, 3.40 mrol) was dissolved in 30 ml of CH C12 and C1 was

bubbled in nq a moderate rate foi 1-1/2 hours. During the addition of C12, a

v.h'te solid precipitated out and the solution turned y eiiow. The volatiles were

remov.i by pumping, and the white solid which remained was transferred (in

the dry box) to a tared flask. The yield of [(CH,)2 NH]2BCil Cl was 0.70 g
+ -
(9'8;, based on [(CH3)2NH]2B12 Cl ), mp = 123. 5-1-2- nal. Calcd. fr

C HN C13C: C, 23.17; H, 6.81; N, .51. Founi. C, 22..72,, 7.;5;
4 fl 2? 3
+-
N. 3. O. Th!e :c-v spect-rum c; [(CE-i 3)Ij2" T 'B:2 CI in Cc S CN was a broad,

-u'ls a.res ronce cantered at 155 Hz. The infra red spectrum (Figure 13)

- vfs very si? x rir to the spe.r1 um ro fr rr.cio (A) in the region 900- In00 cm ,
-.1
!:,.t -,.', r a-'e *r. -a! i il fe ence;s ir- the region below 9tUU cm notably the

.il.s, :.,_'r? ,", th'- 'i ?i. 650 iO' ar.d a -ove'er ;.ter'osit of the pattern t 750-850 r .cm

u c''Jf !i!-i--- hd'2'Tft).. )-diF-:Moro'Doroniui chrcide an..(
'.h.e rt 1-; ) '.n C 0.20 1.0 m ol ) :wa s taken i flask,

attached re a nar.-ui!n ri- a-nd eva:iated. BC13 (2.6 Crc) wa introdluc-.0 irnt:











WAVELENGTH IN MI
3 5 425 5 1 55 6 o5 / I/ B




S I I
S-- ---, '--^-'- -/ --7 k















F-T-+
F re 13 nfrar spectrum of [(C ',NI BCl ,C .,
2- 2 2-
i I
















Fi', e. 13. Infrared spectrum o t [>C)!0, NI11BCI12 Ci


9 10 11 I0 14 16 16 1i s ;5 3C0


1200
R CMm









the flask very slowly so that no heat buildup occurred. After the solid was

exposed to all of the BCl a large -excess (about 20 mmol) of BC3 was

introduced into the flask, and after one hour's exposure, the volatiles were

pumped off. The white solid which remained was transferred to a tared vial.

The weight of the solid was 0.275 g, corresponding to 75% conversion of

Cl to BCi4 (Since there was some material loss during transfer, the actual

percentage may be as high as 80%.) Ti cmpwas 93-98.5.The infrared spectrum

of this solid (Figure 14) was identical to that of fraction (A). The new peaks

in the spectrum, which are thus due to BCI are at 650(w), 735(m), and

780(s) cm*.1

In summary, the direct reaction of dimethylamine and boron trichloride

produces two major products as shown in the following equation:



6 (CH3)2NHBClI

20 (CH3)2NH + 20 BC13- +
"0 (i 3

7 [(CH3 )22NHBCl 2 BCi, (or Cl )


Reaction of dimethvlamine and boron tribromide. -- Boron tribromide

(3. !0 ml, .3. 0 rmmol) was dissolved in 50 ml of dry berzene on the vncuum line

(the p-rssu.re of the solution was about ItO mirm). Iiiietlhylainnic (2.37 iml,

3;.;0 mmoi) waVs Lltrpped out in another part of the line and allowed to warm up.

When tl'h pressure of dimethylamine react.d 500 mm, the stopcock to !he BBr3-

beniz(eri;c ,hl:luionn was opecinld, allowing the dijietlhyinnc to (ente: Lt such a

rate asi;: '-. li the pressure of dimeth ylminP in ice exterr'ui system steady at









500 mm;n a white solid precipitated during this procedure. The addition was

facilitated by stirring. After all the dimethylamine had been added, a sample

of the reaction solution was taken for nmr. The spectrum showed a pattern in

the aliphatic region (about 125 Hz) which appeared to be the 8-peak multiplet

due to (CHI) 2NHBBr3 overlapped by a singlet possibly due to (CH3)2NBBr2

(or some other aminoberane). The volatiles were distilled out into a liquid

nitrogen trap, leaving a white solid in the flask. [An nmr spectrum of the

liquid which wvas trapped out showed only benzene absorptions, but the liquid

fumed, implying the presence of BBr3. ] After transfer to a tared vial, the

product weighed 9. 37 g (88% yield).

A small amount (1. 0 g) of this product was stirred with -10 ml of CH Cl9

for ten hours. The solution was filtered. giving a white solid (A). Th' solvent

was removed from the filtrate on a rotary evaporator, leaving a grcy 'poiadery

solid (B). Each fraction was about half of the original material.

An nmr spectrum was taken in CH2IC12 solution of the CII Ci -soluble

fraction (13), which showed a complex pattern in the aliphatic region. This

Inulitiplet was :; :ic result of the overlapping of three sets of resona,.ce: 1) an

-i-::cak noultipici, centered at 180 Hz, due to (Cil INHIBr3, 2) a 1:2:1 iriplet,
32 3

e'ord_ A t 171 Hz Vdue to (CIH) I2. :NC t = 3. Hz), and 3) singlet at

181 I~z i.-s to UC'[ .Br The .-it of inter.sities of these three resonioc'es

jdeterinni-id h-, stil"r ting visually) is !'c.- ly 2.1:0.3 ;i '(1 )21-i N B r3:

(Cii,, 4 CH .,'r The iifr-ar'cl spectrum tf tlis rrant:io' (T3i

v.Wni-.s id. 'n: t !hat .ao au tl:enic IC! ) NTiBBr3 .
'*' ^










The infrared spectrum of the CH2ClI insoluble material (A) (Figure 15)
-1
was quite different from that of (CH )2NTBBr3. In fact. above 900 cm the
+ -
spectrurn of fraction (A) is very similar to that of [(CH3)2NH 2BC12 Cl ,
+
implying that the material is a salt of [(CH )2NH1 BB2 The region below

900 em snows a series of bands at 650-820 cm- (maxima at 665, 685, 720,

-1 +
785, and 615 cm ) due to amine BBr2 vibrations. In addition, there are

-1 -1 +
two bands at 450 and 470 cm and a band at 550 cm due to BBr2 and BBr4

respectively. 1 [The two bands at 450 and 470 cm-1 are related to two similar
-1+ -
bands at 505 and 540 cm in the spectrum of [(C3 ) NH] BC12 CI .] An nmr

spectrum of this fraction (A) was taken in CH CN, which showed two

resonanCes in the aliphatic region: 1) a broad featureless resonance (about
+
12 Hz wide) at 171 Hz, due to [(CII )NH]2BBr2 and 2) a 1:2:1 triplet at
+
159. 5 Tz due to (CH)2NH2 (J = 5.6 IIz). However, the absence of bands in

the infrared spectrum characteristic of (CH )2NH- shows that the ammonium
3j 2
4--
ion comes from hydrolysis of [(CH)I2NH]2BBr2 .

In summary, the direct reaction of dimethylamine and boron tribromide

produces two major products and two minor products as shown in the following

*equation:


32 23
++-
-.CH -,) h + 4BBr + 1 [(CH~32NH]2BB r2 BBr (or Br )
+
<< [(CB1 ) NH2 BBr4 + (CH r)NBBr
2 2 2,4










Preparation of Diethvlamine-Haloboranes

Synthesis of diethvlamine-boran e. -- Diethylamine-borane was prepared

via a trarsamination reaction of diethylamine with trimethylamine-borane.

Trimethylamine-borane (21.4 g) was dissolved in a large excess of

diethylamine (60 ml) and refluxed for about three hours. The excess

diethylamine was pumped off, and the remaining diethylaminiie-borane, an oil,

was purified by shaking with water and extracting with methylene chloride,

the methylene chloride being evaporated to give a good yield (21. 6 g, 85%) of

liquid diethylamine-borane. Infrared and 1H nmr spectra showed the compound

to be free of impurities. The nmr spectrum showed a triplet in the methyl

region, a quintet in the methylene region (Figure 16),and a broad resonance

farther downfield, due to N-II. The chemical shifts and coupling constants

are given in Table IV. The infrared spectrum (Figure 17) was characterized

by a strong peak at about 3200 em due to N-H stretch, and a strong double
-1,
centered at 2350 cm due to coordinated BH3 stretch.

The B1 nnr spectrum of (CI3CH 2)NHBH3 showed a quartet (J =

7. 5 Hz) centered at 34. C ppm upfield frocn trimethylborate.

,Sthesis u di.thvian~i ie-monochloroborane. -- A sample of diethy lamrine-

borse (about i. 0 g) was dissolved. in 20 iml of CH C1, and a solution -f IIC
2 ,2 a
in! CH; Ci2 ias .dd-., Che nr.r spectrum was monitored as the .c. i so uticn

,,as :dd-i., and .. ir triplet ir the methyl region due to starting material w'.as

cbsey-ed td ,'imiinin in ;ili -is Ity as a ne'v triplet just downfield i n:I'eased in

intensi:y. Ce,:-' ct-tly, a complex multiplet grew downfield from the 5-peak

Iulipict of stalLing mIaterial in t') methylene region, When the tripi -f due to










ET2NHBH3


15 Ilz
I--------


Figure 16. Methylene region of the 1H nmr spectrum of
(CH 3CH ) NIIBi3
2 22 3










ii ... I ... 'i--- _k i h



I r I; i j -


I i


S. i i' i0 I
. -- .... .-





"-'*jy J -C0

.1 ___
25. i60


0


LWAllFNGTH IN .lClON5
d6 a 71 8 10 1i i6 16 I ; 25 C 4.
---- I-. _I__: JI I
S' II
Vi -I i










:1---r
.... ., ..?i, .L -. ., .-.1[ --_ ,


18b0 i. 1U200 1200
WAVEmjmNUMi CM'


Fig ) i7. Infrared sofctrurn of (CH3,i 2NIHBI3.1
3 2'2 3


1000 600 600 600 L0 A 0 ;;'




59


ET2NHBH C)





obs.


A//V
Ij i~ ( j A


SII III iriy/
1 i I, ; \ t
i i i ij
j -'\ I o I' n'


Fig;;:-,- 2. -nIeth-ylce region of the I;U ninr spectrum of
(cIi CH M:~iBHiCi.


_calc.


21Q HZ
I










remained. The volatiles were removed from the solution by pumping and

the oil which remained was taken into the dry box for storage.

A sample of this oil was dissolved in CI C 2, and a fresh nmr was taken.

This spectrum showed a triplet in the methyl region, a complex multiple in the

methylene region (Figure 20), and a broad absorption farther downfield, due

to N-H resonance. The chemical shifts and coupling constants are given in

Table IV. The infrared spectrum of this oil, taken between KBr plates, was

consistent with the formulation (CH3CI I) NHBHC12 (Figure 21). There was
-1
a strong peak at 3200 cm due to N-H stretch, and a single strong peak at

-1
2500 cm due to B-H stretch. The entire infrared spectrum is given in

'able i.

Synthesis of diethylamine-trichloroborane. -- A sample of diethylanine-

borare (about 0. 5 g) was dissolved in 25 ml of CH 2C2 and chlorine gas bubbled

in until a permanent yellow color was imparted to the solution. The voiatiles

were removed by a stream of nitrogen, leaving a white solid, mp = 129-130 .

Anal. Calcd. for C 11 NBC13: C, 225.25;H, 5.83: N, 7.36. Found: C, 25.46;
S411 3

K, 6, 3- N, 7.22.

T';I- imr -!ect!rum of (CH3Cli2)2NNBCl3 in CH2Cl2 consisted of a triplet

in .-'de nat.:i:, r:'gi and a complex multiple in .hec mcthylene region (Figuce 22).

The ..: o-..a a! i;ts and . '; constarnts are. given in Table IV. Thi infrarrej

s ..' ,' ,,': ,jNBCi. 2Figure 23) was c-harnn-terized by a strong !c.C.

a .' C3O :;n ,!, ie t' N-HI steci-ci, and b-- Oih-. :-.b-Sence f B-r' stqtetching

-.i >r<:!rii'r: .it r;, c ,io. 3P09-0 5n0 c'0 i'e. T entire inl arrr _i sp ctrirul ? iS

-i0,-h &~ T












ET2NHBHCI2





obs.


165 Hz


ca c.


Figu;-e 20. i. ..- region of the; H nmn- spectrum of (CH3CF)2NHibItlCl,
3 2 '2,


225 Hz
I









WAVLNCGTH IN 6,'CI ONS
: 5 5 6 6 5 7 7 5


...... ..-. .-..-'-nL...- -- ...-L. Li

I- -- .i.. --

it- -r"' i 'yi r
I K' K'H
] I , i ,
I:
I 7 i
I !1 I
j I l
,-. --r--I i ,
_- -H . . +-

S. ...... .... -- .. I -



. ... .-- ---- - :
.. .... .. -- --- --- ---- C --'----k ....; ... ---i-- t---
!- :.ll 1 '.: i, .0-) = ? :


_ I" I i F I '"" '-I i; '1 I'
|i i F i i i 7 1 F I


I I


KA--B "i TI I




S A i-'- -' --T i- -- -- --
__ ._ ._ ...... - - -- i _ .

-"-YF 7.. 1o -I .-. --

0 00 160 0 1 1200 IOO BOO t>00 .;GO 500 50.35C
\-Vd,'sV6uwEf. CiI


Fi're 14. !nfrar'ed spccr'um of [(CH )2 NHi BCI2 BC! .
.'i2 2 2


6 16 18 70 75 30 C












iI -' i :L i -.,< i
L i '





I I i
. -- -- -
9 I/


,--.1,1 U" i /
I ii


.. : -- ---- ---I . [1


i '' 'l l--. LL. i. |.


V'AVE;[NGH IN MICRONS
i5 7 6 1 7 5 9 I 1i 1 12 I 1 i6
I ,r


i "' '' I rT h i I


.. i I! I ".




..- ._ .._ .... .. -- -- .. ..-I- -1
. . . . :---- - -.. .. ." - - . .


235. 32i0 213 2000 'CO I-l6 lto 10 1200
WAVENUMBER CM'


16 18 T7 25 33 .10
i l' i _ i










WI A
-I




-I--



r-L ..... r ,.-- --.
i- L. --i--
,--i -i .. 7--


Figure 15. Infrared spectrum of fraction (A) from the reaction of (CH3)2NH and BBr3.
3 2^ 3"


10o0 aCO C 600 503 .00 1:.O










Reaction of diethylamine-borane and hydrogen bromide. -- Diethvlarmine-

borane (0. 577 g, 6.64 mmol) was dissolved in CH2C12, and a 0.33 M solution

of 'Br in CH2C12 (20. 0 1l, 6. 60 mmol) was added with stirring. The mrnr

spectrtum of the solution consisted of a triplet in the methyl region, just

downfield from the position of the triplet of the starting material, a complex

multiplet in the methylene region (Figure 24), and a broad absorption farther

downfield due to N-H resonance. This pattern was assigned to (CH3CH )2NHBOHBr.

Addition of excess HBr caused the peaks due to (CH3CI2 )2NHBH2Br to diminish,

being replaced by a new triplet and complex multiple downfield from those of

(CH3 CHl,, NHBIH Br. The new peaks were assigned to (CH3CH2 2NHBIIBr2.

Reaction of diethvlamine-borane and bromine. -- Diethylamine-borane

(0.432 g, 1.97 mnol) w-as dissolved in 25 ml of CH Cl Bromnine (0.791 g,

4. 95 mmnol) dissolved in 2 ml of CH2Cl2 was added to the amine-borane solution

dropwise with stirring. The nmr spectrum of the resulting solution showed

two sets of triplets in the methyl region, of relative intensity 5:1. the positions

of which i:.CTrc ond to the previously assigned (CIH CI) 2N!IBllEr2 and

iCIS,,:CH,2)2,HBHi2Bc, respectively. More bromine (0. 132 g, 0. 83 mmol)

was a'idcd ni 1 CHliCl solution and another nmi spectruir' takenV. The spectrum

ai1 ov', .: y ',"1 e 1-ipiet Clue to (CH CH(') NHITBBr2 in the methyl region. In

additionn, there .a;: a complex muitiplet in the methylene region (Fig.ure 25)

''d a broad absrpo'.pan due to N -If resonance farther dormfield. Thie chemical

shifts ulu coupling consinLs for these tw.o adducts are given in Table IV.
il
'The ;3 n .r specti am of (CIH CH 2)2 F HB:H1 r2 showed a broad (400 H

'\ide) f:atauceless rtuoiiance at z,. a [.ppm lpfield from rrlmethylborate.












ET2NHBH2Br



160Hz




\ obs.


210Hz





f







A





i
/


* / -/11


calc.


FI:gr 2-i. :. .'.,i..-r region of the H nmr spectrum of (CC! CIIN )NEBE.Br.
3 22 2













ET2NHBHBr2





170 Hz


Sobs.


calc.


1
Figure 25. ethylene region of the H nmr spectrum of (CII3CH2),'FHFBHBr2.


220Hz
I










Synthesis of diethvlamine-tribromoborane. -- Diethylamine-borane (about

0. 5 g) was dissolved in CH CI and bromine added slowly with stirring until

a permanent color was imparted te the solution. The volatiles were evaporated

by a stream of nitrogen, leaving a lght yellow solid, mp = 155-157 Anal.

Calcd. for C4H11NBBr3: C, 14.84; H, 3.43; N, 4.33. Found: C, 15.28;

H, 3.60; N, 4.27.

The nmr spectrum of (CH3CH2)2NHBBr3 in CH2C2 consisted of a triplet

in the methyl region and a complex multiple in the methylene region (Figure 26).

The chemical shifts and coupling constants are given in Table IV. The infrared

spectrum of (CCH3CH )2NHBBr3 (Figure 27) was characterized by a strong

peak at 3180 cm-1, due to N-H stretch, and by the absence of B-H stretching
-1
absorption in the region 2300-2500 ci1. The entire infrared spectrum is

given in Table I.
11
The 1B nnmr spectrum showed a single peak at 24.9 ppm upfield from

trimefthylborate.

Synt~hesis of d I ethylamine- irnc:noiodoborane. -- Diethviamine-boranc

(0.985 g, 11.3 tmmol) was dissolved in 30 ml of CH_2C2, and iodine (1.44 g,

5. u7 nmnol) was added piecewise with stirring. The volatile were removed

from the resul";ing solution by purring, leaving a colorless, -iscous oil. which was

t:en into the dry box for storage. Yield was 1.950 g (80.5%, based on

(CII3CH 2NH bLH 3)

A sample cf this ;il was dissolved in CH 2CI, and an nmr opectrim

obtained The spectrum showed a triplet in the methyl region, a complex

mIltiolet in t"'e : -'' '- region (Figure 28), and a broad ribsorpion due to










W'AVL[NGTlH IN MICRONS
55 6 5 7 75 B


00 1700
WAVEcNUV,0R CM'


3 5 4 I 5


Fignure 19. Infrared spectrum of (CHCHii )NIiBH2Ci.


9 i II 1 1 I lo 6 '8 2r 25 30 *0



-0
r 1 I 1 -'' I i _j _










: C ,i 1 ,



+/. i- ,/



0-I 0 j l 60.---'- 5 -- -



1000 600 o000 600 O O 00 3J)















ET2NHBH2I









f 155 Hz

/ obs.


calc.


figure 28 Methylene region of the 1 nmir spectrum of (CH CH ),NHBHJi.
3 _2


215 Hz
I










N-II resonance farther downfield. The chemical shifts and coupling constants

are given in Table IV. The infrared spectrum of (CHI CII ) NHBHi I (Figure 29)

was taken, between KBr plates, and was characterized by a strong peak at

3140 cm due to N-H stretch, and a double at [2450, 2500 cm- due to B-H

stretch. The entire infrared spectrum is given in Table I.

Reaction of diethvlamine-borane and iodine. -- Diethylamine-borane

(0. 840 g, 9.66 mmol) was dissolved in 30 ml of CII2Cl Iodine (2.46 g,

9. 70 mmol) was added piecewise with stirring. The nmr spectrum of the

solution taken 3 hours after all the iodine had been added showed two sets of

triplets in the methyl region (in a ratio of about 6:1). The triplet of lower

intensity and farther upfield was in the same position as the triplet of

(Cli3CiT ) NHBH 1; therefore, the other triplet was assigned to (CPICH2).NHBHI,;.

The nmr spectrum taken after 20 hours showed, in the methyl region, the same

two triplets in a ratio of 20:1, (CHCtI2)2NIIBI 2in excess. The nmr spectrum

after 60 hours showed, in the methyl region, only the triplet due to

(C, CC 2NIHIBHl2. The complete nmr spectrum of (CI3CH2)NIIBHI consisted

of a triplet in the methyl region, a complex multiple in the inethylene region

(Figure 30), and a broad absorption farther downfield, due to N-H resonance.

The chemical shifts and coupling constants are given in Table IV.

In another experiment, diethy;,namine-borane (0. 812 g, 9.31 mmol) was

dissolved in 75 -m of C6 E in a 125 ml flask fitted with a condenser. The system-

was p-c tf-ctcd from the atmosphere by a stream of Ni. through a T-tube above

the eonden;-er. odine (7. 00 g, 27.6 mmoni waaj added pieer-wise with stirring

until a perinonent color was imparted to the solution, thy;: '' rest f the i.Jo(i








WYAVtiNGlli IN MIC(riC'l
6 35 1 75 B 10 It 7 14 16 16 1 70 J 40


Si 5 s0 : 00 0 i

9.11
SI I J- --





I. _ii "


o 53, iH 160 i-00 70050 0
., . .p --: .. . -- -l-- t---- -



6OC 35C 'G 3000 2SOO 10oo 0 0 160


j i rI I"Ii I',"[1P '1 1 I'



..1. \_Il_'_ \1_1J !7 .





m!-J-- :-E::n --


0 1600 1400 1200
WoVENUMIER CM,


I 1-





4f

1.' vGC\.. v:


-I- -- -o
,. -._i It-
-. ---.1 =. 3o


Figure ,. Infrared spcctrun of (CH3 )NIIBF3
32 3


100 800 600 600 0 .110 jCu








,JAVELLNGTO1 IN MiC 4ON^I
.3 34S a i55 6 65 75 5 17 0 1 1 I" IS is ; ?0 ;5 *I; 0
I_l_.. :_r-- l '.LZ 7L-,J 'J _L!.-J '[r "'" i. "* **n -; -r Tn r "i '' 1' 'jl i'-1- "-L 1!
2-'- 1 7-- L- .-i / i : I r r 1 T I 'l i
SI I. ,, ,,i ! i.- I I ' i I T "
Fi i I 1 : .
I I i
.....I /ITl .. 1 4 4 ... . _{ -- .. ...








" 0 "-'^^ 3y 2500 lO00 00 @O lOo iaG I O0 1:00 IOC,0 600 600 000 500 *tC0 .3Ci
I CM
I.i jM I / 1.

Iii /i-I -_L -I _,.


F__ ___,. K-----i~i ilK I L
iLJ F1~i -: ---- I jF j~ J. L r .I I L AJ f LI
5~t ; 25 02 5 0 ,0 0or ,I 6 0 0 08 5 5 S 0 ,
W.VNMO CM-


Figure 17. Infrared spectrum of fraction (A) from the reaction of (CH3)2NH and BC3.
-3 2 3










{ --t--, 1 '--T -4 ,-
.. .. ; i L I . j '
T L: Ii I





I*i 71 a- 1I I I'
,i i 0 6
--.-:--- T --r-

.... ,----- -- ,-- _--. . t ..y -
.-.-1--4 __- L4-- -L-
-- ,..T-- t'I -i .---'t- -- '- -- (--
/ i.LJJ L I ,m t ^J I,

10( 3JOO 300(1 l1^C ;000 0


5 5 6 65 I


GiH i I, "ICRONS
7. 3 P


I0 II I? I 16 16 18 20 's 31 1 0


:iJ lT -I I : -1 ll I ir- J..I ,I;- -.]1 :fI If I 1 V- ; l'r .i- i I
S 1I ',! i i' ti .- i i'
I Ii. i tI ,, i 1 i I; I' i
1I.',


TT.1 I I/ l

I I
F i Q i
1 , I , f , b ,,-

ZJi^ ^ir^'T rj T^ [.l i t i i-i -
'-4C: ____ Iv277
i i-I _______ -~:- Ut I-r
I --J-:----- -
-- r --]-

--l -:----" -- -7 .... [ ------ F-~-----'-----------
5-_. -~_~i I
00 1800 1600 1400 1200 100 800 60( 600 O0 403 6 O 3
WAVENLUMSS CMI


-I-
Figure 12. Infrared ',pectrum of [(CU, ,)pt ,IBI, C1
9 -









separatory funnel. The resulting solution was dried over Na2CO overnight,

then filtered and pumped to dryness. A white crystalline solid remained,

mp =- 34.5-35.5 [literature for (CH3)2NHBH3 = 36 ]. The nmr spectrum

in CH Cl2 showed a single peak at 151.5 Hz, downfield from internal

totramethylsilane taisS). The infrared spectrum was identical to that of

authentic (CH3)2NDBH3 prepared from N-deuterodimethylammonium chloride

and lithium borohydride.

Reaction of N-deuterodimethvlamine-borane and chlorine. -- A sample of

(CH3 2NDBI3 (about 0. 05 g) was dissolved in 20 ml of CH2Cl2,and C12 gas

was bubbled in until the solution turned yellow. (The rate of addition was such

that the color change occurred within 10 minutes.) Volatiles were removed

by pumping, and a white solid remained. The nmr spectrum of this solid in

CH2Cl2 solution showed a six-peak multipict centered at 174. 5 Hz, downfield

from internal TAIS, with an intensity ratio of about 1:3:4:4:3:1. The infrared

spectrum of this solid showed peaks at 3200 cm- (N-H stretch) and at 2400 cm

(N-D strcechl in a ratio of about 1:1, in addition to the pattern foi (CI3 -)NHBCl,,.
-1
(Tiere was an extra peak at 970 em .) Thus this solid was a mixture of

i'CI.3)2NIICi3 and (CH3)2NDBC13 in a 1:1 ratio.

Note that the nmr pattern expected for (CH )2NDBC13 is a 1:1:1:1 quartet,

wiithi pij'ak coincident with the 2nd, 3rd, 4th,and 5th peaks of the (CII)2 NHBCl3

sextet.

Reaction of N-deuterodimethylaminc-bocane and bromine. -- A sample of

CI 3) 2NDBH3. (about 0. 05 g) was dissolved in 25 mi of CT!ICl and 1Br2 as

added oropwise with stirring imtil 'he solution turned yellow. The Lime of









addition was about 15 minutes. Volatiles were removed by pumping, leaving

a white solid. The nmr spectrum of this solid in CHI2C12 solution showed a

quartet (1:1:1:1) with two small peaks, one just upfield and one just downfield

of the quartet. This pattern was centered at 182. 5 Iz, downfield from internal
-1
TMS. The infrared spectrum of this solid showed peaks at 3150 cmn (N-H
-1
stretch) and 2350 em- (N-D stretch) in a ratio of about 1:9, in addition to
-1
the pattern for (CH3)2NHBBr3. (There were extra peaks at 870 cm and
-1
960 cm .) Thus the solid was a mixture of (CH ) NHBBr3 and (CH ) NDBBr3

in 1:9 ratio.

Reaction of N-deuterodimethvlamine-borane and iodine. -- A sample of

(CFI3)NDBH3 (0.0255 g, 0.432 mmol) was dissolved in 5 ml of CH9C12 and 12

(0, 0553 g, 0.218 mmol) added piecewise with stirring. The nmr spectrum of

the resulting solution showed one peak at 165 Hz, downfield from internal TAIS,

with no fire structure. Thus tie product was (CH3)2NDBH2 .

A sample of (CH3)2NDEH3 (0. 1098 g, 1. 83 mmol) was dissolved in 15 ml

of CH2Cl2. 2 (0.7003 g, 2. 73 mmol) was added and the solution stirred. The

nmr spectrum of the solution after 7 hours of stirring showed one peak, with

no fine structure, at 173 Hz, downfield from internal TMS. Thus the product

was (( ) 2NDBHI2

Reaction of N--deuterodimethylamine-borane and hydrogen chloride.-- A

sample of (CI-I2NDBH3 (about 0. 05 g) was dissolved in 25 ml of CI2Ci and

HC. gas was bubbled in for 30 minutes. The HC1 was thus present in large excess.

The v'c1tiles were removed immediately thereafter by pumping. leaving a

gummy solid. The nmr spectrum of this solid in CtriCl showed a 3-peak








multiple centered at 157 Hz, downfield from internal T\IS, which appeared

to be a strong singlet overlapping a doublet. The singlet is thus assigned

to (CH ) NDBH Cl and the doublet to (CH3)2NHBH2C1. The ratio of these two

compounds, as estimated visually, was about 1:3, the deuterated compound in

excess.

Reaction of N-deuterodimethylamine-borane with hydrogen chloride and

chlorine -- A sample of (CH3)2NDBH3 (about 0. 05 g) was dissolved in 20 ml

of CH2CI 2 Both HCI gas and C12 gas were bubbled into the solution, the rate

of addition being such that the HCI was added in about a 10-fold excess over the

CI2. After about 7 minutes of addition, the solution turned yellow, whereupon

the addition of gases was stopped and the volatiles removed by pumping,

leaving a white solid. The nmr and infrared spectra of this solid were nearly

identical in pattern and intensity to that of the solid formed by reaction of

(CH 3)NDBII with C2 alone. Thus, the incorporation of hydrogen on nitrogen

as the result of reaction of (CH3) NDBH3 with CI2 is not affected by the

presence of HC1 in excess.

.'cacti on of N-deuterodimethvlamine-borane with deuterium chloride and

chlorr.e, -- A sample of (CII3)2NDBH3 (about 0. 05 g) was dissolved in 25 mi

of 0C12Cl2. Bth DCI gas and C12 gas were bubbled into the solution, the rate

of addition being such ''tat .he DCI was added in about a 10-fold excess over the

Ci0. (The DC1 wa-, produced from the hydrolysis of benzoyl chloride by D20 as

desp'cribcd in rcference 63. The benzoyl chloride was ia-ted for about 45 minutes

to -.rive off any HC1 produced as a result of hydrolysis by 1120 impurity. D 0

(9). 5%) .vas tLei added,and the resulting DCI was used only matter a good stream










o0 gas was produced. Thus the purity of DC1 was 99% or bcror.) After about

2u minutes of addition, the solution turned yellow, whereupon the addition of

gases was stopped and the volatiles removed by pumping, leaving a white solid.

The nmr spectrum of this solid in CH2 C2 solution showed a six-peak multiple

in the same position as that observed for the product of the reaction of

(CH3)2NDBH3 with C12 alone, but the intensity ratios were quite different.

The outside two peaks had lost intensity while the 2nd and 5th peaks had gained

intensity. The infrared spectrum of this solid showed peaks for N-H stretch

(;200 cn- ) and N-D stretch (2400 cm-1), but the relative intensity was about

1.3, the N-D peak being the stronger. Thus, incorporation of hydrogen on

nitrogen as the result of reaction of (CH3)2NDBH3 with Cl2 is greatly retarded

by the presence of DCI in excess.

Reaction of N-deuterodimethylamine-borane and hydrogen chloride at

low temperature and the synthesis of N-deuterodimethvlamine-mono- and

dichloroborane. -- A sample of (CH3)2NDBH3 (about 0. 05 g) was dissolved in

8 ml of C 2C12 and the solution cooled to about -78 by immersion in a dry
2 2

ice/acetone bath. HCI gaas ws bubbled into this solution for 10 minutes, then

nitrogen gas was bubbled in for 30 minutes to remove excess IIC1. After w\xrmilng

to roon:i ternpetature, the nmr spectrum of this solution was obtained, which

showed one peak iith no fine structure aL 157 Hz,downfield from internal iTS3.

An infrared spectrum of the solute was obtained by using matched NaCl liquid

cell! wiii Cl; Ci as a standard. This infrared spectrum showed no peak at
2 2
-i the a
3200 2m indicating the absence of N-li. Thus the product was (Ct-t9NDSBH12Ci.









in another experiment, the solution of (CH3)2NDBH2Cl and HCI at -78

was allowed to warm up to room temperature without removing HCI beforehand.

Thus (CH3)2NDBH2Cl was exposed to HC1 at room temperature. The nmr

spectrum of this solution after 20 minutes at room temperature showed no fine

structure on the singlet due to (CH3)2NDBH2Cl. Thus there was no exchange

between (CI3)2NDBH2Cl and HCI at room temperature within 20 minutes.

In another experiment, a sample of (CH3)2NDBH3 (about 0.05 g) was

dissolved in 8 ml of CH2C12 and cooled to -78 by immersion in a dry ice/acetone

bath. IC1 was bubbled in at a moderate rate for 50 minutes. The resulting

solution was then allowed to stand, in a dry ice/acetone bath, and the nmr

spectrum twas taken from time to time. The nmr spectrum after S-1/2 hours

showed two singlets. one at 157 Hz [(CII ) NDBH Clj and one at 163 Hz

[(CH3)2NDBHCI2]. downfield from internal TIS. The ratio of products was

about 3:1, the diclloro adduct in excess. After 21 hours the spectrum showed

a ratio of about 9:1 for (CH3)2NDBHC12: (CH)2 NDBH2C1. There was still no

fine structure on the peak due to (CH3) NDB1IC12. After 32 hours the spectrum

showed only (CH3) NDBHCi2. An infrared spectrum was run on this solution

in matched NaC1 liquid cells with CH2C12 as a standard. This infrared spectrum
-1
showed no peak at 3200 cmn Thus the product was (C01,) NDBHC12.
32 2
A portion of the solution of (CH )2NDBHC12 and HCI at -78 as prepared

above was allowed to warm up to room temperature without removing excess

HC! beforehand. The nmr spectrum of this; solution after 20 minutes at room

temperature showed no fine structure on the singlet due to (CH3)2NDBHCI2.










Thus there was no exchange between (CH3)2NDBICI2 and HCI at room

temperature within 20 minutes.

Reaction of N-deutercdimethylamine-monochloroboran.e and chlorine. --

A solution of (CH3)2NDBI2Cl was prepared as described above, the excess

HC1 gas having been removed by bubbling in N2 gas and by a short period of

pumping. C12 gas was then bubbled in until the solution turned yellow. (Time

required was about 15 minutes.) The volatiles were removed by pumping,

leaving a white solid. The nmr spectrum of this solid,dissolved in CH2C C2

showed a six-peak multiple centered at 175 Hz,downfield from internal TMS,

with intensity ratios of approximately 1:12:15:15:12:1. The infrared spectrum

showed both N-Il stretch (3200 cm-1) and N-D stretch (2400 cma- ), but the

N-D stretch was five to six times as intense. Thus the product was a mixture

of (CH3)2NHBC1 and (CH3)2NDBC13 in a ratio of 1:6.

Reaction of N-deuterodimethvlamine-dichloroborane and chlorine. -- A

solution of (CH3 )NDBlHCl2 was prepared as described above, the excess HC1
3 2

gas having been removed by bubbling in N2 gas and by a short period of pumping.

Cl2 gas was then bubbled in until the solution turned yellow. (Time required

was about 5 minutes.) The volatiles were removed by pumping, leaving a

white solid. The nmr spectrum of this solid,dissolved in CH2Cli showed a six-

peak imultiplet centered at 175 Hz,downfie!d from internal TMS, with intensity

rntios of approximately 12:20:25:25:20:1. The infrared spectrum showed both

N-H stretch (3200 cm- ) and N-D stretch (2400 cmn- ), but the N-D stretch was

seven to eight times as intense. Thus the product was a mixture of

(CII)2NHBCl3 and (CH3) NDBC1l in a ratio of 1:8,
3










Reaction of N-deuterodimethylamine-borane and chlorine at low

temperature. -- A sample of (CII3) NDBl3 (about 0. 05 g) was dissolved in

10 ml of CH2Cl2 and the solution cooled to about -78 by immersion in a dry

ice/acetone bath. Cl2 gas was bubbled into the solution for 10 minutes at

which time the solution was yellow. The volatiles were removed by pumping,

leaving a white solid. The nmr spectrum of this solution showed a quartet,

centered at 175. 5 Hz,downfield from internal TMS, with intensity ratio of

approximately 1;1.2:1.2:1. Two very low-intensity satellite peaks were visible,

one upfield and one downfield of the quartet, in positions corresponding to the

set and 6th peaks of the (CH3)2NHBC13 sextet. The infrared spectrum of this

solid showed both N-H stretch (3200 cm ) and N-D stretch (2400 cm ), but

the N-D peak was fifteen to twenty times as intense as the N-HI peak. Thus the

product is a mixture of (CII3 ) 2NHBC3 and (CH3)2NDBCl3 in a ratio of

approximately 1:20.

Reaction of N-deuterodimethvlamine-trichloroborane and hydrogen

chloride. -- A mixture of (CH3)2NHIBCI3 and (CH )2NDBC13 in a ratio of 1:1

was prepared, as described above, by the reaction of (CI3) NDBH3 and Cl2.

A sample of this mixture (about 0. 05 g) was dissolved in 25 ml of CII C!, and

HCi gaI, was ubblied into the solution for 2 hours. (During this time, the CI C!2
2 2

lo.~- Ilpy; evaporation was replaced to keep a nearly constant volume of 25 ml.)

Altr the addition of HCI, the volatiles were removed by pumping, leaving a

white solid. The nmr spectrum of this solid was superimposahle on the nmr

spectrum of the starting mixture. The infrared spectrum of this solid was the









smne as the infrared spectrum of the starting mixture within the limits of

variation expected from variation in pellet composition. Thus (CH32NDBC13

and HCi do not exchange to a measurable extent at room temperature within 2

hours,

A mixture of (CH3) NHBC13 and (CII)2NDBC13 in a ratio of 1:20 was

prepared,as described above, by the reaction of (CH3)2NDBH3 and C12 at

-78 A sample of this mixture (0. 0170 g) was dissolved in 3 ml of 0. 12 M

HC1 in CH2Cl2. The ratio of HCl:adduct was thus approximately 3:1. About

1 ml of the resulting solution was transferred into an nmr tube, and the nmr

spectrum was monitored as a function of time. There was no observable change

in the spectrum, even after 24 days. Therefore, within the limits of

delectability of nmr, HCI does not exchange with (CII3 ) 2NDC13 within 24 days.

Reaction of N-deuterodimethylamine-trichloroborane with hydrogen

chloride and chlorine. -- A sample (about 0. 02 g) of nearly pure (CH3)2NDBCI3

prepared, as described above, by the reaction of (CH3)2NDBH3 and Cl2 at

-7 was dissolved in 15 ml of CH2Cl and both HCI gas and C12 gas were

bubbled into the solution for 30 minutes. The volatiles were then pumped off,

leaving a white solid. The nmr spectrum of this solid was superimposable

on the nmr spectrum of the starting material. The infrared spectrum of this

solid vas the same as the infrared spectrum of the starting material within the

limits of variation expected from variation in pellet composition. Thus

(CH4) NNDBCI, and HC1 do not exchange to a measurable extent at room tempera-

ti;re in tIhe presence of Cl2 within 30 minutes.
2




86




Reaction of ejimethylamine-trichloroboranc and deuterium chloride. --

Dimethylamnine-tri'icbloroborane (0. 0353 g, 0.218 nmmol) was dissolved in

1 ml of a 0.22 AI solution of DCl in CH2 C in an nmr tube, and the nmr

spectrum was monitored as a function of time. There was no observable change

in the spectrum, even after 11 days. Therefore, within the limits of

detectability of nmir, DC1 does not exchange with (CH3) 2NIBCi3 within 11 days.











Table I

Infrared Data for Some Amine-Haloboranes'


(CH1)2NHBF3:




(Ci3)2NHBCI3:









(CH32NHBBr3:







(CH,)2NHEIS:







CH '-' 2)NI .H2 C:
I i F 9.%,


3260(m), 3080-2960(m), 1630-1600(w), 1465(m), 1340(m),
[1150, 1040, 940, 910(s)], 705(w), 560(w), 480(w).


3180(s), 3020(w), 2970(w), 2780(w), 2700(w), 2660(w),
2440(w), [1475, 1465, 1450, 1440(m)], 1410(m), 1380(m),
1340(m), 1230(w), 1195(w), [1150, 1140, 1130(m)],
1050(w), 1010(m), 900(s), 840(m), 810(s), 780-700(s),
510(m), 460(w), 375(w), 360(m).


3220(m), 3150(s), 2780(w), 2680(w). [1470, 1460, 1445,
1435(m)], 1410(m), 1370(m). 1335(n). 1200(w), [1140.
1130(m)], 1000(m), 595(s), S25(m), 790(s), [710, 650,
665(s)], 455(m).


3300-2950(s), 2770(m), 2640(w), 1500-1430(s), 1405(m),
1360(m), 1330(m), 1200(m), [L135, 1125(m)], 1040(w),
1015(w), 990(m), 895(m), 820-790(m), 775(m), [645,
620, 600(s)], 550(w), 420(m).


3190(s), ,040(sh), 2990(s), 2945(w), 2910(w), 2800(w),
2840(w), 2450(s), 2373(m), 1470(m), 1445(m), 1395(m),
1370(w), 1315(w), 1260(m), 1190(m), 1150(s), 1130(s),
j100(mi, 1065(m), 1030(m), 1020(sh), 980(w). 870(m),
790(ni). 710(w), 615(m).









Table I (continued)


(CII3CII2)NHBH2I:







(CHCH2)NHBHCL2:







(CH3CH2)NHBC13:







(CIICH 2)NHBBr3:


3140(s), 2990(s), 2980(s), 2940(w), 2500(s), 2450(s),
1465(m), 1450(m), 1420(w), 1390(m), 1360(w), 1250(m),
11SO(m), 1130(s), 1110(s), 1050(s), 1025(s), 1005(s),
885(m), 865(m), 810(w), 790(m), 760(w), 700(w).


3200(s), 2990(s), 2950(m), 2500(s), 1460(s), 1400(s),
1290(sh), 1265(m), [1175(w), 1140(s), 1115(sh),
1070(m), 1025(s)], 910(m), 880(m), 810(sh), 790(m),
745(m), 680(s), 545(m), 520(m).


3190(s), 2970(s), 1470(s), 1450(sh), 1370(s), 1290(s),
1260(m), 117C(m), 1125(m), 1095(s). 1060(m),
1030(m), 1010(m), 910(m), 870(s), 830(m), [780(s),
740(s), 710(s)], 560(s), 460(w).


3200(sh), 3160(s), 2990(sh), 29G0(s), 2940(sh), 1465(m),
1445(m), 1390(m), 1375(m), 1300(w), 1285(m), 1255(w),
1190(w), 1165(w), 1120(w), 1090(s), 1055(m), 1020(m),
1005(m), 900(w), 875(sh), 870(s), 820(w), 770(s),
700(w). 665(s), 645(s), 520(s), 450(w).


All values ar- in cm Symbols: s strong, n = mreium, w = weak,
sh = shoulder.









Table II

H NMR Data for (CH )2NH and (CH) N Adducts


(CH3)2NH Adducts (CH3)3N Adducts

a b b a b
Bo, ne 5 J J 6 J
CH BHNCH BDNCH CH BNCH
3 3


E3l-i
BH3

BH2CI

BH Br

BH I

MHCI2

PRiBr2,




BF3

3BC!3

BB!-


812


152.0

156.0

161.5

165.5

162.0

163.5

173.5

155

175

182

I 89


1.9

2.9

3.41

3.7


1.6C

2.7

3.1

3.4


,=
J ,KCI! = O. & U z.










Table III
1
H NMR Data for (CH3CH2)3N Adducts


Borane 6a 6 a b b
C13 CH2 H3CCH2 2 BENCH


BH3

BH2CI

BH Br

BH I
BHC]2


BHBr2

EHI2

BF3

BCi3

BBr,,


69.5

69.5

69.5

70.0

76.5

78.0

79.5

72.5

82.0

87.0


165.5

173.0

176.5

179.5

185.5

197.5

201.5

176. 5

203.0

214.5


<2.0c

2.6

2.8


In Hz, downfield from internal T''IS.
b
b Hz.

FNC < 0 z.




91




L I





C-' C-oD N Cl L0 L-

I I i


CCCG G OD CC 00 Cl C Cl ^j


C l 'o 4 Cl 7 C',


4 L- t0 0 -








cM o C)
CID D r1 ^I -7 n
'o CC N l Cl Cl N











aooa
"9 c-- D C,









CC











3 C C

Z; i( IC l- C~ [- C C C LC 1 j C) it



















I #1*< m











CHAPTER IV

DISCUSSION


Halogenation of Amine-Boranes

As noted earlier, the procedures involved in the halogenation of

trimethylamine- and pyridine--boranes were developed by M. A. Mathur and

24,25, 50-52
V. R. Miller, among others. 24, 50-52 In this work these procedures have

been extended to include the synthesis of haloborane adducts of dimethylamine,

diethylamine, and triethylamine. In so doing, some general trends in

reactivity have become apparent, which seem to be consistent with the

existence of a steric effect on reactivity.

The fact that trimethylamine-borane and the pyridine-boranes showed
25
differences in reactivity was noted by Mathur,5 and these differences were

attributed to an involvement of the aromatic ring in stabilizing the boron

during the halogenation of pyridine-borane. However, in this work it has been

observed that the reactivity of secondary amine-boranes parallels that of

pyridine--boranes. Some examples of these differences in reactivity are

enumerlated below:

1) When HCI is allowed to react with R3NBH3, the reaction stops

at monochlorniation. When HCt is allowed to react with

R2NHBH or with C H5 NBH, the reaction does net stop at

mor'ochilorination, but proceeds to dichlorination.

2) The rcactioni of 12 with R3 NBfI3 requires elevated temperature

aroundd 110 ) or a relatively long exposure rime to produce

92









diiodinated products. The reaction of I2 with R2NHBH or

with CsH NBH proceeds smoothly and relatively rapidly

to diiodinated product without reflux. It is important to note

also that more 12 is required to diiodinate R NBH3 than is

required to diiodinate R2NHBH or C H NBH
2 3 5 5 3

a) R3NBH3 + 3/2X R3NBHX + 1/2H2 +HX

b) R2NHBH3 R2NHBHX2
or + X2 or + H2
C H NBH C 5H NBIX

The relative stoichiometries imply that HI will diiodinate 2NHBH3

or C5H5NBH3 but not R NBH3

3) The haloborane adducts of tiimethylamine are not particularly

sensitive to hydrolysis. The haloborane adducts of secondary

amines or pyridine are much less stable towards hydrolysis.

If one looks for a feature common to both the secondary amines and the

pyridine, relative to tertiary amines, one sees that both secondary amines

and pyridines have less steric bulk than tertiary amines. The greater

reactivity of secondary amine-boranes and pyridine-boranes, relative to

tertiary amine-boranes, may be due, therefore, to the fact that there is less

steric hindrance to reaction in the former systems.

Steric effects on reactivity are rot limited to just the differences in bulk

of the amines. There also appears to be an effect which parallels the trends

in the bulk of the halogen, or halogenating agent. The most obvious examples

of this sort of reactivity differences are given below:









1) The reaction of C]2 with any of these amine-boranes proceeds

to trichlorinated product with no difficulty. On the other hand,

1 requires heat and/or long exposure times to fully iodinate

these amine-boranes.

2) The reaction of C12 with the secondary amine-boranes produces

mixtures of products at all stages of reaction, whereas the reaction

of Br2 with the secondary amine-boranes produces nearly pure

dibrominated product before any tribrominated product is

observed. This implies that the reaction of Cl2 with the

intermediate products is not significantly slower than the

reaction of Cl1 with the starting borane. In contrast, the

reaction of Br with the intermediate product 11 NHBHBr2 is

markedly slower than the reaction of Br2 with R2NHBH2Br.

The reaction of 12 with secondary amine-boranes gives clean

product at each step, implying that the reaction of 12 ith

RNHIBH I is markedly slower than with R2 NHBI3 yet

significantly faster than with R NHBHI2.

The last point illustrates a general trend, namely that reaction to

further halogenate an amine-borane is slower the more halogens there are

already substituted on boron. That is, the order of reactivity towards

halogenating agents is:

AmineBH -> AmineBI 2X > AmineBHX, .

it is highly probable that this trend in reactivity arises fIom the increase









in steric hindrance due to the bulky halogen groups on boron. There are

a few halogenating agents, such as C12, which are so powerful that these

differences in reactivity are minimized to the point that at all intermediate

stages of reaction, mixtures of products result.

In one case, an unusual combination of effects is observed. Exhaustive

iodination of dimethylarine-borane or trimethylamine-borane produces

triixodnated product. Attempts to exhaustively iodinate diethylamine-borane

or triethylamine-borane produce, after diiodinated product, ammonium salts.

In this case, the combination of bulky ethyl groups and bulky iodine apparently

results in the cleavage of the B-K bond concomitantly with reaction of the

diiodinated product with iodine.

Reaction of Dimethvlamine and Boron Trihalides

The classical method of synthesis of amine-trihaloboranes involves the

64, 65
direct addition of the amine to the boron trihalide. The reaction of a

tertiary amine with a boron trihalide produces, in good yield and purity, the

mi.ne-trihaloborano. In an attempt to prepare aminoboranes, Brown58

reacted ni; ottiylawminel with boron trichloride, added tricth3 avmine to the

product, and isolated in good yieid dimethylaminodichlcroborane and

:riethyinmirn hycdrochloride. Brown identified the intermediate product as

iti,:thlyiti.mi.o -ho 66
s;i'lihiar mnannr, Go!abeau reacted dimethylamine with boron trbromide,

added 'rieth'*l,;ine to the product, and isolated -n goCd yield dircthyinmino-

ci:bromoblrsaue and trieihylamine hydrobromide. He stated that the i:termediate

pro'uc- wias din"t.hyiamino-triu'ormoborane but offered no basis for that









identification. Goubeau attempted to prepare dimethylamine-boron trichloride

by two methods. In the first case, he reacted dimethyiamine with boron

trichloride and isolated a product by sublimination in vacuum. This product

gave a good analysis for dimethylamine-boron trichloride but was still

impure, even though sublimed. In the second case, he reacted hydrogen

chloride with dimethylaminodichloroborane and isolated a pure compound

with properties similar to trimethylamine-boron !richloride, but no other

identification was given.

The first quantitative examination of the reaction between secondary

amines and boron trihalide was reported by Gerrard, who examined the

reaction of several secondary amines with boron trichloridce. It was determined

tlhat side products were formed in considerable quantity in all cases and that the

ratio of various products formed was dependent on the conditions of reaction,

especially the solvent used and the organic substituent on the amine. The

equation beiow (not balanced) shows the products observed:

+ -
1I2N: B:X -* R2NHBX3 +-'2NHf2 BXt + R PBX2 + (RNI 2BX2 B (?

In thie casn of dimelhyLnmine reacting with boron tiichloride, Gerrard

reportia the in. nation of the 1:1 adduct, dimethviamrnonium tetrachloroborate,

in, di mt!'-,n riedichl:oreooranc. Ho \wevcr, inflartei evidence ea s mentioned

n anotherer :prndut, his -(dmieth: lai,;ine)-dich!eorobc:'onllu tetrac!iloroburate,

a fo t-he a':ta a-'count fuo on!y 2/;l of tic starting material. Also it was

proposed; tor.t thie 1:1 adduct was ionic, nr.!_iw.c, ;Ci "i.N.IiBCi "", wi ch

66
pita pa coitf:is x'c, ,tth tie evidence ohin-fincd by UoAeau on the same










compound prepared by addition of hydrogen chloride to dimethylamino-

dicliloroborane. In summary, Gerrard served notice that the direct reaction

of secondary amines with trihaloboranes did not yield pure adducts, but it

was not yet clear just what the products were.

It is noteworthy that this present work gives the first detailed report

of the preparation of the dimethylamine adducts of boron trichloride, boron

tribromide, or boron triiodide, free of impurities.

The preparation of the dimethylamine adduct of boron trifluoride has

57, 5S
been reported elsewhere; these reports were confirmed in this present

woik, This is the only dilnethyiamine-trihaloborane adduct which is produced

free of impurities by the reaction of dimethylamine and the boron trihalide.

There is, however, some evidence for side reactions, even in this system.

If (CH3)2 NI is in excess in the reaction of (CH3)2NII and BF3, the nmr signal

observed for the adduct collapses to a 1:1:1:1 quartet (J = about 2 Hz), and a
+
singlet appears upfield, the chemical shift corresponding to (CII )3NHI The
+
lack of coupling, in the peak due to (CH)2 NH2 implies an equilibrium exists,

as shown in the equation below:
-t-
(CLH3 NlHBF3 + (CH ) NH (Ci13) NBF + (CHl3)NH2
S32 3 2 3 3 2 2
67
This sor. oi reactivity has been examined by Ronan and Gilje.6 The septet

pattern o the adduct -is restored when more BF3 is introduced. Excess BF

has no efi'ct on tme septet pattern; therefore, reaction to produce BF4 and

C'i3) 2Ni-BF2 does not occur. If the reaction did occut, the septet pattern

shod cl to a doublet.68
should collapse to a doublet.









The reactions of dinethyhmine with boron trichloride and with boron

tribromide were examined in detail in this present work. These reactions are

discussed in the paragraphs which follow.

The first step in the reaction of dimethylamine with a boron trihalide is

certainly adduct formation. Kistiakowsky69 has shown for several cases that

adduct formation proceeds with little or no activation energy. However, once

the adduct is formed, several reactions can occur, for example:

1) Proton abstraction by (C 3)2NH:
3-


(CH ) NIHBX2 + (CH ) NH ->


(C3)2NBX3 + (CH32NH2
or


(CH3)2NBX2 + (CH3)2NH2 X

following which, reaction with BX3 would yield:

(CI3 )1NBX + (CII3)2NH2
32- 3 3 2 2


+ 3 (CH3) NBX2 + (CII3)2 NH 1 .


(CH) ;NBX2 + (CH3)2NH X

Tihe overall equation is givmn below:
+
2(CH3) NH 2BX3 ( CH3) NBX2 + (CH 3)NH NBX

2) DispLicement of X by (CH)2NH:
tj -

(ClI 2 NHBX. + (CHR ),H [(CHt3)2NHI2BX2 X ,
3,C : 2 3 2 2 2

following which, reaction with BKX would y;eld:
+ +
[(CH~, ,-' BXr X BX -B [(CI- ) Nil 2BXBX4.
2 "z <2 2 2 4

The overall equation is given below:
+
2(C ,) NH + 2BX, [(CIt~2NIBX2 BX4

3) aii: .ihbstiraciUon by BX :
+ -
(Cl.Lj NHSX + BX, -4 (CII- .NIHBX BX
2,? 3 j 2 4


4









following which, reaction with (CII )2NH would yield either:
+ + -
(CH3 2NHBX2 BX + (CII3)2NH [(CH3)2NH]BX2 BX4 ,

or:
+-
(CH) 2NHBX2 BX4- + (CH3 2NH -* (CIINBX + (CI3 2Nl BX.

The overall equations are given below:

a) 2(CH3)2NH + 2BX3 [(CH3)2NH]2BX2 BX'
+ -
b) 2(CH3)2NH + 2BX 3 (CH3)2NBX2 + (CH32"H2 BX4

There are, of course, a variety of alternative reactions which could be

written, but 31a would be derived from these three, and with one exception all

would result in the same side products. (The exception is that bis- or

trisaminoboranes might be produced according to ite following reaction

scquence:

R NH
P.NBX2 + R2NH 2 R2NBX 2NHR2 -2--E

MR2N)21)X + R2NI2 X
+-
2 2BX + R2 2N)2B NH2R2----
(RN) BX + KRNHX BX 3
2 2 22

(R N, BX+RN -N (R N) BX*NHF I ZM -
+- BX-
(R2N)3B + R2NH2 X -

(R2N)3B R2NH2 BX4.)

it is important to note that in the three reaction schemes shown above,

the reaction p oduicts for each scheme have a ret stoichi/metry of 1(CH 3)NH:

ImX3, which is tlie same as the 1:1 adduct. Therefore, elemental analysis of

the reaction mixture will not distinguish bewveoe possible reaction paths;

only a compiet.e product analysis will ailow one to comment on the reaction

path.









An examination of the products of the reaction of dimethylamine with

boron trichloride or boron tribromide shows clearly that either sequence 2)

or sequence 3-a) is being followed in these systems after initial adduct

formation, since the product analysis showed the product mixture to contain

principally the adduct and bis-(dimethylamine)-dihaloboronium tetrahaloborate

salts. The other sequences would require the product mixture to contain

ammonium salts and aminoboranes.

In light of the reaction of (CH )2NIIBF3 with excess (CH3)2NH, which

is certainly a proton abstraction reaction, it is surprising that proton

abstraction is not observed to occur to any great extent in these systems. It

is quite likely that experimental conditions play a role in determining reaction

stoichiometry; thus under different conditions the proton abstraction route

might be more favored. Nevertheless, it has been established in other work
70,71
in this laboratory that halide displacement by amine 1 or halide
68
abstraction by a Lewis acid is a laciie r-ure to boron cations.

It aiould he noted (hat the infrared dat- do not require that 73X be the

only anion formed in the reactions which lead to ionic products. If is quite

r.cba ble that some X is present, and in the case of X = Br, the experimental

results (tih observation of BBr, in the volatiles) require the presence of

consi.idc able quantities of Br

Finally. there remained tle possibility that the cati;o could be formed

dire't;y ;'ron the adduct without reaction :rith exccss aji r e or BX This

72
,,.-: s',,getcd by a recent report, tin whi:;h Nth ieoocted the synthesis of

(C Hi3), N1BC10 by direct reaction of (C- ,Ntli and BC ondl the subsequent
I a 1 *3 j








+-
isolation of [(CH3)2NHJ]BCI Cl in 74% yield as the result of rcfluxing that

'ClHi2,N'HBCI3" for 7 hours in Et20. (Another product was Et20 BCI,.)
+ 2
Our dala show a near 70% fieldd of [(CH3)2 NH]2BCIl BC4 as the result of

addition of (CH )2 NH to BC1 To test what happens with authentic

(CIl NHBC13I a sample (0.28 g) was refluxed in 30 ml of Et2 0 for 8 hours.

A s-ialla amount of white solid in the solution was filtered off (0. 016 g, 6%).
+
An infrared spectrum of the solid was very similar to that of [(CH3)2NH]2BC12

BC1 The EL20 was evaporated from the solution to give, after transfer

to a tared vial. 0. 23 g (82%) of a solid whose infrared and nmr check for

(C! )2 NIIBC13 (a small amount of aminoborane was present, as indicated
2 2 3

by nm'r. but this was much less than 5% of the fraction).

In another experiment a sample of authentic (CH3 2NHBC13 (0.27 g)

was dissolved in 40 ml of C 6H, heated at 40-50 for 6 hours, and then

refluxed for 1 hour. A very small amount of solid precipitated during the

refElu, but not enough for any analysis. The benzene was evaporated from

the soiution to give, after transfer to a tared vial, 0.19 g (71%) of a white

soid whose infrared and nmr check for (CH 3) NHBCI3

T;ic, data clearly show that Nbth's report72 as in eiror. The data

I"dicrate hat ',iht NAth called (CIIHj),HBC3 was probably a mixture of adduct
3
r'.i boroJ:i ';- salt iwh'ose composition was not appreciably affected by

I f.-luxig in Et2C.

In *-!r:T.. ~ry, the direct reaction of din'3;ethylam~ine and boron trichloride

r rimOijle i suls i~n the formation of his--(diinoihylamine)-r!haloburoniun

..LalWuua,- ...ats.n along w'ith the -xpected adduces. To obtain pure addicts,









one may take advantage of the low solubility of the boronium salts, or one

may directly halogenate dimethylamine-borane.

Reaction of N-Deuterodimethvlamine-Borane with Various Halogenating Agents

The initial nmr results from the diethylamine-haloboranes were quite

unexpected. The nmr signals for the methylene protons in mono- or

dihaloborane adducts of diethylamine were not a doublet of quartets as

expected on a first order basis, but were instead complex multiplets

containing many more than eight lines. The nmr spectra of the two trihalo-

borane adducts of diethylamine were more complex yet. In an attempt to

simplify the spectral pattern, N-deuterodiethylamine-borane was prepared

(in a manner exactly analogous to that described in the experimental section

for the preparation of N-dcuterodimethylamine-borane*) and used as the

starting material for several halogenation reactions. It was expected that

the deuterium atom in place of the proton on nitrogen would not couple to the

meltylene protons, and would thus produce a simplified pattern for those

protons. iHcewever, the patterns were not any less complex,** and indeed the

infrared spectra of the trichloroborane adduct thus prepared clearly showed

the presence of both N-H and N-D bonds. This indicated that some exchange

h-: taken place in the course of reaction. Some qualitative experiments



1UPAC approved name is dimethylan.ine (ND)-borane.73
Sit should he noted that the complex pattern observed for the methylene
protons of the two expected products (CH3CHg)2NDBHI2I and (C113CH 2)NDBHI2
,wrc'i exactly that which one would predict for the pattern of (CII3CHi2)2iNHBHI
and (CITCH,) N 1B-IIII 2, if all coupling constants between the proton on
nitrogen and t.e methylene protons were set equal to zero.









were performed which seemed to indicate that the extent of exchange of N-H

for N-D as the result of chlorination of (CH 3CH2)2NDBH3 by Cl2 was

unaffected by added HC1, nor was the product affected by exposure to HIC1

or DC1. However, the system was analyzable only by means of the infrared

spectra since the nmr spectra were much too complex. Accordingly, a

thorough examination of the analogous dimethylamine system was performed,

as described in the experimental section.

The following facts stand out as the principal results of these experiments:

a) At -78 HC1 does not exchange with (CH3)2NDBI3, (CH3)2NDBH2C1,

or (CH3)2NDBIHC12.

b) HC1 alone does not exchange with (CH) 2NDBH2CI or (CH3 )NDBHCI2

within a half-hour at room temperature.

c) Exchange and chlorination both result from the reaction of HC1 and

(CH3)2NDBH3 Therefore, it is not known whether HCI will

exchange with (CH3)2NDBH3 without reaction occurring.

d) HCI, even with C12, does not exchange with (C113)2NDBC13 within

a half-hour at room temperature.

a) (CH3 2NDBC13 and HCI do not exchange within 24 days at room

temperature. (CH 3)NHBC13 and DC1 do not exchange within 11

daiy at room temperature.

), Dring the reaction of (CH3)NDBH3 (CH3)2DBI2C1, or

(C;II32NDBHC2, with C1, exchange occurs. In the case of (CHi, NDBH

exchan-ge occurs even at -78 although the extent of exchange is much

pess than at room temperature.









g) The addition of HC1 in large excess during the reaction of

(CiH3)2NDBII and Cl2 has n6 noticeable effect on the extent of

exchange, while the addition of DC1 in large excess during the

same reaction markedly reduces the extent of exchange of H for

D on nitrogen.

h) Bromination of (CH3)2NDBH3 to the tribromoborane adduct with

Br2 is accompanied by very little exchange. Iodination with I to

the mono- or diiodoborane adduct does not induce exchange at all.

Several plausible mechanisms could be proposed for this exchange

process; however, only one mechanism appears to be consistent with all

the ibcts.

The idea that direct exchange between an N-D bond and an IIC1 molecule

might be the principal mechanism is certainly attractive, but if it were,

rhe addition of excess HCI to the system (CII3) NDBH + Cl2 should have
3 3 2
increased the extent of hydrogen incorporation, whereas in fact it had no

effect at all{ g) above.

A mechanism requiring that exchange only occur on (CH )2NDBH3 before

neL reaction takes place to produce halogenated amine-borane is also attractive,

bur the fact that significant exchange is observed during the reactions of

(CH3)NGDBH CII or (CH 32 NDBHC2, ii1h CI2 [f) above effectively disproves

the postulation.

Conver:eyiv, any mn.echanism involving exchange only afterall halogenation

reactions ar~ ocver 's disp-uved by the fact that HICI does not exchange with

(CH3),)NDBC1,, eoen with C02 present, within the lime limits of the reactions

Id) abo',;,).




105


It would not be incorrect to assume that, during the first step of reaction

of (CH3)2NDBH3 and C12, direct exchange could occur between HC1 and

unreacted (CH3)2NDBH3 As evidenced by the reaction of D20 and (CH3)2NHBH3

to produce (CH ),NDBH3, exchange may occur at nitrogen in the presence

of an acidic group, and HC1 would certainly qualify. However, as soon as

an amine-borane is chlorinated, HCI alone can no longer cause exchange if

the molecule is at room temperature, yet exchange does occur in the presence

of C1 Since the function of C12 is to chlorinate the amine-boranc, it would

seem reasonable to conclude that the chlorination process momentarily

"activates" the molecule for exchange.

The one mechanism which appears to be consistent with all the facts

involves the loss of DX from an activated (or "hot") molecule still in an

excited vibrational state as the result of halogenation, thus producing an

aminoborane as an intermediate. If the aminoborane picks up HX, present

in the solution as a product of the halogenation reactions taking place, then

exclhage occurs. If it picks up DX again, no net exchange occurs. The

stepwise equations for this mechanism for the first step of chlorination are

shown below:

(CI 3,) NDBH2C1 +
collision /-" HCI (no exchange)
etc. T

(CHA,)NDB3N + iC -* [(CH )2NDBHCI]* + HCI (CH )2NBII +
3 2 2 3 2 2
(excited DCi + HCI
molecule)
(CH3) 2NHBH2CI


DC1
(exchange)





106


The mechanism would suggest that if chlorination occurs at room temperature,

nearly all of the excited molecules produced would have the energy required

to undergo loss of DCI, but that most of the molecules would quickly

dissipate that energy by collision or some other thermal path. Thus in the

presence of a certain minimum amount of HC1, the extent of exchange would

be determined almost entirely by the probability that an excited molecule will

dissipate its excess energy through the loss of DC1. In other words, the

addition of excess HC1 would have little effect on the ratio of H to D in the

product. On the other hand, addition of excess DCI would tend to sweep out

all of the HCI produced by reaction, leaving very little HCI for exchange, thus

reducing the amount of II incorporated into the product.

Note that this mechanism requires that the intermediate aminoborane

have a lifetime sufficiently long so as to allow escape of DCI fiom the solvent

cage. Otherwise, the aminoborane would simply re-add DCI, and no net

exchange would have taken place.

The mechanism would also suggest that the energy imparted to the

an'rin-boraneas a result of bromination is less than that imparted as a result

of chloriraliin, so that the chances are lower that a broininated molecule

wouid have sufficient energy to allow it to rmdergo loss ofDEr. iodination

would impart still less energy so that no exchange occurs at all. These ideas

are reasonable when one recalls that Cl2 is a much more powerful halogenating

'ig(en: than Dr 2and Br more powerful than 12.

Two noilnts concerning this mechanism deserve further comment. The

first point is that the idea of a "hot" molecule as the key intermediate requic(.,










not only a reactive step which is very energetic but also one which isolates

(or concentrates) the energy of reaction in a B-X bond for a finite period of

time. A concerted process (as opposed to a radical process, for which

precedence exists in systems of this sort) would not concentrate the energy

effectively since the energy of reaction would be distributed over at least two

bonds and would also be released in kinetic energy of more than one molecule.

Thus this mechanism would seem to require a radical process for the halogenation

reactions.

A second and related point is the fact that although on statistical grounds,

one would predict more exchange would occur when more than one Cl is on

boron. one actually finds less exchange occurring. The rationalization which

seems the most reasonable is that the energy isolated in one E-Cl bond at the

moment of reaction is quickly distributed over other B-Cl bonds if they exist.

This is reasonable because the B-C1 motions would be highly coupled to each

other when more than one Cl is attached to boron. Therefore, the first

h':loge-ation step would give the most exchange because the energy would

remain concentrated in one B-C1 bond for a longer period of time.

Tt is interesting to note that one can determine the more

thermodynamiically favored products of these exchange processes by

taking into account the changes in zero-point energy as a result of

exchange:



Zero-point energy is the energy of vibration possessed by a molecule
in 't li;wesi rationall energy state.










If one assumes that entropy effects for these reactions are negligible* and

that the AE values calculated above correspond to enthalpy changes, t then,

since AG = AH-TAS, it follows that for these systems, AG = AE. Now since

AE is positive in all three cases, the reactions to produce N-H and DX from

N-D and HX are not favorable. That is, the thermodynamically Favored

products are N-D and HX.

At this point one may calculate equilibrium constants for these systems

(as shown below):

R2NDBX3 + HX R2NHBX3 + DX,

maidng no more assumptions than those made above. The equation used to

calculate the equilibrium constants is:

log K = -AE/2.303RT

where R = 1.987 cal/mole- K and T = absolute temperature. The results of

the calculation are given below:

X K

Cl 0.776

Er 0.634

I 0.550

Unfortunately, attempts to verify these values failed, because the systems

do rot apparently have a mechanism available for achieving equilibrium

[ e) above].


Tlis is reasonable since there is no change in the number or type of
molecules.

SThis is reasonable since there is no change in volaue associated with
the process.









General H NAIR Results

There can be no question as to the place of nuclear magnetic resonance

in chemistry today. Nmr has become one of the most powerful tools

available for studying a wide variety of chemical systems including such

diverse problems as reaction kinetics, structure, product yields, product

identification, and reaction mechanisms. However, in spite of its usefulness,

there is still considerable controversy in the literature concerning the origin

of chemical shifts.

The IH nmr data for the haloborane adducts of dimethyl- and trimethyl-

amine are given in Table II, and are shown graphically in Figure 32. It is

clear that substitution of halogen (CI, Br, or I) for hydrogen in these amine-

boranes produces a downfield shift of the resonance of the methyl groups.

It is also clear that this downfield shift is larger the larger the halogen

introduced or the greater the number of halogens introduced. It is significant

that the bulkier amine, trimethylamine, shows a greater downfield shift for

a given change in borane substituents than does the less hindered dimethylamine.

There is no a priori reason for this difference if the shift is determined

solely b, an inductive mechanism because the two amines are nearly the

sme i: base strength. It is therefore proposed that the greater steric

interaction, bet.C ece the halogens and the nitrogen substituents in the

trii'ethylanmine adducts results in a greater downfield shift for the methyl

protons in the i, cimetlyiamine--haloboranes. It is also significant that the

trifiuoloborane .Kdducts show almost no change in chemical shift from that of

















200


CI (Hz)
0 ^


i704(CH 3 NH Adducts
I C3 2


(CH3)3N Adducts
U i


BH3 ET2X BHX2 BX3


0 :C

' 'Br


BH3 BHI2X BIIX2 BX3
3 2 1 2 3


Figuir 32. Chemical shift of methyl protons in haloborane adjucts of
(CH ) NH and (ClHI,) N as a function hf extent of halogen
substilution on boron.


______~_I T 1 1_____7_____ ~~_




112


the BH3 addicts. Little or no steric interaction would be expected in these

cases. (In other work in this laboratory, chemical shifts have been

determined for mono- and disubstituted fluoro-75 and oxo-76 borane adducts

of trimethylamine. In these cases, small upfield shifts from the borane are

noted, probably caused by the anisotropic effects of the B-O and B-F bonds.)

Tables III and IV give the H nmr data for the haloborane adducts of

triethyl- and diethylamine, respectively. These data are displayed graphically

in Figure 33. From this figure it is again clear that the substitution of a

halogen (Cl, Br, or I) for hydrogen in an amine-borane results in a downfield

shift of the resonances due to protons in alkyl groups in the amine. The

relative changes in chemical shift for the methylene protons in these

ethylamine-boranes are greater than for the methyl protons in the methylamine-

boranes mentioned above. The order of steric bulk, as indicated by these

chemical shift changes is:

(CH3)2NH < (CH3)3N < (CH3CH )2NH < (CH3CH2)3N,

which order would have been expected.

It is interesting to note the effect of substitution on the chemical shift

of terminal methyl groups in the adducts of the two ethylamines. Mono-

halogeiation has almost no effect at all, dihalogenation produces moderate

downfield shifts, and trihalogenation causes larger shifts yet. This indicates

incccnsing interaction between borne halogens and these methyl groups,

which is consistent with this model, since such interactions should be mininmal

with only i -: hIilogcn on boron but of increased importance with two or three

halogens on boron. A consideration of space moneis suggests that the




113


6C2 (Iz)
2


Sf (CH3CH2)2NH
I / Adducts




CH
i CI(Hz) /'




80 .





I0 1



BH B H 2X BHX BX,


figure 33. Chemical shift of methyl and I
adduct of (CIICH2) 2NH and [
of halc),n substitution on boron,


(CH.CH2)3N
Adducts


//












BI- 2IT BHX2 EX3

Br ne-thi'ylene protons in Ihaoborane
S (ChiCit23IN as' a function of ex.Lent









methyl groups are easily able to nove out of the way of one halogen on boron

but not two or three.

Possible Explanations of the Effect of Boron-Attached Halogens on the
N3MR Shifts of Alkyl Protons in Donor Groups Attached to Boron

For the particular systems under discussion here (namely halogenated

amine-boranes) the chemical shift has been shown to parallel the steric

interaction between halogens on boron and protons in alkyl groups in donors

coordinated to boron. In particular, the effect is a downfield shift upon

substitution of halogens for hydrogen in an amine-borane, the shift being

greater the more halogens substituted or the larger the halogens are.

There are several possible explanations for this phenomenon, all of

which certainly contribute to the overall effect, though probably not to the

same degree. These include:

1) Intramolecular van der Waals' forces between halogens and the

a kyl protons.

2) Magnetic anisotropy around the halogen, in the boron-halogen

bond, or in the boron-nitrogen bond.

3) AMagnetic anisotrcpy in C-C or C-N bonds adjacent to the C-H

bonds, which is induced in some undefined manner by the presence

of the boron atom.

Our analysis of 'he system strongly favors the first explanation, hut we cannot

completely rutle out the other two with the information available.

Studies on proton shielding in haloallka:es (sy;:cms analogous to those

under discussion here) have been carried out by several -roups. Although the










systems studied did not have the steric interference of our systems, the

conclusions reached are enlightening as to the factors to be considered.

Bothner--by and Naar-Colin7 concluded that factors other than electronegativity

were at work in determining shifts of groups more than one carbon away from

the halogen. No hard conclusions were reached, but the suggestion was

made that either mesomeric effects or long-range bond anisotropy effects

may be the cause of the observed shifts. The mesomeric effect would be

best explained as an increase in the contribution of resonance forms which,

through r-bonding effects, increase the electron density of the halogen, as

shown below:

+ +
H H 11 H IH H H H
Si I + I I -
R-C-C-X R-C=C X R- CX R-C-C=X
I I I I I I I i
R'H" R'R" I' R' R'R"

The larger the halogen, the more forms II, III, and IV would contribute. This

effect would be greatly attenuated (as should a purely inductive effect) by the

interposing bonds between the halogen and the proton in the amine-borane

systems.

The overall effect of bond anisotropy was found to be a combination of

two factors, one producing a paramagnetic effect, the other a diamnagnetic

effee!. and in practice, the two effects partially cancel one another. The

more ionic the carbon-halogen bond becomes, the more the paramagnetic term

diminishes; thus, for a C-CI bond, the effect is nrincipall3 a result of

electronegativiry, while for a C-I bond, the diamagnetic anisotropy predominates

(The relative contribution of these factors is apparently affected by the structure




116


of the rest of the molecule.) It should be pointed out that in the equation used

to calculate the quantitative effect, a term for the distance between the aniso-

tropic bond and the affected proton entered as an inverse cube, implying

that this effect would be sensitive to small changes in distance.

The same authors mentioned that the shielding of the protons a to

halogen correlated very well with the dipole moment of the molecule; thus

the effect of the halogens may be to polarize the C-H bonds as a result of

the total molecular dipole moment, which would certainly be largely

determined by the halogens.

78,79
Cavanaugh and Dailey discussed the shielding effects in the same

compounds, but reached different conclusions. They concluded that there

is a contribution to the chemical shift due to the presence of C-C bonds

adjacent to the affected proton, in addition to a contribution as a result of

inductive withdrawing of electron density by halogens. The chemical shift was

viewed as arising from a linear combination of these two effects, and the

effect at positions U, B, and y to the halogen x\as evaluated. Thus,

comparing two compounds:


X-CH CH3 and X-CH(CH2


(I) (I)

the chemical shifts of the o protons in these two compounds are

different because there are different numbers of C-C bonds in the two

cases: .









6*(aH-I) = 6t(elect. of X at aH) + 6**(C-C)

6(aH-II) = 6(elect. of X at aH) + 26(C-C)

However, the chemical shifts of the B protons are about the same, since there

is only one C-C bond adjacent to each B proton:

6(S-I) = 6(elect. of X at HI) + 6(C-C)

6(SH-II) = 6(elect. of X at }H) + 6(C-C).

This sort of correlation worked very well for a wide variety of substituents:

X = Cl,Br,I,OH,CN,COOH, C6H5 -0-,-S-, CHO,NH, NO3,-S -.

The question of the origin of this C-C bond shift was not answered in this

study. The idea that it could be a purely anisotropic effect seemed to the

authors to be incorrect since anisctropy played such a small role in determining

the shifts of the methyl derivative, and since the magnitude of the effect is so

large for the ethyl and propyl derivatives. It should be noted that the magnitude

of the C-C bond shift was found to be proportional to the size of the substituent,

which implies a steric origin for the effect.
SO 13
Spiesccke and Schneider0 examined both H and '1 nmr shifts of a series

of CH X and CH2CH X compounds. It was concluded that the major contributions

;o these shifts came from the inductive and magnetic anisotropy effects of the

I substituent. (The comment was made that these results contradict the

conclusions of Cavensugh and Dailey. 7) It was also noted by Spiesecke and


6((a-I) is the chemical shift of the protons a to the group X in
compound I,

t 6(eleet. of X at aH) is the contribution to the chemical shift of a proton
due to the electronegativity of the group X at a position a to the proton.
** 6(C-C) is the contribution to the chemical shift of a proton due to the
presence of a C-C bond on the same carbon as the proton.
a.









Schneider that even their conclusions were not complete, since several

aspects of the problem remained inadequately explained. The presence

of an as yet uncharacterized contribution to the chemical shift was suggested

as the explanation for these aspects.

Schaeffer, Reynolds, and Yonemoto1 brought a fresh viewpoint to the

problem and came the closest to suggesting a reasonable explanation

applicable t tle system under discussion in this work. It was pointed out that

two factors disfavor anisotropy playing a major role in detennining chemical

shifts in alkyl halides. One factor is that the equation relating anisotropy to

shielding effects (based on the point-dipole approximation of McConnell82

is not valid in cases where the radius of the charge distribution giving rise

to the anisotropy is of the same order of magnitude as the distance between

the point-dipole and the point of interest. The second factor is that in the

case of these alkyl halides, the experimental observations are in many cases

opposite to what would be predicted on the basis of anisotropic effects. In

particular, an upfield shift is predicted for several cases where a downfieid

shift is actua ly observed.

Sclaeffer. et al. resolved this quandry by invoking shifts due to von

der wVaais' interactions. Citing the work of Buckingham, Schaeffer, and

S:cneider 3 on dispersion forces in intermolecular solvent effects, these

jtiuthors pointed out that a downfield shift is expected from each of two types

"f: i .teraction;

1) Interaction in tl;e equilibrium configuiration causes a distortion of the

electron cloud surrounding the nucleus, which cistortion is probably

Xii 't.X-; oiin.










2) The motion of the molecules at moderate temperatures leads to

a time-dependent distortion of the symmetry of the C-H bonds. For

molecules whose C-H bonds are exposed to a "sideways" attack,

this effect could become important.

The first type of interaction is possible if the molecule is at all crowded, and

the second could arise from internal rotations and vibrations of the molecule.

These effects should increase as the number of electrons in the preturbing

atom increases. That is, the effect should be greater for the larger halogens.

Ilaigh, Palmer, and Semple4 expanded on this idea by observing that

three factors involving intramolecular van der Waals' forces contribute to

the determining of chemical shifts (much of the argument is derived from the

work of Marshall and Pople ):

1) At all ranges, London dispersion forces polarize the atoms' electron

clouds towards each other (according to an inverse sixth-power

law), reducing the diamagnetic term, i.e., shifting the absorption

downfield.

2) For ranges a little above or below the conventional van der Waals'

separation, the repulsive overlap forces produce an opposite effect.

3) At very sort range, the major interference of the electron clouds

hinders precession and thus produces a downficid shift.

Clearly in severely hindered cases, the third factor will predominate

Eg\peri;men.t-i evidence has been presented in several cases to supco..: this

86
contention. igaa., Terasawa, and Tori reported observing the dashielding

of protons as a result of steric interference by other proximate hydrogen atoms.










The degree of deshielding was found to be closely related to the distance

between the interfering protons. Winsein, Carter, Anet, and Bcurn' 7'88

reported on the effects of steric compression on chemical shifts and coupling

constants in half-cage and related molecules. These systems involved steric

interaction between an OH group (or-O ) and a proton in another part of the

molecule. Large downfield shifts were again noted, the larger shifts noted

for the more sterically crowded system.

In the system under discussion in the present work, there is certainly

tremendous steric interaction between the protons in the alkyl portion of the

molecule and the halogens attached to boron. Taylor's X-ray work 1 confirms

this. as does a consideration of approximate bond lengths, bond angles and

van der Waals' radii for an H-C-N-B-X system, as shown in Figure 34.

Thus, the nmr results obtained in this work reflect the great steric

strain inherent in these systems. Other effects such as magnetic anisotropy

are not of sufficient magnitude to be important in determining the net chemical

shifts of these compounds.

The only other reasonable explanation for the trend in chemical shifts

of iiiese compounds is Onyszchuk's inductive argument. This argument

is effectively destroyed by the results obtained from tie stud\ of the nmr

spcutra of tlie diethylamine-halobo;anes, as discussed it detail in the section

wv.lch fol!uws this section. The fact that such i large non-equivailence exists

oe ,tweetn 'to protons the same nu!o:ber ot bonds away from bor-on cannot be

rationalizcoa orn the basis of an inductive effect transmitted through the bonds.


































SCALE 12 cm = 108 cm.


Figure 34. Sketch of approximate bond lengths, bond angles, and van der Waa!s'
radii for an H-C-N-B-X system (X Cl, Br, I) showing the potential
for severe overcrowding.










Such an effect must of necessity affect the two protons to the same extent.

The data are, however, easily and reasonably explained on the basis of

storic interference resulting in a downfield shift, which shift increases with the

severity of the interference.

It must be noted that the proposal of a relationship between nmr shifts

and steric factors does not require a relationship between those steric factors

and net stability of the molecules involved. In other words, the assertion that

steric interactions between halogens on boron and alkyl groups on nitrogen

increase with increasing size or number of halogens does not assert that this

increase in steric interaction is accompanied by a concomitant destabilizing

of the adduct (or, more correctly, of the B-N bond), since other factors may

increase the bond strength. Indeed, one only has to compare the gns-phase

dissociation data of Onyszchuk with the X-ray work of Taylor,12 both

concerning the trimethylamine-haloboranes. Onyszchuk showed clearly that

boron tribromide forms the most stable adduct, boron trichloride next, and

boron trifluoride the weakest. Taylor showed that the B-N bond lengths in

the BBr3 and BC13 adducts were, within experimental error. the same. One

:must conclude, therefore, that the bond length is determined by a combination

:if bndiong t eingh and steric repulsion, both of which increase from the BC13

to the BBr, aciduct, sc that the bond length is not ch.,nged sigr.ificantiy. Had

the increase ini steric repulsion caused a decrease in the B-N bond strength,

tie B-N bond length in the BBr3 adirict should have been consider 'bly longer

than in the aCi., adduct. Had there been no increase in steric repulsion, the
'3










increase in B-N bond strength should have caused the B-N bond in the

BBr3 adduct to be considerably shorter than in the BCI adduct.
3

Diethylamine-Haloborane H NAIR Results

The pattern of the H nmr resonance observed for the methylene

(CH2) protons of the diethylamine-haloborane adducts is not a simple doublet

of quartets, as one might predict on a first order basis. Instead the pattern

observed is extremely complex, due to the non-equivalence of the two

methylene protons. These spectra were analyzed as the AB portion of an

ABCD3 spectrum using a computer program, as described in the procedures

section. The results of these analyses are given in Table IV and are

shown graphically in Figure 35. That the methylene protons of these adducts

should be non-equivalent is not surprising. Magnetic non-equivalence of this

sort is possible in any system in which a methylene group is attached to a

tetrahedral center with three different groups on it. [This is illustrated

in Figure 36, in which Newman projections are shown for 1) RCH CXYZ

and 21 CH CH I2N(H)(Et)(BXY2), where Et = CH CII3, and X, Y II, Cl, Br,

T Depending on the differences in size among the three groups, different
39-91
arnoou'ts of ncr-equivalence may arise as the result of two effects:

1) intrinsic non-equivalence- near free rotation but with different

tfrsiona' preferences for similar positions of the methylene protons.

Thi? corresponds to a rotational potential ernery' curve which has

mi:in'ra in e-eregy whici o cur at positions other than at 0 dihedral

angles and which his relativel- small differences in energy between

maxima and mininm.











6 (iz) //





//













C1Ii2 () CII2 (A)







d t







H, BHiX BHX BX BHX- BRE X BUN2 BX
2 2 *3 [7 T 3 2 3
Br

F;~.u. eS. CLaemical shift of non-equivalent methylene prctois in haloborane
;ddCcts of (CiHCfHT .,NH as e function of extent of halogen
slbstitution on coron.


















13






"








CH







Et






Figure 36. Newman projections showing similarity between CH 3CH CXYZ
: a, CH3CH22N(H)(Ft)(BXY Y).









2) Non-equal conformer population- the relative rotation population

is such as to make one (or tho) rotation(s) favored over the otherss.

This corresponds to a rotational potential energy curve which has

one (or two) minimum (a) in energy much lower in energy than the

others) and which has relatively large differences in energy

between maxima and minima.

Non-equixalence of methylene protons in a borane system has been observed in

two instances; however, in neither case was the resulting pattern analyzed.

Coyle and Stone34 observed a complicated methylene spectrum for

(CH 3CTI2)2SBH which they attributed to non-equivalence of the methylene

orctons; in this case, the lone pair of electrons on the sulfur acted as a

stereocheiically significant group. Rothgery and Hohnstedt92 observed an

AB pattern for the methylene protons between nitrogen and the earbonyl group

in HCH CHOC(O)CH2NH(CH3)- BX (X = H, Cl). Additional comments on
23 2 C 130

their work will be made in a later section.

The values for 6 for the diethylamine-haloborane adducts, as shown
Ct2
in Figure 33, are averages of the values fir the two non-equivalent protons.

Figure 35 shows graphically the actual values for the chemical shifts of these

non-equivalent protons. (Using standard notation, the downfield proton is

iubzL
shoa:',s iiat proton A is extremely sensitive to halogen substitution on boron,

vh!e prn.ot;.n P is relatively insensitive. This implies that proton A is, on

the averag-:,''ory close to the borane moiety, while protoi; B is insulated from









it. (The anomalous changes in the shift of proton B between BH I and

BHI2 and between BHBr2 and BBr3 are significant. More will be said about

this later.)

At this point it is possible to establish more definite information

concerning preferred rotational conformers by considering, in addition to

chemical shifts, the changes in the coupling constants between the nitrogen-

attached proton and the two methylene protons, as a function of substitution

on boron (Table IV). The values for JNCI(A are in the range 4.5 to 5.5 Hz
IINGIi (A)

for all the adducts. On the other hand, the values for J HNCH(B drop to
HNCH(E)

about zero when more than one halogen is substituted on boron. The Karplus
93
rule for ethane-type systems predicts that J iCC will be at a minimum
HiCCH'

when the dihedral angle, t, between the two protons is 90 :
0 2 [
J = J cos 4 + C [0 < D 5 90 ]

J = 80 cos2% + C [90 S 1800]

0 180 0 180
(J J and C are constants; J and J are the coupling constants for

S= 0 and S10, respectively, and C is the coupling constant for i = 90.

0 380
Typically J J 5-10 Hz and C 0 Hz.) The coupling constants

observed in our system imply, then, that when more than one halogen is

substituted on horon in diethylamine-borane, one rotamer is strongly preferred,

and that this rotame- has a dihedral angle between the N-H bond and the

C-1H(B) bond of about 90.

The three possible rotamers for this system are shovn. as Newman

pro.ictio:ns looking down one C-N bond, in Figure 37. [Rotamers I and II are

shown torsionally twisted by interaction between the methyl group and the




128














X -i Y )90 ( 19 --Et

E- H
Et BXY2



A B




CH3
Et BXY2

l2

H Hi









Figure 37. Newman projections showing roi-meric isomers of
(CH CI2 2,NHBXY2 (with torsional preferences).










nearest bulky group on nitrogen. Protons A and B are labelled according

to dihedrai angle, proton B being 90 (dihedrally) from the N-H group.]

Rotamer III is ruled out because it has the bulky methyl group between the

two bulky groups on nitrogen. Rotamer II is ruled out because it would

require that proton B be more sensitive to halogen substitution on boron.

Rotamer I is the preferred rotamer since it has a dihedral angle of about 90

between the N-H bond and the C-H(B) bond, and since it has proton A near and

proton B well-insulated from the borane moiety.

Recall now the anomalous changes in the chemical shift of proton B

between B!2 I and BHI2 and between BIIBr2 and BBr3. The fact that proton B

is totally insensitive to the new halogen substituted on boron implies that this

is the point at which rotamer I is nearly completely "frozen out. Thuls it

takes one iodide or two bromides to freeze the rotation, but even with three

chlorides, there is still some freedom of rotation. This trend is perfectly

in line with the trends in steric bulk of the halogens.

Further Comments

Comments on the Work of Rothgery and Iohnstedt

It is interesting to apply the sort of viewpoint developed in this present

i.ork Lo toe observations of Rothgery and Hohnstedt on non-equivalent

i.ethviene protons in borane adducts of amino acid esters. In particular, the

compound CHI3CH2 OC(O0)CH N(H)(CH3)- BCl shows non-equivalence of the

meihxlene protons between the nitrogen and the carbonyl group. An

exariintion of the nnmr spectrum of this compound, as recorded in Figure 2









of this reference, shows JH JHN (B) SHz, JBNCH(A)
HNCH(A) HNCH(B) BNCH(A)

411z, and JBNCH(B) OHz. where CII(A) and CH(B) are downfield and upfield,

respectively. The zero value of JBNCH(B) strongly suggests some sort of
BNCH(B)
preferred rotameric population distribution. If one assumes that a Karplus-

Lype relationship exists for boron coupling to hydrogen across three bonds,

then these observations require that there be a ~90 dihedral angle between

born and CH(B) in the preferred rotamer, and.secondly, that this preferred

rotamer be much more favored than other rotamers, since large contributions

of other rotamers would cause the average JBN (B) to be non-zero. In addition,
BN CH(B)
the near equal JNHCH values require that this rotamer not have a 90 dihedral

angle between N-II and either CH2. Additionally. the magnitude of the value

suggests that the nitrogen proton is not between the two CIIH protons, since

if it were, the two dihedral angles would be expected to be equal (i. e., 60 )

for the values to be equal, and the maximum value of J predicted for a dihedral

angle of 60 would, by analogy to carbon systems, be about 6 Hz. Finally,

since proximity to borane has been shown by our work to cause downfield

shifts, the greater shift arising from the greater proximity, the proton which

is farther away from boron is the proton which must have a dihedral angle to

boron of 90 This requires that the boron be between the two protons, 90

from one and 30 from the other. This leaves two possible rotomeric forms

(A and B in Figure 38). Rotumer A can be ruled out since the 90 dihedral

angle between boron and CH(B) would require a 90 dihedral angle between

N'-H and CI(A). ,vhich means JNCH(A) would be zero. Rot:rmer B on the
HNOH(nA)




131










O OEt





CH3



90
EC1

A



O OEt

C
CH3






90 BCI

B





Figure 38. Newman projections of two rotameric forms of
CH3CH2OC(O)CH2N(II)(CHI3) H BC13'










other hand has angles between N-H and the two CH protons of 30 (proton A)

and 150 (proton B), and the Karplus rule would predict near equal values for

these two coupling constants. Rotamer B is thus the preferred rotamer.

It must be pointed out that this deduction is based on an unestablished

assumption, and thus the result is not to be taken as surety. Indeed, steric

considerations would at first glance lead one to predict that this rotamer is

not so favored as the ones which have no bulky groups eclipsing one another.

Nevertheless, the ester group is not spherical, and it is possible that in

this particular molecule, the group is rotated such that its interaction with

the nitrogen-attached proton is more significant than its interaction with the

nitrogen-attached methyl group.

Comments on the Solvent Dependence of NMIR Spectra of Diethylamine-Holoborines

The CII2 regions of the IIi nmr spectra of the diethylanine adducts

were quite sensitive to changes in solvent. In methylene chloride, the borane

(BH3) adduet showed a 1:4:6:4:1 quintet in the CHI2 region, which is consistent

with the CH2 protons being equivalent, but in benzene solution, the pattern

shifted upfield. and showed the complexity associated with non-equivalence.

Similarly, on changing from methylene chloride to benzene, the CII, region of

the spectruir of the dibromoborane (BIIBr) adduct showed a large increase in

'he shift difference between the two non-equivalent protons, an increase from

:rboui 17 to about 40 Hz. These two observations would seem to be most

kogically e-xplained as follows. Clanging from CI2C12 to benzene as a solvent
causes the resonances due to the thy prots to shift upheld, but t
causes the resonances due to the methylene protons to shift upfield, but the











B proton (the upfield proton in CH Cl ) is shifted farther by this solvent

change than is proton A. This implies a specific solvation of one of the two
94
protons,94 and since the overall shift of both protons is upfield, it would seem

reasonable that the more affected proton, proton B, would be the one

specifically solvated.

It must be noted that the shifts observed in benzene are due to the large

anisotropy associated with the benzene molecule. Indeed the pattern and

position of the resonances of these amine-boranes were not appreciably

affected by changing from CH2CI2 to solvents such as CIICI3, CC14, and

CIi3CN.

Comments on the Temperature Dependence of NMIR Spectra of Amine-
IIa loboranes

The work of Kessler91 on barrier- to rotation of tert-butyl groups in

organic molecules suggested that in some mono- and disubstituted

trimethylamine-boranes, hindered rotation might be observed. If the rate of

rotation could be slowed down enough, the methyl groups would become non.-

equivalent as shown in Figure 39. Accordingly, the nmr spectra of

(C 113)NBHI2, (CHQ)3NBHBr2, and (CH3)NBBH2I were examined at low

temperature in CH2C12 and CS2, In all cases the signal remained a sharp

singlet down to below -80 C. It becomes clear then that the barrier to rotation

in these adducts is low; it is probably not low because of a low transition state

energy, but rather because of a high ground state energy. In other words,

these addicts are so highly hindered that there is little energy difference

between staggered and eclipsed forms.

















(1) CII3 CH (1)






CH3 (2)







H

(1)CH3 CH3(1)




X x

CJ3 (2)





F''igure 39. Newnan projections showing possible non-equivalence of
methyl groups in trimethylimine-ha ioaborn n ps rin roducd by
hindered rotati(.n at low temperature.





135


Tile conclusion that there is highly hindered rotation in the diethylamine-

haloboranes suggested that at higher temperature this rotation might be more

likely to occur freely, thus making the methylene protons in these adducts

more nearly equivalent. Accordingly, the methylene region of the spectrum

of diethylamine-dibromoborane was examined at temperatures up to about

140 in s-tetrachlorocthane. (This solvent was chosen because of its high

boiling point, 149 and its resemblance to methylene chloride, the solvent

for most of this work.) Although the spectrum lost some resolution at

higher temperature, probably due to easier relaxation, it did not change in

the number or location of peaks. Thus, the methylene region of the spectrum

of diethylamine-dibromoborane (in s-tetrachloroethanu) is insensiti e to

increases in temperature. This strongly implies that the molecule is so

highly hindered that rotation is not likely to occur, even at 140.

It should be pointed out that these two conclusions are not inconsistent,

though they may at first glance appear to be so. The first case examines

rotation about a B-N bond, the nitrogen end of which is highly symmetric.

The second case examines rotation about a C-N bond, neither end of which is

symmetric. Since the two cases examine rotation about very different types

of bonds, the conclusions should be examined separately; that is, the one

conclusion has no bearing on the other.












BIBLIOGRAPHY


1. D.R. Martin, Chem. Rev., 42, 581 (1948).

2. A.W. Laubengayer and D.B. Sears, J. Amer. Chem. Soc., 67, 164
(1945).

3. J, M1, Miller and M. Onyszchuk, Can. J. Chem., 43, 1877 (1965).

4. H.C. Brown and R.R. Holmes, J. Amer. Chem. Soc., 78, 2173 (1956).

5. F.A. Cotton and J.R. Leto, J. Chem. Phys., 30, 993 (1959).

6. C.M. Bax, A.R. Katritzky, and L.E. Sutton, J. Chem. Soc., 1258 (1958).

7. Wl. Dutton, W.G. Paterson, and MI. Onyszchuk, Proc. Chem. Soc., 149
(1960).

P. J.M. Miller and SI. Onyszchuk, Can J. Chem. 41, 2898 (1963).

9. .I). Coyle. Proc. Chem. Soc. 172 (1963).

10. B. Swanson, D. F. Shriver, and J.A. Ibers, Inorg. Chem., 8, 2182 (1969).

11. P.S. Bryan and R.L. Kuczkowski, Inorg. Chem., 10, 200 (1971).

;2. P.11. Cliippard, J.C. Hanson, and R.C. Taylor, J. Cryst. Mol. Struct.,
in press.

1i. H.G. Schirdewahn. Doctoral Thesis, University of Freiburg, 1965.

V1. P.H. Clippard, Doctoral Thesis, University of Michigan, 1969.

135. B. Swan-on and D. F. Shriver, Inorg. Chem. 9, 1406 (1970).

I6. G.F. LF!vi'der and J. Mi. Miller, J. Chem. Soc. A, 346 (1971).

7. D.G. Br-own, R.S. D'rago, and T. F, Boiles, J. Aner. Chem. Soc., 90,
5705i (1 68).

18. M.F. I.appert, J. Chem. Soc. 542 (1962).










19. D. Cook, Can. J. Chem. 41, 522 (1963).

20. M. F. Lappert and J.K. Smith. J, Chem. Soc., 5826 (1965).

21. I.R. Beattie and T. Gilson, J. Chem. Soc., 2292 (1964).

22. K..S Purcell and R,S. Drago, J. Amer. Chem. Soc., 88, 919 (1966).

23. T.D, Coyle and F. G.A. Stone, in "Progress in Boron Chemistry,"
Vol. 1, ud. II Steinberg and A.L. McClosky, Macmillan, New York,
1964, p. 118.

24. J.W. Viggins and G.E. Ryschkewitsch, Inorg. Chim. Acta, 4, 33(1970).

25. M.A. Mathur, Doctoral Thesis, University of Florida, 1970.

26. V.,D, Phillips, H.C. Miller, and E.L. Muetterties, J. Amer. Chem.
E-e., 8K. 4496 (1959).

27. a. P.N. Gates, E.J. McLauchlan, and E.F, Moonley. Spectrochim.
Acta. 21, 1445 (1965).

b. P.G, Davies and E. F. Mooney, Spectrochim. Acta, 22, 953 (16CC).

c. E. F. Money and B.S. Thornhill, J. Inor.. Nucl. Chem., 28,
2225 (1966).

d. E. J. McLauchlan and E. F. Mtooney, Spectrochim. Acta, 23A,
1227 (1967).

e. E. F, Mooney and M.A. Qaseem, J. Inorg. Nuel. Chem., 30, 1S38
(i96s).

f. E.F. Mooney, M.A. Qaseem, and P.H. \Vinson, J. Chem. Soc., B,
224 (.1 l68).

g. E. F, Mooney and M.A. Qaseem, J. Ilnorg. Nucl. Chem., 30, 1439
(10 6 3).

25C.CW. HeitFch, Irorg. Cher- _, 1019 (1965).

24. '.G P'.inderso and E, F. MoolM'y, in "Annual RIev'iews of' NS, Spectroscopy,"
Vol. 2, ed. E, F. Mocrey, Aciademic P-'ess, New York, 1969, p. 239.

-0. B.F. Sp.ic vogel and J.M. Purser: J. Amn.a. Ch.cm. Soc. 93, 4418 (1971).

31. 1S, CG1towsky, D.W. McCall, and C.P. Siichter ,, J Cheti. Phys., 21,
279 I 5, ). a

32. P.M. Grant and W.M. Litchman, J.Amer.Chmin.Soc 87. 3994 (1965)










33. J.W. Wiggins, Doctoral Thesis, University of Florida, 1966.

34. T.D. Coyle and F. G,A. Stone, J. Amer. Chem. Soc., 83, 4138 (1961).

35. R.R. Holmes and R.P. Carter, Inorg. Chem., 2, 1146 (1963).

36 J.M. Miller and M. Onyszchuk, Can. J. Chem., 42,1518 (1964).

37. J.M. Miller and M. Onyszchuk, Can. J. Chem., 44, 899 (1966).

38. J.S. Hartman and J.M. Miller, Inorg. Nucl. Chem. Letters, 5, 831
(1969).

39. W. Anderson and H.M. McConnell, J. Chem. Phys.. 26, 1496 (1957).

40. A. Derek, H. Clague, and A. Danti, Spectrochim. Acta, 23A, 2359
(1966).

41. D.E. Young, G.E. McAchran, and S.G. Shore, J. Amer. Chem. Soc.,
8. 1390 (196G).

42. A.G. Massey and A.J. Park, J. Organometal. Chem.,, 5, 218 (1966).

4;.. A.H. Cowley and J.L. Mills, J. Amer. Chem. Soc., 91, 2911 (1969).

44. P.H. Clippard and R.C. Taylor, Inorg. Chem. 8, 2802 (1969).

45, C.W. Haigh and J.M. Williams, J. Mol. Spec. 32, 398 (1969).

4. A.A. Boihner-by and S. Castellano, in "Computer Programs for Chemistry.'
Vol. 1, ed. D. F. DeTar, W.A. Benjamin, New York. 1968, p. 10.

47. S. C?:;-ellano and A.A. Bothner-by, J. Chem. Phys., 41, 3863 (1964).

-'. J.A. Poe, V'.G. Schneider, and H.J. Bernstein. "High-Resolution
Nuclair Magnetic Resonance,'' MicGraw-Ilill, New York, 1959, p. 103.

S:. 5. Castelaneo. private comrnunican ion to R.W. King, 1969.

3C. G. E. yischkewitsch nnd J. V. Wiggins, Inorg. Syntheses, 12, 11i (1970).

31. V,A. MI,ithur and G.E. Ryschkewitsch, Inor.. S-rtheses, 12, 123 (1970).

52. M,A. li.thnur. S. Chindra, '.R. Miller, and G.E. Ryschkc.witsch, in
Iprcpa.raJ ion.

:,3. Vj. miller pri'vte co.minu.nict'o n 1969.










54. W.H. Myers and G.E. Rvschkewitsch, Inorg. Syntheses, in press.

55. R.L. Amster and R.C. Taylor. Spectrochim. Acta, 20, 1487 (1964).

56. W. Gerrard, H.R. Hudson, and E.F. Mooney, J. Chem. Soc., 5168
(1960).

57. H.C. Brown and S. Johnson. J. Amer. Chem. Soc., 76, 1978 (1954).

58. J. F. Brown. J. Amer. Chem. Soc. 74, 1219 (1952).

59. W. Kynaston, B.E. Larcombe, and H.S. Turner, J. Chem. Soc., 1772
(1960).

60. V.R. Miller and G. E. Ryschkewitsch, J. Amer. Chem. Soc., 92, 1558
(1970).

01. T.C. Waddington and J.A. White, Proc. Chem. Soc., 85 (1960).

C2. D.F. Alton. R.D. Brown, J.C. Carter. and R.C. Taylor. J. Amer.
Ca2m. Soc. S1 3550 (1959).

,1. H.C. Brown and C. Groot. J. Amer. Chem. Soc.. 64_, 2223 (1912).

64. E. Wiberg and W. Shtterlin, Z. Anorg. Allg. Chem., 202. 31 (1931).

C5. R.C. Osthoff, C.A. Brown, and F.H. Clarke, J. Amer. Chem. Soc.,
73, 40,45 (1951).

;. Goubeau. M. Rahtz, and H.J. Becher, Z. Anorg. Allg. Chem., 275.
161(1954).

"1. R.J. Ronan and J.W. Gilje, Inorg. Nucl. Chem. Letters, 7, 359 (1971).

;S. G.E, Ryschlewitsch and J.W. Wiggins, J. Amer. Chem. Soc., 92,
1790 (197C).

C3. G. Garvin and G.:B, Kistiakowsky, J. Chem. Phys., 20, 105 (1952).

7. G.E. Rschkewitsch, J. Amer. Chem. Soc. .89, 3145 (1967).

1i. C.W, '2kosky. G.I.. Gallowvay. and G.E. Ryschkewitsch, Inorg. Chem.,
3, 972 (1567).

72. H, Ndth P. Schweiger, and F. Ziege!lginberger. Cheni. Bev. 99, 1089
i3006).










73. "Nomenclature of Inorganic Chemistry. J. Amer. Chem. Soc., 82,
5523 (1960).

74. K. Nakamoto. "Infrared Spectra of Inorganic and Coordination Compounds,"
2nd Ed., John Wiley and Sons, New York, 1970, p. 78.

75. J.R. Bergerud and G.E. Ryschkewitsch, private communication, 1971.

76. G.E. Ryschkewitsch, Inorg. Nucl. Chem. Letters, 7, 99(1971).

77. A.A. Bothner-by and C. Nanr-Colin, J. Amer. Chem. Soc., 80, 1728
(1958).

73. J.R. Cavanaugh and B.P. Dailey, J. Chem. Phys., 34, 1094 (1961).

79. J.R. Cavanaugh and B.P. Dailey, J. Chem. Phys.. 34, 1099 (1961).

80. H. Spicsecke and W.G. Schneider, J. Chem. Phys., 35. 722 (1961).

81. T. Schaeffer, W. F. Reynolds. and T. Yonemoto, Can. J. Chem. 41,
2969 (1963).

52. H.M. McCcnnell. J. Chem. Phys. 27. 226 (1957).

83. A.D. Buckingham, T. Schaeffer, and W.G. Schneider, J. Chem. Phys.,
32. 1227 (1960).

34. C.W. Haigh, M.II. Palmer, and B. Semple, J. Chem. Soc., 6004 (1965).

';. W.. Marshall and J.A. Pople, Mol. Phys. 3, 339 (1960).

E \\ Nagata, T. Terasawa, and K. Tori, J. Amer. Chem. Soc., 86,
:746 (I 964).

87. S. Winsreit~ P. Carter, F.A.L. Anet, and A.JR. Bourn, J. Amer.
Chein. oc. S7, 5247 (1965).

A. F.A.L. Anct, A.J.R. Bourn, P. Carter. and S. Winstein, J. Amer. Chem.
Soc. : 97, 524J (1005).

$0. E.I. Snydcr, J. A.ner. Chem. Soc. 85, 2624 (1963).

90. G.'V. Whitesides, D. Holtz. and J.R. Rloberts. J. Amer Chem. Soc., 86,
2628 (1964).

91. H. Keslc. Angew. Chem., Internat. Ed., 0_, 219 (1970).




141





92. E.F. Rothgery and L. F. Hohnstedt. Inorg. Chem., 10, 181 (1971).

93. L.M. Jackson and S. Sternhill, "Applications of Nuclear Magnetic
Resonance Spectroscopy in Organic Chemistry, 2nd Ed. Pergamon
Press, New York, 1969, pp. 281 ff., and references therein.

94. R.D. Bertrand, R.D. Compton, and J.G. Verkade. J. Amer. Chem.
Soc. 92, 2702 (1970).











BIOGRAPHICAL SKETCH


William Howard Myers was born on January 26, 1946, in Oak Ridge,

Tennessee, the second of three sons of Dr. and Mrs. Albert L. Myers. He

grow up in Jefferson City, Tennessee, graduating from Jefferson High School

in May of 1963.

Mr. Myers attended Houston Baptist College in Houston, Texas. He

graduated nmana cum laude with a B.A. in chemistry in May, 1967. During

his college years, Mr. Myers attended special institutes in chemistry at the

University of Arkansas, Fayetteville, Arkansas, and at Argonne National

Laboratory, Chicago, Illinois.

Mr. Myers entered graduate school at the University of Florida in the

fall of 1967, on a National Science Foundation Traineeship. As the result

of a problem with the draft, Mr. Myers resigned the NSF traineeship to

accept a position as an instructor in the Chemistry Department of the

University of Florida for the academic year 1968-69. He was awarded one of

five P;illips Petroleum Corporation Teaching Axwards for his teaching during

this time. Mr. Myers re-entered graduate school 0a the University of Florida

in ihe s;um!.Ine of 196I on another NSF traineeship. In the fail of 1069, he

.as a.ard.:t:! ia tl'. uc.--ear Nation.i. Aeronautics and Space Ad-,iniistration

T'aineesbip.

Mr. :-yers was co.nnaissioned a 2nd Lieute:ant in the Chemncal Corop

i,^ Jhe UFS. Army in June, 1971, as the result of t 'o years' pa cticipatio i.n

R:LTC at the University if Florida.


1 -2




143






William Howard Myers is married to the former Barbara Sue MlcElvany

of Houston, Texas. He is a member of the Baptist Church and the American

Chemical Society.












I certify that I have rcad this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate. in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.







George E. Ryschkewitsch. Chairman
Professor of Chemistry





I certify that I have read this study and that in my opinion it conforms
to acceotabl, standards of scholarly presentation and is oully adequate. in
scooe and quality. as a dissertation for the degree of Doctor of Philosophy.







R. Carl Stoufer
Associate Professor of Chemistry





1 certify that I have rcad this study and that in nm\ opinion it conforl'lls
to Iecu table Paniards of .-cho!rly i ore'cntation and is tolly : pil .I e, in
scope and quality. a? a dissertation for the degree of DocP r 'il'-,onl,.y.







Mrerlc A, jiac(iste
Professor of C':eiliusry















I certify that i have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality. as a dissertation for the degree of Doctor of Philosophy.







Richard D. Dresdner
Professor of Chemistry





I certify that I ha\e read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.







Zoran R. Pop-Stoj>movic
Associate Professor of Mathematics





T'iis dissertation was submitted to the Deoartment of Chemistry in the
Co lege of Arts and Sciences and to the Graduate Council, and was accepted
as partial fulfillment of the requirements ior the degree of Doctor of
Philosophy.

March, 1972


Dean, Graduate School









47 7 S m*lA[ltNGTH ill MICRONS
7 5 a0 I5 10 1 1 16 i3J 3)
1 I T* I I I I, I''I" I i


I I. !- :I I-L. iL r -I 1

'^lF l^^it I i
I I, O




i-"- ..i..... Ii J L hL 7
. .. --T.-' .



mo 16O 1400 200 10,0 000W O 6o0 E CM'
WAIE4UM Cm'


.gu.re 3 Inh'are-i mp:itrum of (CHiI NIiBCl.
312










L.... ..J.I I4 ..l i I V F. -] I II i- '_ 1 l F. '.l IF i


,o l -".- ; I -' i i.
. i -. .i ti F F II
AVE, LENCCl IN c--CO








S . i ...-....-.--- r. "-- | .... l----- - -- -
5I --- -- - .5 6 6 i-5 7 -5 9 i 11 12 1 16






,I f -
-L 7


... t T----i;i r ---i. -!-i-- -.-.. ----- .---p---- -



51 -00 2500 200 00 1800 1600 140O 1200 1000 800 60
SWAVENUMBiI Cm'
______ -----




W6V6N6M868 C-'


:6 18 27 75 30 40







I .- '%






L o


50 600 ,)O0 r0 30)


Figure 7. Infrared spectrfin of (CH3)2NHBI3.









starting mater-al had disappeared completely, addition of HCI solution was

stopped and the volatiles removed from the solution by pumping. The oil

which remained in the flask was transferred to a dry box for storage. A sample

of this oil was dissolved in CH C12, and a fresh iimr was taken. This spectrum

showed a triplet in the methyl region, a complex multiple in the methylene

region (1"igure 18),and a broad absorption farther doiwnfield, due to N-H

resonance. The chemical shifts and coupling constants are given in Table IV.

The infrared spectrum of this oil, taken between KBr plates, was consistent

with the formuhttion (CH CH )QNHBH C1 (Figure 19). There was a strong

peak at 3190 cm due to N-H stretch, and a doublet at [2450, 2370 cm-,

due to Ell2 stretch. The entire infrared spectrum is given in Table I.

Synthesis of diethylamine-dichloroborane. -- A sample of diethylamine-

borane (about 1. 0 g) was dissolved in CI2 Cl1 and IIC1 gas bubbled in at a
2 2
moderate rate. The nmr spectrum of the solution taken after adding tHCI for

1/2 hour showed two triplets in the methyl region and a complex multiple in

the methylene region. One of the triplets, the one farther upfield, corresponded

to (CH3CH2,)NHBHIICl. The other triplet was assigned to (CHLCII H2)NHBHCI,.

The ersio of ith, two compounds as evidenced by the ratio of intensities of the

tv/o triplet was about 4:1. the monochloroborane adduct in excess. Addition

of 1CI ,v. s cont,,::ed,and the emr spccira .aken at various time-' showed the

Iripler 'ue to rnonochlo-oborane adduct diminish as the triplet due to dichlorn-

borane adduct increased (no i:w' triplet appearCr during the course of this

reaction,. Aflei about 4 hours, only tho triplet due to dichloroborane adduct









was added and the flask heated until the solution was brought to a gentle reflux.

After 15 hours of reflux, the nmr spectrum of the solution showed two triplets

in the methyl region in a ratio of about 3:1, and in the methylene region, a

quartet overlapping an unresolved absorption. After 40 hours of reflux, the nmr

spectrum showed only a triplet in the position of the triplet which was more

intense in the previous spectrum, and a quartet in the methylene region, in

the same position in the quartet in the previous spectrum. The ratio of intensity
+
was 39:62 (or about 4:6) and the pattern was assigned to (C 3CH2)2NH2 ion,

the coupling of Nil to CH being lost because of rapid exchange of the

N-H protons.

Reaction of diethylamine and boron trifluoride-otherate. -- A solution of

diethylarnine in ether was added rropwise to an ether solution of boron

trifluoride-ethorate. Evaporation of the volatiles left low-melting, colorless

crystals of diethylamine-triflioroborane. On standing in air these crystals

decomposed to a brown oil. An nmr spectrum of this material in CI12C12 solution

showed a triplet in the methyl region, a complex unanalyzable multip'lot in the

mrethyleoe region (Figure 31),and a broad absorption farther downfield, due to

N-H rescnan'ce. The chemical shifts and coupling constants, where assignable,

are ive .n in Table IV.

rU, ez cti,: of :-.oDeurcrodim't hyla mine -Borane witivh various H !oge'ating A gnts

Synthesis of N deiutcrodimcthvlamine-hora-iie. -- A sampic of dimethyl-

amine ilhorane (:bout 5 g) vs added to 40 m! of D20 ar.d stl red for 2 l.oars,

during "ichi tim- most of the borane dissolved. The borane was extracted

from tnoe D, with two 40-ml portions io methylene chloride, using a 125-ml




















ET2NHB1






160Hz


Figure 31. Methylene region of the I nrnr spectrum of
(CHI3CHI2)2NHBF3.


200Hz
I









2 i0

"oiij "-" -

[-L- i


-- l
0 -- -



:! . 5 ii
II


I f-- i-
ll! l,!
Fi-j-a .:!772


VWAVELENr,'" ;N MICROCNS
3 5 4 55 6 65 7 /5 1 9 10 II 12 14 16 t 16 I 0 25 30 40
_;] _[rj_ *_IL__ -, fTi_ L LI i I !: i4 I 1 ii i -- ii !

iT YK v-v 7 7r-IF --






t- V -o : :







WAVINUM3ER CM"


Fpigur 21. Infrared spectrum of (CII CI 2)2 NIBIIC12.







ET2NHBC!-
<_j


220 Hz
i


/I
(II
'i",~/


ill ijil I ii


Figure 22. Methylene region of the 1H nmr spectrum of (CE CII ) NHBCI3.
3 2'2 3'


170Hz

bs

obs.


Il l 1 II










i r Ii 1 3 i 60i





I' i i I tI iI i I
L I Ir






_. I




..l' L.I- K .A-4L--_rl '--=i -


4;).5 3S01 3250 0 2000 ;0 18CO lot00


7






1-







p,


75 8 10 II I 14 16 16 I 6, 30 40




S -/ .40
fT' I I F I












S I ll _l
I J

00 12 00 1000 600 600 500 400 300
WAVENUMBft CM


Figure 23. Infrared spectrum of (CII CH ) NliBCI,.
22 32


WA\'[lf IGlHl Irl MICPO'NS








ET2NHBBr.


240 Hz
i


170 Hz



obs.


/.Y/' II I I/ V I iIiC


Figure 26. MoLhylene region of the H nmr spectrum of (CH CH2)2NHB5r3
F, 1m s3crp of ( C 2)2NHBBr3.























1* i_--.--

c0,3


W'AVLI iGTH IN MICROINS
53 6 6 7 7 8


3 35 4 1 i





I .. .
Z:- ---_ tF T-





I L _t


-:_10 30 Z5 0-. L_
.1 O 3( 500 2000 00


I r


l lll


.. I _I._Ill: .L_ L. Li. i
1800 1600 1400 1o00
W/AVYNUMOrR CM'


10 II 5 1i ; 1e is 0 25 30 40


T' 7 "J~ :1.!


fl!i



LtiL F i~





I-HY


I:








Ii
4] '


-LIL


h in/ I/




i n ,



; rf_ p


'000 d00 00 0 4003 .3 450 300


F'iguru 27. Infrared spectrum of (CI,3CHI2)2NH3BrY.
3j 2, 2 NFD 1


i I 1







II

-f-I 1II 1-;- t-!
-l I v
!21-_ -I :L!_-,:









V AVFEENG'C IN MICFON5
S5 4 5 5 5 5 7 7 U P 10 11 1 14 6 16 IC 1. 25 33 40
T 'F 1 I | 1 fl '1 1 I Il r I l : |" I'TI I 1 0

i J I 7



I _ '' I0I I
i i -- i -i_ 1 i




I ,


.. : .
... I..- -- -- ... .. ,
: -T21 ----_-- .: -- L---_
I1 [_ I_._.


40uO 3.00 3000 05CU 2000 00 10o0 1.0o0 100 1200 1o00 800 600 600 LC C 0 AC
WAVENUMIEER CM'


Figure 29. Infrared spectrum of (CH3CI ) 2NIBiI2 I.






ET2NH-BH2


170 Hz

bs
obs.


calc.


Figure 30. MethyIene region of the 1H nmr spectrum of (CII C 2)2NHBHI2.
3 2 jjC-I


220 Hz
!









-1
Bond vem..

N-H 3100* (3200**)

C1-H 2885*

Br-H 2560*

I-H 2230*

N-D 2160t(2400P)

C1-D 2050 (2090*)

Br-D 1810f

I-D 1540t

From the data above one can calculate the change in zero-point energy for

the process, R R2R3ND + IIX R1R RRX3N + DX, by first determining the

change in frequency, Av, for the process and converting the change in

frequency to a change in energy, AE, by use of the equation, A E = 1/2hcA,
-27 10
where h = 6. 625 x 1027 erg-sec, c = 2.998 x 110 cm/sec, and 1.00 erg =

1. 14 x 1016 cal/mole. The results of this calculation are given below:

X Av (for ND + HX NH + DX) AE(cal/mo!e)
-1
Cl 105 cm1 150

Br 190 270

I 250 355




Values taken from reference 71.

t Cualul:ied from the equation V =\2v, .

Observed in this work for (CH3)2NHBCG1.

Ober',cci in this work for (CH )2NDBClI3
3'2 t















Trends in the Proton NMIE Spectra of Some
Amine-Ha'loboranes: Steric Effects













By

WILLIAM lHO\VARD MYEiS


A OISS .TATION PRESENTED TO 'iTH ORAUi7'AT COlNCIL O'
TI!E LNIVERSITY OF FI OPIDA T-N PAR'.ITAL
FUL. jJLITMENT CF THFI REOQL;IE;ENTS FOI' Ti 17 iEGlRE OF
DOCiTOR I INTLOSCOPH


IJNJV.C:j 2 LIkD


































to Barbara

what has been
is exceeded only by
what is yet to be










TALLE OF CONTENTS

Page


ACKNOWLEDGMENTS ............. ..........ii

LIST OF TABLES .............. ........ viii

LIST CF FIGURES .................. ...... .ix

ABSThRACTT . .. . .. ... .. . ... ......... xii

CHAPTER I. INTRODUCTION ...... ........... 1

CHAPTER i, MATERIALS AND PROCEDURES . . ... .. . .17

Materials .......... ........ ..... ... 17

Apparstus and Methods.... . . . . . .17

Coniocuter Analysis of IH NMIR Spectra of Diethylamiile-HIa'bolrancs 18

CHiAPTER II. SYNTHESES AND REACTIONS . . . . .. 22

Preparation of Triethylamine-Haliboranes . ... ...... 23

Reaction of trieihylamnine-borane and chlorine . ... .23

Reaction of triei:hlanmine-burane and bromine . . . 24

Eeaction of triethylamine-bharne and iodine . . . .. S

Reaction of trithylnamine and boron tiifhcjride- thcrate . 29

F;eparaticn of Din-ietbhlamine- Haloboranes . . ... 29

Snith.esis of diroethyimine-mcnor- ner dichlorobnran-tl . 29

Sy\:lthsic of dirrethy!namin:e-trichlorborate ... .. . 229

-esaction of -dimrnltblvli inir,-toraie Uid bromine .. .. .30

S ,. is of dif th'yla in tribronrobo n . . . .. 33

RPceO ,O of anin thylamine-,ioriane and icdin. .. ... . .S









ACKNOWLEDGMENTS


The author wishes to express his gratitude to his fellow graduate

students, whose friendship and support was and is invaluable.

The interest and concern of the faculty and staff of the chemistry

department will always be remembered, especially Drs. R.C. Stoufer,

W.M. Jones, R.D. Dresdner, and H.H. Sisler. Dr. W.S. Brey is thanked

for obtaining the IB nmr spectra reported in this work. Special thanks are

due to Dr. R.W. King for his aid in the computer analyses of some of the

proton nmr spectra.

The patient, usually cheerful, and always helpful guidance and counsel

of Dr. G.E. Rysc'hkewitscli throughout the course of this work is most

gratefully acknowledged.

The author wishes to thank the Nationai Science Foundation and the

National Aeronautics and Space Administration for their financial support,

and the University of Florida Ccrc. uting Center for making time available on

the IBM1 360-65.

rirnlly. the author w~ -ld hike to e-.rress lis 'ec-c alpreciatior to his.

wife, 1ar bara, whose patient .ipL 'rli cI;d coar ima-', mnc)t of this possible.








Page


Reaction of N-deuterodimethylamine-borane and
brom ine . . . .... . . . . .. . . 78

Reaction of N-deuferodimethylamine-borane and iodine 79

Reaction of N-deuterodimethylamine-borane and
hydrogen chloride.... ........ .... .. .79

Reaction of N-deuterodimcthylamine-borane with
hydrogen chloride and chlorine. .... . . .. . 80

Reaction of N-deuterodimethylamine-borane with
deuterium chloride and chlorine . . . . .... 80

Reaction of N-deuterodimethylamine-borone and hydrogen
chloride at low temperature and the synthesis of
N-deuterodimetbylamine-mono- and dichloroborane 81

Reaction of N-deuterodimethx lamine-monochloroborane
and chlorine . . . . . . . . . . 83

Reaction of N-deuterodimethylaiine-dichloroborane
and chlorine . . . . . . . . . . 83

Reaction of N-deuterodimethylamine-bor ine and
chlorine at low temperature . . . ... ..... 84

Reaction of N-deuterodimethylamine-trichloroborane and
hydrogen chloride. . ..... .. .. .. .. 84

Reaction of N-deuterodimethylamine-t richlorborane with
hydrogen chloride and chlorine ... ...... 855

Reaction of dimlethylamine-trichlorobor-no and
deuterium chloride .................. 8S

CTiATER IV. DISCUSSION . . . . . . . ... 92

Haio;genetion of Amine-Boranes . . . ... ... . .92

Reaction of Dimetbylanrine and .oron Trialides . . ... 95

Reaction cf N-Deut(-rodimethyilarine--Boraoe -ith Variou,.
Hli I :ao tini Agents . . . . . . . . . .102







Page

Synthesis of dimethylaminiie-triiodoboa~e .. ..... 37

Synthesis of dimethylamine-trifluoroborane . . . 3E

Reaction of Dim ethylamine and Boron Trihalides . 42

Reaction of dimethylamine and boron trichoride . 42

Synthesis of bis -(dimethyiamine)-boroniurr chloride .. 45

Synthesis of bis-(diimethylamine)-dicl loroboroniimn
chloride . . . .... . . . . . 47

Reaction of bis-(dimethylamine)-dichloroboronium
chloride and boron trichloride . . . . . 47

Reaction of dimethylainine and boron tribromide . . 50

Preparation of Diethylamine-Halooranes . . . .... 55

Synthesis of diethylamine-borane ... . .. . 55

Synthesis of diethylamine-monochloreborane. .... . 55

Synthesis of diethylamine-dichlcroborane . . . 58

Synthesis of diethylamine-trichlorborae . . ... 61

Reaction of diethylamine-borarn and hydro.et! bromide 66

Reaction of aiethylamine-boviane and brcmine .. . ,

Syinihesis oi diothylarnine-tribron.;.bu'rno ... . 69

STthlesis of diethyiumine-mcnciodohorare .. .. .. 69

fReu, ction of dierhyian:ine-borane Pnd iodine ...... T7

Reaction of aithyianmine and oron' t"iftluoride--er ernte 7(

He c.R *.i ~. -Deut erCi L- cthyiami e-crane v .w' th V rious



Sy" t" ^'-is" r,[ b,-deL er'cl.o-.-,tt};:yi J'-l;;lic,-L'.i' . .. 7(

C gll A ll:i .. . . . . . . .
SN -T v /





Page

General H NMR Results ................. 110

Possible Explanations of the Effect of Boron-Attached Halogens
on the NMR Shifts of Alkyl Protons in Donor Groups Attached
to Boron . . . . . . . . .. . . . 114

Diethviamine-Haloborane H NMR Results . . . ... 123

Further Comments ................ .... . .129

Comments on the Work of Rothgery and Hohnstedt . 129

Comments on the Solvent Dependence of NMR Spectra
of Diethylamine-Haloboranes . . . . ... 132

Comments on the Temperature Dependence of NLMR
Spectra of Amine-Haloboranes . . . . . . 1

BIBLIOGRAPHY ........ ...... .... 136

BIOGRAPHICAL SKETCH . . . . . . . . 142









LIST OF TABLES


Table No. Title Page

I Infrared Data for Some Amine-Haloboranes ..... 87

II IH NMR Data for Haloborane Adducts of (CII3),N . .
and (CII3 2NH S9

III H NMR Data for Haloborane Adducts of (CHI3CH)3N . 90

IV 'I NM3R Data for Haloborane Adducts of (CH CH 2NH . 91
i 5.2.










LIST OF FIGURES


Methylene region of the H nmr spectrum of (CHC1-1 BC2 .13

Mptehyiene region of the 111 nmr spectrum of (C1I CII)BBr
S(BB3 .


Infrared spectrum of (CH3)2NHBCi3 .

H nmr spectrum of (CH3)2NHBC13 ..

Infrared spectrum of (CH3) NHBBr ...
3
H nmr spectrum of (CII3)2NHBrBr3

infrared spectrum of (CH3)2NHB

S2 3
H nmr snectr'um of (CIIH3)2NHBI3 .......

infrared spectrum of (CH)2 NHBF3 ....
3 2 3

IH umr spectrum of (CII3)2NHBF3

Infrared spectrum of fraction 'A) from the reaction
(Ci13),NH and BCl3
+
Infrard spectrum of ((CH,)2NH]2BH2 CIl
2 2
Infrared spectrum of [(CH ) NH] 4~ C
+
Infrared spectrum of [(CH .) 2N' IBCk, BCI4 .
05 4


Page

25

27


of


15. InfrarJ spectim of fraction (A) from the reaction of

3 3

3. :erothylb>ein region of the III 5nmi sp,-crumr, of (CH. CI 2)',NHEH .

7. Icn ied spectrum of (C 3CH2NHBH, .. . .....


: 22 2
P. 'ic-thy;le:e r-eglon of the H 1Iur s11 ,-crum of (CH3.CH1)14 )B".H

13. Infrared spectrum of (CHj C 2) 9NHBH 2Cl . . . . .


. . . 51




Trends in the proton NMR spectra of some amine-haloboranes
CITATION SEARCH THUMBNAILS DOWNLOADS PDF VIEWER PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097629/00001
 Material Information
Title: Trends in the proton NMR spectra of some amine-haloboranes : steric effects
Physical Description: xiii, 143 leaves. : illus. ; 28 cm.
Language: English
Creator: Myers, William Howard, 1946- ( Dissertant )
Ryschkewitsch, George E. ( Thesis advisor )
Stoufer, R. Carl ( Reviewer )
Bathite, Merle A. ( Reviewer )
Dresdner, Richard D. ( Reviewer )
Pop-Stojonovic, Zoraan R. ( Reviewer )
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 1972
Copyright Date: 1972
 Subjects
Subjects / Keywords: Amines   ( lcsh )
Haloboranes   ( lcsh )
Nuclear magnetic resonance   ( lcsh )
Chemical Engineering thesis Ph. D
Dissertations, Academic -- Chemical Engineering -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Abstract: Borane adducts of triethylamine, diniethylaminc, and diethylamine were halogenated using free halogens or hydrogen halides, and the proton nmr spectra of these sininc-haloborane adducts were obtained. The resonances of these adducts and of the haloborane adducts of triethylamine showed a shift to lower field with increased size of halogen or with increased number of halogens on boron. This shift to lower field had been previously attributed to inductive effects, but in this work it was shown that the shift to lower field was due to steric interaction between halogens on boron and alkvl groups on nitrogen. Proton mmr spectra for diethylamine-haloboranes were complex and showed patterns attributable to non-equivalent methylene protons. Computer analyses of the spectra allowed assignments consisted with preferred rotational configuration. . The reactions between the secondary amine, dimethylaiaine, and three boron triflouride were examined in detail- diethylamine and boron trifluoride Borane reacted to produce the 1:1 adduct, diethylamine-boron trifluoride. Diethylamine reacted with boron trichloride or boron tribromide Lo produce mLxtiires of both the 1:1 adduct and bis-( diethylamine)-dihaloboronium tetrahaloborate salts. In order to obtain pure 1:1 adduct, it vas necessary to take advantage of the low solubility of the boronium salts in methylene chloride or to directly halogenate dimethvlamine-borane. During the reaction of N-deuterodimethylamine-borane with chlorine, considerable exchange of hydrogen for deuterium on nitrogen occurred. An extensive investigaition of the halogenation reactions of N-deuterodimethylaniineborane indicated that the exchange occured only via the loss of deuterium chloride from a molecule activated as a result of halogenation.
Thesis: Thesis -- University of Florida.
Bibliography: Bibliography: leaves 136-141.
Additional Physical Form: Also available on World Wide Web
General Note: Typescript.
General Note: Vita.
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000577521
oclc - 13986489
notis - ADA5216
System ID: UF00097629:00001

Downloads

This item has the following downloads:

( PDF )

( INSTR )


Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
        Page vii
    List of Tables
        Page viii
    List of Figures
        Page ix
        Page x
        Page xi
    Abstract
        Page xii
        Page xiii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
    Materials and procedures
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
    Syntheses and reactions
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
    Discussion
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
    Bibliography
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
    Biographical sketch
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
Full Text













Trends in the Proton NMVR Spectra of Some
Amine-Haloboranes: Steric Effects












By

WILLIAM HOWARD MYERS


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA
1972



































to Barbara

what has been
is exceeded only by
what is yet to be












ACKNOWLEDGMENTS


The author wishes to express his gratitude to his fellow graduate

students, whose friendship and support was and is invaluable.

The interest and concern of the faculty and staff of the chemistry

department will always be remembered, especially Drs. R.C. Stoufer,

W.M. Jones, R.D. Dresdner, and H.H. Sisler. Dr. W.S. Brey is thanked

for obtaining the 11B nmr spectra reported in this work. Special thanks are

due to Dr. R.W. King for his aid in the computer analyses of some of the

proton nmr spectra.

The patient, usually cheerful, and always helpful guidance and counsel

of Dr. G.E. Ryschkewitsch throughout the course of this work is most

gratefully acknowledged.

The author wishes to thank the National Science Foundation and the

National Aeronautics and Space Administration for their financial support,

and the University of Florida Computing Center for making time available on

the IBM 360-65.

Finally, the author would like to express his deep appreciation to his

wife, Barbara, whose patient support and care made most of this possible.












TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS .. . . ... . . . . iii

LIST OF TABLES ................ ....... viii

LIST OF FIGURES ................ ......... ix

ABSTRACT ... . ... . .. . .. . . . . . xii

CHAPTER I. INTRODUCTION .... . . . . . 1

CHAPTER II. MATERIALS AND PROCEDURES . . . . . .17

Materials .. .. .. .. .. . .. . .... ...... 17

Apparatus and Methods .. . . . . . . . . 17

Computer Analysis of 1H NMR Spectra of Diethylamine-Haloboranes 18

CHAPTER III. SYNTHESES AND REACTIONS. . . . .... 22

Preparation of Triethylamine-Haloboranes . . . . .... 23

Reaction of triethylamine-borane and chlorine . . ... 23

Reaction of triethylamine-borane and bromine ...... .24

Reaction of triethylamine-borane and iodine . . ... .26

Reaction of triethylamine and boron trifluoride-etherate . 29

Preparation of Dimethylamine-Haloboranes . . . . .. 29

Synthesis of dimethylamine-mono- and dichloroborane . 29

Synthesis of dimethylamine-trichloroborane . . ... .29

Reaction of dimethylamine-borane and bromine ...... 30

Synthesis of dimethylamine-tribromoborane . . ... .33

Reaction of dimethylamine-borane and iodine . . ... .34







Page

Synthesis of dimethylamine-triiodoborane ...... 37

Synthesis of dimethylamine-trifluoroborane . . .. 38

Reaction of Dimethylamine and Boron Trihalides . . .. .42

Reaction of dimethylamine and boron trichloride . 42

Synthesis of bis-(dimethylamine)-boronium chloride .45

Synthesis of bis-(dimethylamine)-dichloroboronium
chloride .. .... ... . ..... .. 47

Reaction of bis-(dimethylamine)-dichloroboronium
chloride and boron trichloride . . . .... 47

Reaction of dimethylamine and boron tribromide . . 50

Preparation of Diethylamine-Haloboranes . . . ... 55

Synthesis of diethylamine-borane . . . .... 55

Synthesis of diethylamine-monochloroborane. .... .55

Synthesis of diethylamine-dichloroborane ...... 58

Synthesis of diethylamine-trichloroborane ...... 61

Reaction of diethylamine-borane and hydrogen bromide 66

Reaction of diethylamine-borane and bromine .... .66

Synthesis of diethylamine-tribromoborane ...... 69

Synthesis of diethylamine-monoiodoborane. . . 69

Reaction of diethylamine-borane and iodine . . .. .73

Reaction of diethylamine and boron trifluoride-etherate 76

Reaction of N-Deuterodimethylamine-Borane with Various
Halogenating Agents ................... 76

Synthesis of N-deuterodimethylamine-borane .... .76

Reaction of N-deuterodimethylamine-borane and
chlorine ...... . . .. .... .. 78









Page


Reaction of N-deuterodimethylamine-borane and
bromine .......... ........ ..... .78

Reaction of N-deuterodimethylamine-borane and iodine 79

Reaction of N-deuterodimethylamine-borana and
hydrogen chloride. .................. 79

Reaction of N-deuterodimethylamine-borane with
hydrogen chloride and chlorine. . . . . . 80

Reaction of N-deuterodimethylamine-borane with
deuterium chloride and chlorine . . . . ... 80

Reaction of N-deuterodimethylamine-borne and hydrogen
chloride at low temperature and the synthesis of
N-deuterodimethylamine-mono- and dichloroborane . 81

Reaction of N-deuteroiimethy lamine-monochloroborane
and chlorine .......... ..... ........ 83

Reaction of N-deuterodimethvlamine-dichloroborane
and chlorine . . . . . . . . . . 83

Reaction of N-deuterodimethylamine-borane and
chlorine at low temperature . . . . . .... 84

Reaction of N-deuterodimethylamine-trichlorcorane and
hydrogen chloride. . . . .... ..... 84

Reaction of N--deuterodimethvlamiinc-ftrichljorborane with
hydrogen chloride and chlorine . . . ..... 5

Reaction of dimethylamine-trichloroborine and
deuterium chloride .......... . ... .6

CHT-P'TEP. IV. DTSCUSSION .. . . . . . . . 92

Hajogenstion of Amine-Boranes . . . . . ... 92

Reaction of Dimethylanminc and Porcn Tri!a'ides .. ... 95

i:r'.t.io f; N-Deitcroinietlhyiom.ine.Bora-e *ithl Variou
:2 ,-r tin Agents . . . . . . . 02









Page

General H NMR Results ................. 110

Possible Explanations of the Effect of Boron-Attached Halogens
on the NMR Shifts of Alkyl Protons in Donor Groups Attached
to Boron . . . . . . . . ... .. ... . 114

Diethylamine-Haloborane 1H NMR Results . . . ... 123

Further Comments .................... 129

Comments on the Work of Rothgery and Hohnstedt . 129

Comments on the Solvent Dependence of NMR Spectra
of Diethylamine-Haloboranes . . . . . .. 132

Comments on the Temperature Dependence of NMR
Spectra of Amine-Haloboranes . . . . ... 13

BIBLIOGRAPHY .... .............. .... . 136

BIOGRAPHICAL SKETCH .................. 142










LIST OF TABLES


Table No. Title Page

I Infrared Data for Some Amine-Haloboranes . . ... 87

II H NMR Data for Haloborane Adducts of (CH3)3N . .
and (CH3)2NH 39

III H NIMR Data for Haloborane Adducts of (CII3CH2 )3 .. 90

IV IH NMR Data for Haloborane Adducts of (CH3CH ) 2NH 91
3 22












LIST OF FIGURES


Figure Page

1. Methylene region of the H nmr spectrum of (CH3H 2)BC13 . 25

2. Methylene region of the H nmr spectrum of (CH3CH2)NBBr3 . 27

3. Infrared spectrum of (CH3)2NHBC13 . . . . . 31

4. 1H nmr spectrum of (CH I3)2NHBCl . . . . . 32

5. Infrared spectrum of (CH3)2NHBBr3 . . . . . . 35

6. H nmr spectrum of (CH3)2NHBBr .... .... . 36

7. Infrared spectrum of (CH3)2NHBI3 .. ............ 39

8. 1H nmr spectrum of (CH3)2NHBI3 ...... ......... 40

9. Infrared spectrum of (CH3)2NHBF3 . .. . . . 41

10. 1H nmr spectrum of (CH3)2NHBF3 ...... ...... 43

11. Infrared spectrum of fraction (A) from the reaction of
(CH 3)NH and BC13 . ................... .. 46

12. Infrared spectrum of [(CH3)2NH]2BH2 Cl . . . . . 48
+

13. Infrared spectrum of [(CH3)2NH2BC12 CI . . . . . 49

14. Infrared spectrum of [(CH3 )2NH]2BC12 BCl4 .... . 51

15. Infrared spectrum of fraction (A) from the reaction of
(CH3)2NH and BBr3 ... . . . . . . .. 54

16. Methylene region of the 1H nmr spectrum of (CH3CH2)2NHBH3 . 56

17. Infrared spectrum of (CH3CH2)NHBH . . . . ... 57

18. Methylene region of the 1H nmr spectrum of (CH3CH2)NHBH2Cl 59

19. Infrared spectrum of (CH3CH2)2NHBH2C1 . . . . . 60
15 nrrd pcrmo frcto (A jfro h ecino










Figure Page

20. Methylene region of the H nmr spectrum of (CH3CH2)NHBHCI 62

21. Infrared spectrum of (CH3CH ) NHBHC12 ........ . 63

22. Methylene region of the H nmr spectrum of (CH3CH2)2NHBC13. 64

23. Infrared spectrum of (CH3CH2)2NHBC13 . ...... .65

24. Methyleeregio petrum of (CH3CH2 HB r 67
24. Methylene region of the H nmr spectrum of (CH3CH 2)2NBHHBr 67

26. Methylene region of the H nmr spectrum of (CH3CH2)2NHBBr3 70


27. Infrared spectrum of (CH3CH2)2NHBBr3 .. ...... 71
28. Methylene region of the H nmr spectrum of (CH3CH2)2NHBH2I .. 72

29. Infrared spectrum of (CH3CH2)2NHBH2 . . . . . . 74

30. Methylene region of the H nmr spectrum of (CH3CH2)2NHBHI .. 75

31. Methylene region of the H nmr spectrum of (CH3CH2)2NHBF . 77

32. Chemical shift of methyl protons in haloborane adducts of
(CH3)2NH and (CH3)3N as a function of extent of halogen
substitution on boron .. . . . . . . . 111

33. Chemical shift of methyl and methylene protons in haloborane
adducts of (CH3CH2)2NH and (CH3CH2)3N as a function of extent
of halogen substitution on boron . . . . . . .. 113

34. Sketch of approximate bond lengths, bond angles, and van der
Waals' radii for an H-C-N-B-X system (X = Cl, Br, I) showing
the potential for severe overcrowding . . . . .... 121

35. Chemical shift of non-equivalent methylene protons in haloborane
adducts of (CH3CH2)2NH as a function of extent of halogen
substitution on boron . . . . . . . . . .124

36. Newman projections showing similarity between CH CH CXYZ
and CH3CH2N(H)(Et)(BXY2) . . . ........... 125










Figure Pv ge

37. Newman projections showing rotameric isomers of
(CH3CH2)NHBXY2 (with torsional preferences). . . . . 128

38. Newman projections of two rotameric forms of
CHI3CH2OC(0O)CH2N(H)(CH3) BC3 ............. 131

39. Newman projections showing possible non-equivalence of methyl
groups in trimethylamine-haloboranes produced by hindered
rotation at low temperature . . . . . . ... 134










Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy

TRENDS IN THE PROTON NMR SPECTRA OF SOME
AMINE-HALOBORANES: STERIC EFFECTS

By

William Howard Myers

March, 1972

Chairman: Professor G.E. Ryschkewitsch
Major Department: Chemistry


Borane adducts of triethylamine, dimethylamine, and diethylamine were

halogenated using free halogens or hydrogen halides, and the proton nmr spectra

of these amine-haloborane adducts were obtained. The resonances of these

adducts and of the haloborane adducts of trimethylamine showed a shift to

lower field with increased size of halogen or with increased number of halogens

on boron. This shift to lower field had been previously attributed to inductive

effects, but in this work it was shown that the shift to lower field was due to

steric interaction between halogens on boron and alkyl groups on nitrogen.

Proton nmr spectra for diethylamine-haloboranes were complex and

showed patterns attributable to non-equivalent methylene protons. Computer

analyses of the spectra allowed assignments consistent with preferred rotational

configurations.

The reactions between the secondary amine, dimethylamine, and three

boron trihalides were examined in detail. Dimethylamine and boron trifluoride










reacted to produce the 1:1 adduct dimethylamine-boron trifluoride.

Dimethylamine reacted with boron trichloride or boron tribromide to produce

mixtures of both the 1:1 adduct and bis-(dimethylamine)-dihaloboronium

tetrahaloborate sats. In order to obtain pure 1:1 adduct, it was necessary to

take advantage of the low solubility of the boronium salts in .r-i', le,,? chloride

or to directly halogenate dimethylamine-borane.

During the reaction of N-deuterodimethylamine-borane with chlorine,

considerable exchange of hydrogen for deuterium on nitrogen occurred. An

extensive investigation of the halogenation reactions of N-deuterodimethylamine-

borane indicated that the exchange process occurred only via the loss of

deuterium chloride from a molecule activated as a result of halogenation.











CHAPTER I

INTRODUCTION


In the last decade, a variety of research techniques have been used to

examine the relative acid character of various boranes and boron halides. On

the basis of electronegativity and relative steric requirements of the halogens,

the accepted order of acid strength had been BF3>BC13>BBr3, but the evidence

presented in these later articles indicates the reverse order to be correct.

Laubengayer and Sears2 were the first to suggest that BF is not the

strongest acid of the boron trihalide series, as the result of a calorimetric

study of the formation of the BF3 and BC3 adducts of acetonitrile. In a later

article, Miller and Onyszchuk added to this work with a calormetric examination

of the formation of the BBr3 adduct of acetonitrile. In a similar study, Brown and

Holmes4 made calorimetric measurements on the formation of pyridine-boron

trihalide adducts, which indicated an order of acid strength of BF3
Stimulated by this new evidence, Cotton and Leto5 performed molecular orbital

calculations of the reorganization energies involved in the formation of donor-

acceptor molecules by boron trihalides. The theoretical evidence showed that

the major hindrance to adduct formation for a boron trihalide involved the break-

up of 7T-bonding between boron and the halogens, and thus that the order of acid

strength was determined by relative extent of i-bonding rather than by relative

electronegativity or relative steric bulk. That is, BF3 is the poorest acid of the

series because it has the most T -bonding to break up.









Additional support came from the work of Bax, Katritzky, and Sutton,

who determined the dipole moments of boron trihabde and borane (BH3) adducts

of trimethylamine and pyridine.

In two related articles,7 8 Onyszehuk and others reported on a study of

the gas-phase displacement reactions of boron trihalides from their trimethyl-

amine adducts. The ordeal of acid strength was shown, unambiguously, to be

BBr >BC 3>BF3>1/2 (B2H6), CoylS9 showed that this displacement can occur
3 3 3 2 6
with or without B-N bond cleavage.

In the recent literature, two articles have appeared giving structural

evidence which confirms this order of acid strengths. Ibers10 reported the

X-ray structure of CHCN'BF3 and CH CN-BC1 The B-N bond : r :ih shortens

by 0.07 A going from the BF3 to the C13 adduct, implying a stronger B-N

bond in the BC13 adduct. Bryan and Kuczkowskil reported on the B-N bond

length in (CH3)3NBF as determined from the microwave spectrum of the compound,

and the complete crystal structures of the other (CH3)3NBX3 adducts (X = Cl, Br,

I) are to be published by Taylor. 12 In addition Bryan and Kuczkowvski reported

a B-N bond length for (CH3)3NBH3 as determined by Schirdewahn13 from the

microwave spectrum of the compound. These bond lengths are given below:

Comrmd Bod l h

C 13CNBF3 1. 630 ( 0. 004)

CH3CNBC3 1i. 62 (= 0.008)



(CH3),3NBH, 1. (65 (-i 0.02)

(CH ) NBF 1. 636 t 0. 004)








Compound Bond length (A)

(CH3)3NBC13 1. 610 ( 0.006)

(CH3)3NBBr3 1. 603 ( 0.02)

(CH3)3NBI3 1. 584 (i 0. 025)

The B-N bond length is seen to become shorter as one goes from BF3 to

BI3 as the coordinating acid. This is consistent with the strength of interaction

(i.e., acid strength of BX3 towards a given donor) increasing in the order

BF3< BC13
would expect the B-N bond to be shorter for a given (CH3)3NBX3 adduct than

for a corresponding CH3CNBX3 adduct. The data above show the B-N bond

length to be about the same for the two BF3 adducts, and significantly shorter

for the CH CNBC13 complex compared to the (CH3)3NBC13 complex. One

possible rationalization for this anomaly lies in the fact that nitrogen in CH CN
3
uses an sp hybrid orbital to bond to boron while nitrogen in (CH3) N uses an sp

hybrid orbital. A consideration of some simple organic molecules predicts a

shortening of 0. 05 to 0. 08 A for a change from sp3 to sp hybridization for one

carbon in a C-C single bond. Thus, CH CNBF3 and (CH3)3NBF3 would have

similar B-N bond strengths only if the B-N bond in CH CNBF3 were at least

0.05A shorter. (Valence force constants were determined for the B-N bond in

(CH3)3NBF3 by Clippard14 and in CH3CNBF3 by Swanson and Shriver.15 The

values are 3.53 mdynes/A [(CH3)3NBF3] and 2. 5 mdynes/A (CH3CNBF3),

demonstrating that the interaction in (CH3)3NBF3 is indeed much stronger.)

One would also have expected the change from BF3 to BC3 to have affected the

two adducts similarly, yet the B-N bond in the CH3CNBC13 adduct is shortened
3 3j









by 0.04 A more than the B-N bond in (CH3 ) 3NBC13. One might conclude that

steric interaction (F-strain) in (CH3)3NBC13 precludes the B-N bond from

shortening as much as expected. One should note that, within experimental

error, the B-N bond length is the same in the (CH3)3N adducts with BC13, BBr3,

and BI although bond shortening would have been expected. This is also

consistent with steric interaction (F-strain) interfering with B-N bonding.
12
Taylor2 comments that these three adducts are sterically quite crowded. In

fact, a comparison of the intramolecular non-bonded distances with the sum of

the van der Waals radii shows that the X-CH3 distance is shortened by about

0. 63 A; this implies a considerable interaction.

Having determined with good confidence that the order of acidity in the

boron trihalides was BF < BC13
to correlate spectral properties of donor-acceptor complexes with acidity. The

basic tenets for three major areas of work* are set down below, with references

to pertinent articles:
14-20
1. Infrared Spectral Shifts. These correlations are based on the effect

of the acceptor group on electron density in the donor group. For example,

coordination of an acid group to a carbonyl group removes electron density from


A fourth type of spectroscopic investigation of donor-acceptor compounds
has recently been reported by Lanthier and Miller.16 They examined the mass
spectra of trimethylamine adducts of the boron trihalides and some mixed
trihalides and found that the results could be explained in terms of relative
acceptor strength of the boron trihalides and borane towards trimethylamine.
The results are consistent with the Lewis acidity increasing with size of the
halogen, and evidence is presented which indicates decreasing boron-halogen
bond strength with increasing size of halogen. This last trend is discussed in
terms of "residual r-bonding" between boron and halogen in the amine-borane
molecule. (For a full discussion of this last point, see Brown, Drago, and
Bolles.17)









the oxygen, thus polarizing the C-O bond. The electrons in the ir-system would

be more localized on oxygen, and the C-O bond strength would decrease.

Another way to express this idea is to say that there is an increased contribution

of a resonance form which has the I-bonding electrons localized on oxygen, and

the formal positive charge localized on carbon. The lowering of the C-O bond

strength would cause the C-O stretch to shift to lower frequency, and the

magnitude of the shift is a measure of the strength of interaction, i. e., a

measure of the acid strength of the acceptor group. Of course, the real

variable one is attempting to evaluate in this manner is the force constant of

the bond, and one must be careful to establish that the bond whose force

constant is to be qualitatively examined by this method has its stretching

motions isolated from other molecular vibrations. If the stretching motion

is strongly coupled to other motions in the molecule, problems in interpretation

may arise. A case in point is the B-N bond in CH CNBX3 adducts, which is

highly coupled to other motions in the molecule. This system has been examined

by Beattie and Gilson,21 Purcell and Drago,22 and Swanson and Shriver.15

An interesting correlation is reported by Coyle and Stone23 for donor

strength in adducts containing B-H bonds. This correlation is based on Cotton

and Leto's argument of reorganization energy, 5 and suggests that the more

stable an adduct is, the more tetrahedral the groups around the donor and

acceptor atoms should be. Thus, since the s character of B-H bonds in such a

system would be less in a tetrahedral case (sp3) than in a case where the

acceptor moiety approaches planarity (sp2), the B-H bond should be weaker

in the more strongly bonded adduct; that is, the B-H stretch should shift to









lower frequency. The data presented by Coyle and Stone are, for the most

part, consistent with this picture. However, a more general point of view

would be to look at the system as a B+ ion coordinated to three donor groups

(X, Y, and Z) and hydride (H ). Changes in the donor groups will affect the

stretching frequency of the B-H bond in a predictable manner, namely, the

more electron-releasing the donor groups become, the more electron density

will be pushed onto the hydrogen, and the more the B-H stretch will shift to

lower frequency. Conversely, the more electron-withdrawing the donor groups

become, the more electron density will be pulled away from the hydrogen, and

the more the B-H stretch will shift to higher frequency. In other words, the

B-H stretching frequency should be a measure of the "hydridic character" of

the boron-attached hydrogen. Thus, the B-H stretching frequency in a series

of (CH ) NBHX2 adducts (X = Cl, Br, and I) may be explained on this basis.

The values are as follows:

Compound VB-H(cm-1) Reference


(CH3)3NBHC12 2480 24

(CH3)3NBHBr2 2510 25

(CH 3) NBHI2 2525 25

One may conclude that the net withdrawal of electron density by X in (CH3)3NBHX2

is in the order Cl
outweigh electronegativity effects, even in tetrahedral boron. (For a full

discussion of this last point, see Brown, Drago, and Bolles.17

2. 1B NMR Data.26-30 There have been two major attempts to correlate
11 26 d ith acid. M s 11 H
B nmr data with acidity. Muetterties was the first to suggest that B- H









coupling constants, in compounds containing B-H bonds, might be a measure of

donor-acceptor interaction, on the basis that the value of the coupling constant

is related to the amount of s-character in the boron orbital used in bonding to
13
hydrogen. (The idea originated in the study of 1C nmr data; see, for example,

Gutowsky et al. 31) One can rationalize the idea as follows. The greater the

acidity of a given borane, the more easily the borane "reorganizes" to

tetrahedral geometry. Since the more tetrahedral the borane, the less s-

character in B-H bonds (sp3 vs. sp2), it follows that if the 11B 1H coupling

constant is directly proportional to the amount of s-character in the B-H bond,

the coupling should decrease with increasing acidity. Conversely for a given

borane, the greater the base strength of the donor group, the more tetrahedral

the borane, thus the less the s-character and the less the coupling constant.

In practice, this concept works only for limited situations, that is, for

systems in which the donor group is varied, or in which the borane group is

varied within limits. A case in point where the correlation does not hold is

in the case of (CH3)3N coordinated to BH BH2C1, and BHC12. The 11B-1H

coupling constants for these three adducts are 101, 123, and 152 Hz, respectively.24

The acidity of these groups surely increases in the order BH3
but the J 1 values indicate the reverse order.
B- H
A rationalization of sorts is obtained by considering the various factors
32
which affect the value of J Grant and Litchman pointed out that J13
1B-1H 13C-1H
values are not only a function of relative s-character but also of effective

nuclear charge at the carbon:


Jobs ( (relative s-character) (relative effective nuclear charge) (J in CH4).
obs









Thus, the correlation of J values with relative s-character works only when the

relative effective nuclear charge is constant. WTen a system is considered in

which great changes occur in effective nuclear charge at carbon, one cannot

use J values as indicators of relative s-character.

In the system described above [(CH 3)N coordinated to BH3, DH2C1, and

BHC12] the changes in effective nuclear charge at boron as a result of chloride

substitution have apparently swamped out any effect of relative s-character.

That is, increasing the effective nuclear charge at boron will increase the

value of J11 B1H, no matter what the effect on relative s-character of the boron
B- H
orbitals used to bond to hydrogen.

The change in B-H stretching frequency in these adducts supports this

argument. The frequencies33 (symmetrical stretch) are 2260-2290 (BH3 adduct),
-1j
2340 (BH Cl adduct), and 2480 (BHC12 adduct) cm-1. The trend in frequencies

implies a lowering of hydridic character in the order BH3>BH CI>BHCI which
3 2 2
would be expected if the acidity increases in the order B13
It should be noted that a rough correlation of structure to J11 1 can be
B- H
made for tetracoordinate boron as shown below:

Secis Approximate JB in Hz
B- H
BH 80

X2B1., XBH3 100 K =N,O donor

borazcres, '. XYBH 120-140 Y Ci, Br, I

boroxiles, XY -'i 140--170

Tiie predo.rinani factor would appear to I, lec,':,"o;ejt.ivity, or, in other words,

for a wide variety oi compounds, J is n I'Uintion of relative effective

nuclear ch'rTCe iLoron.









The second major attempt to correlate 1B mnr data with strength of

27a-g
donor-acceptor interaction was based on Mooney's idea that the change in
11
1B chemical shift from uncompleted to completed borane is a good measure

of the strength of the donor-acceptor bond. Heitsch28 directly contradicted this

idea, pointing out that the order of basicity as determined by 1B chemical

shifts for a series of amines coordinated to BH3, BF3, and B(CH3)3 cannot be

rationalized on the basis of any known or assumed order of complex stability.
29
Mooney, in a review article, suggested that the data Heitsch compiled might

be suspect because of the solvents he used -1, 4-dioxane and acetonitrile. In

addition, Mooney sidestepped the issue by suggesting that the orders of basicity

which Heitsch's data imply might actually be the true orders of base strength

towards the reference acid in question.
27a 11
Mooney has given B chemical shift data for triethylamine adducts
28 11
of BX3(X = F, Cl, Br, I). Heitsch gave a value for the 1B chemical shift

of triethylamine-boron trimethyl, and Muetterties26 gave a value for the 11B

chemical shift of uncomplexed trimethylboron. These data are given below

(all 6 relative to external Et20 BF,):

Borane i6 of free borane 6 of Et3N adduct AS (after Mooney)
B--B- -B--- ---

BF ----- 0.2 + 9.5

BC ----- -10.0 + 37.6

BBr3 ----- -5.1 + 45.1

31_ ---- +59.8 + 53.1

B(CH3)3 -86.4 +13 +100.0










Thus, according to Mooney's criteruin, B(CH)3 would be a stronger acid than

any BX3 (X = halogen) toward Et3N. This perhaps conclusively demonstrates

that Mooney's criterion is not universally applicable, and thus that care should be

taken in its use.
30
Although the work of Spielvogel was not directly concerned with the

question of strength of donor-acceptor interaction, it is nevertheless of such

significance that it should be briefly discussed. Essentially, Spielvogel attempted

to correlate 1B chemical shifts with the nature of the substituents on boron.

13
Such a correlation has been relatively successful in the study of 1C shifts, and

the idea that the same type of correlation exists for 1B is not surprising.

Briefly. the correlation works as follows. So-called "p:u.r w.ise additi'ity

parameters" are determined for possible substituent combinations. Four

substituents may be grouped into six groups of two subhtituents each. For each

group, a contribution to the shift is assumed, and the sum of the six contributions

is a good measure of the 1B chemical shift of the molecule. As an example,

consider Et NBHBr2. The four substituents are Ut N (;cltu l H?: for an

approximate calculation), H, Br, and Br. The six gioup:- are then: LEt N,tH],

[Et3N. Br], [Et3N, Br], (H, Brl, [H, BrI, and [Er, Wrl,. 'hc contributionss for

these groups are (in ppm from Et20- B.1;Y

Group C-',tribur, o to (pm)
B
r r ,:;. H] -1. 6j

[Et3N, Br] -4. 57

[iH Br] 4.52

Fqr, Br] t+, 27









The chemical shift is then calculated to be:

6 E[X, Y = (-1. 63) + 2(-4.57) *2(4. 52) -(6. 2,') = 4. 54 or 4. 5 ppm.

In the present work, the observed value for the 11B chemical shift of this

compound was 2.1 ppm referred to the same standard.

In this manner Spielvogel also calculated 6(Et3NBHI2) = 39.4 ppm, while

the observed value in the present work was 6 = 24.3 ppm. The agreement in

this second case is not so good; however in most cases the agreement between

calculated and observed values for the 1B chemical shifts was very good,

typically with less than 1.0 ppm error.
1 34-38 34
3. H N Data.34-38 Coyle and Stone3 were the first to attempt to

correlate HII nmr shifts to donor-acceptor properties and adduct stabilities.

They examined a series of donor molecules (which contained -CH CH, groups)

coordinated to BI 3, BF3 and B(CH3)3, to see if any correlation existed between

the stabilities of the adducts formed (as determined by other means) and the

"internal shift" of the ethyl group in the ligand.* Although in all live of the cases

studied (Et N, EtMeN, Et2S, Et2 0, and EtMeO) no such correlation was found,

the idea, that a correlation might exist between donor-acceptor properties and

proton chemical shift, encouraged other research groups to examine other systems.

Holmes and Carter35 examined the reaction of a series of dimethylamino-

,aethylohosphines with triethylborane. The rela.nvF base strengths were determined

by a displacement study, and the proton rnmr dat.' vere examined for correlations.

It was found ihat tin' chain g in chemical sIjift of tbhe boreo-attached ethyl


SThe termC".nternai sMhif" ref:rrs t t'he st -fi.aion of i .u ', :-,opripons due
to the methylene and methyl protois of an tth;3 grc'c, and is taken as positive
when th, methyl f'Ihsorrtion is nt '", er fiHell,










group' correlated well with adduct stability:

order oi base strength as determined by displacetr.et L.BE6 ;ppm)*
B-EL
reactions

(strongest) (CH3)3P + 0.35 (farthest upfield)

> (CH3)3N + 0.31

S(CH3)2PN(CH3)2 + 0.29

> (CH3)3P[N(CH3)212 + 0.21

(weakest) > P[N(CH3)2 3 + 0. 11 (least downfield)

This correlation is explained in that coordination of B(Et)3 to a basic center

should increase the electron density on the ethyl groups attached to boron,

causing an upfield shift of 6 B-, and that the more basic the ligand attached to
B-it
the borne is, the greater that upfield shift should be.

Miller and Onyszchuk6 reported the It nrmr shifts of the (CI13)3; addicts

of boron trihalides and borane and showed that the chemical shifts correlated

with the previously determined8 order of acceptor power: BBr3 > BC13 > BF3 >

1/2(B I6). In a related study, Miller and Onyszchuk reported37 the 1H nmr
2 6
shifts of the CH CN adducts of boron trihalides and showed that these chemical
2,3
shifts also correlated with the order of acceptor power. 2, Ali these values

are shown below:

X in :BX, 6 of (CIIT 3N adducil 6 of CH CN adduct**
S33 3

B 159.5

56 it determined by tho iomei t imcthfc.l: referenced to cyclohexane.
B0 -6 -t . A positive value id cates an upfield shift.
B-Et addamt fr'' aecid
t Values in Hz, d3v.'oiield oro- in'e;nal TM?.I, it C1ICI, solirLt.
** ,'nlu"s in Hz, dl-onfield from" internal TMS. i, CH ":02 solt'i"n.









X in BX3 6 of (CH3) N adduct 5 of CH3CN adduct


F 157.3 161.8

Cl 180.2 182.8

Br 190.8 189.9
37
I 201.6

The idea in this correlation is that coordinating an acid to a base causes a net

removal of electron density from the donor, thus deshielding protons in alkyl

groups in the donor moiety. The greater the acceptor power of the acid, the

greater should be the downfield shift of the alkyl protons in the donor group.

Other H nmr data on donor-acceptor complexes are given in papers by
28 38 40
Heitsch, Hartman and Miller, and Derek, Clague,and Danti. Other papers

which discuss donor-acceptor properties in terms of 1H nmr shifts include

those by Young, McAchran, and Shore,41 Massey and Park,42 Cowley and Mills,43

and Clippard and Taylor.44

The correlation of relative acid strength of boron acids with changes in

the 1H nmr shift of protons in alkyl groups in the donor portion of donor-acceptor

complexes formed between the acids and a reference base is perhaps a difficult

idea to accept. The basis of the argument is that the effect (a downfield shift

when stronger acids are coordinated) is a deshielding due to the inductive with-

drawing power of the acid. This argument clearly applies in those cases where

there are only simple a-bonding interactions between donor and acceptor.

However, the effect of coordinating BX3 groups (X = F, Cl, Br, I) is opposite to

that expected on the basis of the trend in electronegativity of the halogen substituents

(that is, the most deshielding occurs when X = I, and the least when X = F). As









stated earlier, the effect has been attributed to the fact that the acid strength

increases in the order:

BF3< BC13 < BBr3 < BI3 ,

and thus that the inductive withdrawal effect of the BX3 group on coordinated

donor groups should be in the same order. To evaluate this argument let us

consider in detail the reason that the order of acid strength is BF3
BI3. One can distinguish three major factors:

1) The order of electronegativity of the halogens is F >> Cl > Br > I.

On the basis of electronegativity alone, one would thus predict an

order of BF3 >> BC13 > BBr3 > BI This order is opposite to

the order observed.

2) The ability of the halogens to IT-bond to boron (in the free acid)

increases in the order I < Br < Cl << F. If IT-bonding in the free

acid must be destroyed to form the adduct, the strongest acid would

be the one which has the least TT-bonding. Thus, on the basis of

extent of Ir-bonding, one would predict an order of BF << BC <

BBr3 < BI. This order is the same as the order observed.

3) The relative steric bulk of the halogens increases in the order

F < Cl < Br < I. Front strain in the adduct is a destabilizing

influence, and one would predict that the acid strength, if

determined by steric bulk alone, would be in the order BF >

BCl3 > BBr3 > BI3. This order is opposite to the order observed.

If the actual order of acid strength is determined by these three factors, one must

conclude that the predominant effect is that of factor 2), since only factor 2)









correctly predicts the order of acid strength. That is, the energy required to

break up ir-bonding in the free acid determines the trend in the relative acid

strength of the boron trihalides.

It should be obvious then that if all of the ir-bonding in the borane is

destroyed in forming the adduct, the inductive withdrawing power of the

coordinated borane would be determined by the other two factors, principally

by the electronegativity of the halogens. Thus, the inductive withdrawing power

should be in the order:

BF3 > BC3 > BBr3> BI3

To the extent that there is residual ir-bonding or "memory of the free acid" in

the acceptor portion of the adduct, this order may be modified somewhat, but it

is difficult to believe that there could be enough residual T-bonding in the adduct

to cause the order of inductive withdrawal power of the coordinated borane to be

the reverse of that predicted by electronegativity considerations.* One must

conclude therefore that some other factors besides inductive effect are at work

in determining the chemical shifts of these adducts. It is the purpose of this


It must be pointed out that this argument is not without conflicting
evidence. Arguments in favor of residual r-bonding in these donor-acceptor
molecules have become common in the recent literature, and certainly
experimental evidence exists supporting the idea of such residual iT-bonding
being a factor in these systems. The argument for residual w-bonding is best
put forward by Brown, Drago, and Bolles; additional experimental evidence is
presented by Swanson, Shriver, and Ibers,10 Clippard, Hanson, and Taylor,12
and Lanthier and Miller.16 Nevertheless, even if residual r-bonding does play a
role in the bonding of these adducts, one should expect the opposing effects to
cancel out in all the haloboranes except perhaps BF3. The fact that large changes
in chemical shift occur between BC13 and BBr3 and especially between BBr3 and
BI3 coordinated to a reference base would seem to clearly indicate that the
explanation of the trend in chemical shifts in terms of inductive withdrawal effects
is at least incomplete, if not altogether wrong.





16




work to present evidence which supports one such factor steric interaction

between the halogens on boron and protons on alkyl groups in bases coordinated

to boron.











CHAPTER II

MATERIALS AND PROCEDURES


Materials

Trimethylamine-borane, dimethylamine-borane, and triethylamine-borane

were obtained from Callery Chemical Company and were used without further

purification. Other chemicals and solvents were reagent grade and, except

where noted, were used without further purification, but were dried over

Molecular Sieve 3A where appropriate.

Apparatus and Methods

Infrared spectra were obtained using a Beckman IR-10 spectrophotometer.

Solid samples were prepared as KBr pellets; liquid samples were prepared as

liquid films between KBr plates.

Melting points were recorded on a Thomas-Hoover apparatus and were not

corrected.

Elemental analyses were performed by Peninsular Chem-Research, Inc.,

Gainesville, Florida.

Except where noted, proton nmr spectra of the amine-boranes and haloboranes

were recorded at ambient temperatures in CH Cl2 solution with tetramethylsilane

as an internal standard. Concentrations were typically in the range 0. 25 to 0. 65 M,

although some of the diethylamine-haloborane spectra were obtained on more

concentrated solutions. The spectra were obtained at 60 MHz, using a Varian

Model A60-A spectrometer. The chemical shifts are reported in Hz, downfield

from internal tetramethylsilane. In two cases, to be mentioned later, spectra









were also obtained at 100 1MHz, using a Varian Model XL-IJ0 spectrometer.

Calibration of the spectra of the diethyla mine-haloboranes was accomplished

by means of sidebands generated from the peak due to internal tetramethylsilane.

Computer Analysis of H NMR Spectra ef Diethyiamine-Haloboranes

The 1H n!mr spectra of the diethylanine-haloborate adducts were analyzed

by means of the program LAMP2, a version of LAME,45 written n for an IBM

360-65 computer with a Calcomp plotter, and kindly supplied by Dr. R.W. King

of the University of Florida Chemistry Department, LAME is a modification of

LAOCOON3 6,47 to include magnetic equivalent e ctoring, which allows a

group of magnetically equivalent nuclei to be treated as one nucleus of spin equal

to the total spin of the group, as regards interacti-. 'with he' r.ui-:e'i, but vhich

weights the energy level populations for th:-- a' oup according to the binominal

expansion. This modification greatly simplifies the calculation in cases where

such groups of magnetically equivalent nuclei exist.

The program LAMP2 is made up of three different subroutines: non-

iterative, iterative, and plot,

The non.-iterative subroutine has the capability of causing the cemnputcr to

generate, from an arbitrarily chosen set of chemical slhits and coupling constants

for a system of two to seven ro upls 0o *- 1.,-:i: '" eqiuiva een. nuclei of spin-1/2,

a tnble cf frequencies and intensities of the lines o:xp'tect ;ir the nnr spectrum.

'Ihe program uses the input data te fonr and solve th.? Hamiltonian rnatrix for

the stationary state energies of the Esystemr frillo'i'g !he pe, eduren .:Itlined by

Poi"le. Schi'.cider, anc' Berustein.48 TirP trcquencies oj; I.1- ::'d transitions

correspond to energy differences of the station rv s'toire: gring -ri-e tf' '-: rr. .









If a spectrum thus calculated bears a recognizable resemblance to an

observed spectrum, the iterative subroutine can cause the computer to perform

an iterative calculation by means of which the calculated frequencies of assigned

lines are brought as close as possible (by the least-squares criterion) to the

corresponding observed lines. The chemical shifts and coupling constants

which yield this best fit are then printed out along with information about the

expected error in them and along with a table of observed and calculated line

frequencies, calculated intensities, and errors in fitting the frequencies of

observed lines.

The procedure involved in the iterative subroutine is as follows. The

partial derivatives iv./ip. are found in the region of the input parameters,
1]
where 6v./6p. is the rate of change of the i-th line position with respect to the

j-th parameter. For small, finite changes, one assumes that Ai = (./6p.)Ap..

During each cycle of the iteration the program seeks to correct each parameter

by the amount required to produce agreement with experiment. This produces

equations of the form, (6v./6p.)Ap. = (Vbs -Valc)i for each selected line

position v.. In matrix form this group of equations becomes DA=N, where D is

the (ixj) matrix of partial derivatives, A is a row matrix of corrections to the j

parameters, and N is the column matrix of i line position errors. As long as

i >j, the equations are overdetermined,and if the initial guesses are good, the

program is strongly convergent. A best least-squares fit to observed line positions

is sought, rather than exact agreement.

It should be emphasized that iterative programs typically seek convergence

either on energy levels or on transition frequencies (i. e. differences between









energy levels). The program used in this work iterates to convergence on

transition frequencies.

The plot subroutine can cause the computer, using the Calcomp plotter, to

generate a calculated spectrum, based on Castellano's hybrid line shape

function,49 vhich includes a histogram showing exact line positions.

The spectra of these diethylamine-haloborane adducts were analyzed as

ar. ABCD3 spin system (A,B = non-equivalent methylene protons; C = nitrogen

proton; D3 = methyl protons) since the full eleven-proton system vas too large

for the program as currently dimensioned. Little error should be introduced

by this approximation, since any cross-coupling between the ethyl groups

should be exceedingly small, and since the nitrouen-attachedil proton is only

weakly coupled to all others (J/A6 does not exceed 0.1). The system

approximates to ABMXI3

'he 100 MIz spectra of the dibromo- and monochloroborane adducts of

diethylamine were obtained, and analysis of these, made simpler by the spreading

out of the patctrn at the higher field strength, allowed subsequent assignments

in the GO MHz spectra to be made with more certainty. The initial assignment

of parameters was made as follows. Values of 6 for the various protons were

esrimater' by visual approximation. The centers *f mass of the two halves of

the meticylene pattern were used as i'.itial estimates of 5 H(A) and 6C2 B).

The two CH CH3's were assumn: d t o be a pprox;ima tely equa 1, since, in every

case, the methyl resonances were 1:2:1 triplets; the value used .v::v the spacing

of the triplet. Values for J W-eC"cre assume Io be either zero, by visual

observation, or boutt 5Hz. by ana!cgyv to the limetpylar.ine-halobo-ane results.









Values for JAB were estimated visually on the basis of spacings in the methylene

pattern. Signs for the coupling constants were assumed to be positive for the

3-bond constants (J2 CH3 and J INCH2) and negative for the 2-bond constant
2 3 2
(JAB) by analogy to carbon systems. After one spectrum had been fully analyzed,

assignment of parameters in other spectra were made by a straightforward

comparison of patterns and/or by visual estimations.

The spectra of diethylamine-trichloro- and tribromoborane were poorly

resolved, probably due to the added feature of coupling to boron, and as a result

it was not possible to carry out full analyses on these two spectra. Approximate

values for the parameters were obtained by simulating the spectra. The

accuracy of these results is obviously somewhat lower than those for the six

mono- and dihaloborane adducts, which results \ ere obtained by rigorous

analysis.

No fine structure was observed in any of the NH resonances, probably on

account of quadrupolar broadening by coupling to nitrogen and/or boron.

Accordingly, approximate chemical shifts were assigned to these resonances, and

these shifts were not refined.

On the average,20 of 48 lines (and in no case le-s than 17) were assigned

in the methylene region of a spectrum and r.ere given equal weight. In each case

16 of 3: lines were assigned in the inethyl region. The root-mnepn-square error

oi observodv vs. calculated transition fiequencies did not exceed 0. %5 Hz and

averaged 0.16 fii. Go o agreementt was fo.n'd between observed 60 MHz spectra

and calcu:!'ate spLctra as plotted by the computer. These observed and

calculated spectra are show in the cxptriinental section.










CHAPTER III

SYNTHESES AND REACTIONS

The syntheses of amine-haloborane adducts examined in this study were

accomplished by following methods established in previous work in this laboratory,

which involved the preparation of mono-, di-, and trihaloborane adducts of

trimethylamine and pyridine via halogenation of the respective amine-borane4'25'50-52

In that work it was determined that pyridine-borane was more reactive than

trimethylamine-borane towards various halogenating agents. The reactions

of interest to the work discussed here are shown in the equations which follow

(TMA = (CH3)3N, py = C5H5N):

(1) TMABH3 + HX TMABH X + H2 (X = C, Br)*

(2) TMABH3 + 1/2X2 TMABH2X + 1/2H2 (X =I)

(3) TMABH3 + 3/2X2 TMABHX2 + 1/2H2 +HX (X= Br, I))

(4) TMABH3 + 5/2X2 TMABX + 1/2H2 +2HX (X= Cl, Br, I)**

(5) TMA + BX3 TMABX3 (X =F, Cl, Br, I)

(6) pyBH3 + HX pyBH2X + H2 (X = Cl, Br)

(7) pyBH3 + 1/2X2 4 pyBH2X + 1/2H2 (X =I)

(8) pyBH3 + 2HX pyBHX2 + 2H2 (X = Cl, Br)*

(9) pyBH3 + X2 pyBHX2 + H2 (X =I)

(10) pyBH3 + 2X2 -pyBX + H2 + HX (X=C1, Br)tt

This reaction proceeds no farther, even with excess HX, unless the
solution is heated to about 100 .
t Reflux is required for X=I.Reaction produces mixture of products forX=Cl.
** Reaction stoichiometry is assumed for X=C1. Reflux is required for X=I.
The reaction proceeds no farther, even with excess HX and heating.
it Reaction stoichiometry is assumed for X=C1.

22









In some preliminary experiments, it was established that triethylamine-

borane reacted in a fashion similar to trimethylamine-borane, and that dimethyl-

and diethylamine-borane reacted in a fashion similar to pyridine-borane. If, as

25, 52
has been suggested in some of the work previously cited,2 52 these reactions

are affected by steric requirements in the reacting molecule, it is reasonable

that tertiary amine-boranes should all react similarly and that secondary amine-

boranes should react more easily, hence the resemblance to pyridine-borane

which also presents less steric bulk.

Preparation of Triethvlamine-Haloboranes

Reaction of triethylamine-borane and chlorine. -- The reaction of

triethylamine-borane and chlorine was examined by monitoring the change in the

nmr spectrum of a -olution of (CH3CH2)3NBH3 in CH2C12 as a function of the

amount of C12 added. The original pattern for (CH3Ci2 )3N BH3 ( a triplet in the

methyl region and a quartet in the methylene region) was observed to diminish

in intensity as a new triplet and quartet grew downfield from the original triplet

and quartet. This new set of peaks grew to a maximum then began to diminish

as a third set of peaks (again a triplet and a quartet) grew downfield from the

second set. As this third set of peaks grew to a maximum and began to

diminish, a fourth set of peaks (a triplet and a complex multiple) emerged

donfi&iEd from the third set. This fourth set grev t.- a maximum as the other

three sets disappeared, and did not diminish on further addition of C1. These

hbservat'oni ere explained on the basis of Ftfevoise chlorination of

(CHI '; NIT producing successeivel (C Cli , .C Cl, (CH3CH )3NBIHCl2,

ond finally (CH CINi.).BC1 The reaction was such as to prxuce a ternary
3 21 f 3'









system at most points in the reaction; that is, before all the starting material

had been monochlorinated, some of the product monochloroborane adduct had

been dichlorinated to give dichloroborane adduct. The nmr spectrum observed

for (CH3CH2)3NBC13 is the same as that reported by Miller and Onyszchuk,36

who prepared (CH3CH2) NBC13 by direct combination of (CH CH ) N and BCI3

The spectrum reported here gave better resolution of lines so that the tentative

assignment of Miller and Onyszchuk is confirmed. The complex multiple in the

methylene region of the spectrum of CH 3CH2)3NBC13 was produced by coupling

of the CH2 protons to boron as well as to the CH3 protons. The pattern (Figure

1) actually appeared as a 13-peak multiple because of overlap of some of the

peaks in the expected 16-peak pattern.

The chemical shifts and coupling constants for these adducts are given in

Table III.

Reaction of triethylamine-borane and bromine. -- The reaction of

triethylamine-borane and bromine was also examined by monitoring the change

in the nmr spectrum of a solution-of (CH3CH2 ) 3NBH3 in CH2C12 as a function of

the amount of Br2 added. As in the reaction of (CH3CH2)3NBHf3 and C12, the

peaks due to starting material were replaced by two sets of peaks (each set

( .--..'. of a triplet and a quartet) downfield from the original set, which

successively increased and then decreased in intensity as more bromine was

added. These two sets of triplets and quartets were due to (CH3CH2)3NBII 2Br

and (CH3CH2)3NBHBr2 respectively. A triplet and a complex multipiet appeared

downfield of the triplet and quartet of (CH3CH2)3 NBHBr2 as more Br2 was added,

and this set of peaks grew to a maxiunumi and did not diminish on further addition






















ET3NBCI3


20 Hz
__~---- --- ---- -I


Figure 1. Methylene region of the 11 namr
spectrum of (CI 3C2 ) NBC13.









of Br2. This fourth set of peaks was due to (CIIHCH2)3NBBr3. The multiple

in the CH2 region (Figure 2) showed 15 peaks of the expected 16-peak pattern,

one peak being masked by a 13C satellite of the solvent, CH C12l

The chemical shifts and coupling constants for these adducts are given

in Table III.

The 11B nmr spectrum of (CH3CH2)3NBHBr2 showed a doublet (JBH

161 Hz) centered at 20. 3 ppm upfield from trimethylborate.

Reaction of triethylamine-borane and iodine. -- The reaction of triethylamine-

borane and iodine was likewise examined by monitoring the change in the mnr

spectrum of a solution of (CH3CH2 ) 3NBH in CH Cl2 as a function of the

amount of 12 added. Triethylamine-borane (2.22 g, 19.3 minol) was dissolved

in 25 ml of CH2Cl2. Solid iodine (2.44 g, 9. 62 mmol) was added in portions

with stirring. The nmr spectrum of this solution showed a triplet in the methyl

region and a quartet in the methylene region, each downfield from the triplet and

quartet of the starting material. This pattern is due, therefore, to (CH3CH 2)N-

BH2I.

To the solution of (CII3CH2)3NBH2I in CII2Cl2 prepared above, more

iodine (3. 21 g, 12. 6 mmol) was added with stirring and reflux. The omr

spectrum of this solution, takcn several times over a period of about 20 hours,

showed the tri;l.et ard quartet due to (CH3CI2)3 3NBT2 slowly disappearing,

while a new triplet and quartet appear,:' diownfiell. After 22 hours of reflux,

the new triplet and quartet were tle only peaks in the specirium other than

solvent peaks. These peaks thus correspond to (CH sCH2 ) NBIIHI,




























ET3NBBr
35 3


20 Hz


figure 2. ilAdIylene region of the IH inr
spectrum of (CI3CFIl), NBBr3
3 3









11
The B nmr spectrum of (CH3CI2 )3NBHI2 shov.ed a doublet (JB
3 .10 2 BH
155 Hz) centered at 42. 5 ppm upficld from trimethylborate.

To the solution of (CH3CH2)3NBHI2 in CH2C12 prepared above, more

iodine (4.89 g, 19.3 mmol) was added with stirring and reflux. The nmr

spectrum of this solution still showed only the triplet and quartet due to

(CH3CH2)3NBHI2 after a short (1/2 hour) period of reflux, but the spectrum

taken after about 4 hours of reflux showed a new set of peaks, the peaks for

(CH3CH2)3NBHI2 having almost disappeared. This new set of peaks was a

triplet ir the methyl region and a doublet of quartets in the methylene region.

On refiuLing for an additional 24 hours, the new pattern remained unchanged.

This pattern is not consistent with (CH CH2) 3NI3, but is consistent with
+
(CII3CHI)i3NH

Accordingly, a sample of (CHCH2) 3NH I was prepared and its nmr

spectrum in CII2Cl2 obtained. To this solution, some iodine was added, so as
+-
to give (CII3CII2)3NH I1 and the nmr spectrum again obtained. Both spectra

wore very similar in pattern and position to the spectrum obtained above,

namely, a doublet of quartets in the methylene region and a triplet in the methyl
F- -,
region. As a final check, the two solutions [(CH 3CH2)3NH 13 and the unknown

solution above] were mixed snd the nmr spectrum taken, The spectrum showed

coly oe. set of peaks, the pattern being unchanged irom the spectra of the

solutions taken separately. Therefore, the product of the exhaustive iodination

of (CH CiI :1NBI3 is not (CH3CH .3 NBT b su is instead (CII3CH2)3NHI A
s L' 23 A 3
where A =1, (Ir some other anion).

The chemical shilts and couplmns constants for these adducts are given in

Thble III.










Reaction of triethylamine and boron trifluoride-etherate. -- A solution of

triethylamine in ether was added dropwise to an ether solution of boron

trifluoride-etherate. Evaporation of the volatiles left low-melting, colorless

crystals of triethylamine-trifluoroborane. On standing in air these crystals

slowly decomposed to a brown oil. An nmr spectrum of this material in CH 2C1

solution showed a triplet in the methyl region and a broadened quartet in the

methylene region. The chemical shifts and coupling constants are given in

Table III.

Preparation of Dimethylamine-Haloboranes

Synthesis of dimethylamine-mono- and dichloroborane. -- (CH3)2NHBH2Cl

and (CH3)2NHBHC12 were prepared by V.R. Miller53 by the reaction of HC1 and

(CH3)2NHBH3. These procedures were confirmed in this present work, and the

nmr data for these two compounds, quoted from Miller's work,are given in

Table II.

Synthesis of dimethylamine-trichloroborane.54-- (CH 3)2NHBH3 (1.90 g,

32.2 mmol) was dissolved in 25 ml of CH2Cl2. Gaseous Cl2 was bubbled into

this solution at a moderate rate, with stirring, until the solution turned yellow.

At this time a white solid was present, and the flask was warm from the reaction.

All the volatiles were stripped off by pumping on a rotary evaporator, leaving a

light yellow solid. The product was stirred with 25 ml of fresh CH Cl2 and then

20 ml petroleum ether was added to precipitate the portion of the product that

was in solution. The white solid thus obtained was filtered, washed with two

5 ml portions of petroleum ether, and dried by pumping. The yield was 4.56 g

(28.1 mmol; 87.0%, based on (CH3)2NHBH3), mp = 125-7 Anal. Calcd. for

C2H7NBCl3: C, 14. 81;H, 4.35; N, 8.63. Found: C, 14. 87; H, 4. 52; N, 8. 27.









The infrared spectrum of (CH 3)2 NHBC13 (Figure 3) is characterized by
-1
a sharp band at 3180 cm assigned to N-H stretch, and a broad band at

700-780 cm-1, assigned to coordinated BC13 vibrations.55 The sharp band at

900 cm-, which Gerrard5 assigned to BCI2 is probably due to C-N stretch.

The entire infrared spectrum is given in Table I.

The proton nmr spectrum (Figure 4) showed a 1:1:2:2:1:1 sextet in the

aliphatic region. This pattern arises from coupling of the methyl protons with

both the nitrogen-attached proton and boron, giving a doublet of quartets; the

coupling constants are such as to cause overlapping of two sets of two peaks.

The chemical shift and coupling constants are given in Table II.

The 11B nmr spectrum showed a single peak at 9.4 ppm upfield from

trimethylborate.

Reaction of dimethylamine-borane and bromine.-- Dimethylamine-borane

(0. 651 g, 11.0 mmol) was dissolved in 30 ml of CH C 2. Bromine (0. 885 g,

5.54 mmol),dissolved in 5 ml of CH C12, was added dropwise with stirring.

The nmr spectrum of this solution consisted of a strong doublet downfield from

the doublet due to starting material. There was a small amount of starting

material left, since the doublet due to starting material had not entirely disappeared.

There was also evidence for a third product since a small peak was observed

downfield from the strong doublet. These observations are consistent with most

of the (CH3)2NHBH3 having been converted to (CH3)2NHBH2Br, plus a little

(CH3)2NHBHBr2'

To the solution of (CH3)2NHBH2Br in CH CI2 prepared above, more bromine

(0.914 g, 5.72 mmol), dissolved in 5 ml of CH2CI2, was added slowly with stirring.














5SA'LNGI 6i MICR7 7 O
55 6:. 7 75) 8


'O I. 12 14 t d 16 4 1 3 40


00 1200

WAVENUM4tR CM'


Figure 3. In;rarei spectrum of (CH3 ) NIIBCI.
32 2


35 4 ~ di 5
























ME2NHBCI3


10 Hz


figure 4. H nmr spectrum of (CI3 )2NHBC
~32









The nmr spectrum of this solution showed the doublet assigned to (CH3)2NHBH2Br

to have nearly disappeared and the doublet assigned to (CH3)2NHBHBr2 to have

become very strong. There was a shoulder on the downfield side of the

(CH3)2NHBHBr2 doublet which was assigned to a small amount of (CH3)2NHBBr3.

To the solution of (CH3)2NHBHBr2 in CH2C12 prepared above, more

bromine (0.986 g, 6.17mmol),dissolved in CH2Cl2, was added slowly with

stirring. The nmr spectrum of this solution showed the doublet previously

assigned to (CH3)2NHBHBr2, reduced somewhat in intensity, along with a

complex pattern of similar intensity at the position of the shoulder previously

assigned to (CH3)2NHBBr3. More bromine was added to this solution until a

permanent color was produced. The nmr spectrum of this solution showed

only the complex pattern assigned to (CH3)2NHBBr3. Chemical shifts and

coupling constants for these adducts are given in Table II.
54
Synthesis of dimethylamine-tribromoborane. -- (CH3)2NHBH3 (2.95 g,

50.0 mmol) was dissolved in 40 ml of dry CH2C12. Br2 (16.05 g, 100.4 mmol)

was dissolved in 40 ml of dry CH2Cl2. The Br2 solution was added dropwise,

with vigorous stirring, to the (CH3)2NBH3 solution. As the last of the Br2

solution was added, a white precipitate formed, and the solution, which was

previously colorless, turned slightly yellow. All the volatiles were stripped off

by pumping on a rotary evaporator. The white solid which remained was taken,

still in the flask, into a drybox where the product was transferred to a tared

bottle. The yield was 13. 0 g (44. 0 mmol; 88. 0%, based on (CH3)2NHBH3),

mp= 150.5-152 Anal. Calcd. for C H NBBr3: C, 8.13; H, 2.39; N, 4.74;

Br, 81.09. Found: C, 8.14;H, 2.36;N, 4.64; Br, 81.21.









The infrared spectrum of (CII ) NHBBr3 (Figure 5) showed a strong,
--1
sharp band at 3150 cm assigned to N-H stretch, and a broad band at 650-,
-1 55
700 cm assigned to coordinated BBr vibrations5. The entire infrared
3
spectrum is given in Table I.

The proton nmr spectrum (Figure 6) showed a multiple in the aliphatic

region, which resembles a sextet of 1:1:2:2:1:1 intensity, but which shows eight

peaks under high resolution. The pattern is a doublet of quartets due to coupling

of the methyl protons with both the nitrogen-attached proton and boron. The

chemical shift and coupling constants are given in Table II.

The 11B nmr spectrum showed a single peak at 24.8 ppm upfield from

trimethylborate.

Reaction of dimethylaniine-borane and iodine. -- To dimethylamine-borane

(0. 534 g, 9.06 mmol),dissolved in 20 ml of CII2Ci2,was added solid iodine (1.14 g,

4.49 mmol) piecewise with stirring. The nmr spectrum of the resulting solution

showed only a doublet downfield from the position of the double due to starting

material. This doublet was thus assigned to (CH3)2NHBH2I.

The 1 nmr spectrum of (CH3)2NHBIH2I showed a triplet (JBH = 143 Hz)

centered at 3i.0 ppm upfield from trimethylborate.

To dintahyvainine-boran (1.05 g, 17.8 i mmol) in 25 ml CH2Cl2 was added

icdine (4. 2 g, 1T. E- n.ol), piecewise vhilt stli,'rlg. The irrr spectrum of the

-resulting sotuti-'. taken 45 mimlites after a l the luid been added, showed a

strong do:'.let down ied -oi om F.b' position" of the doubl t i p cvrously assigned to

(Cil .,. I; T1I. I doiblet was nss ignl.d' t (Cl),,;:!l':IB i. Th cre was in t.he

sp~Ociti, a i1 ,1 peakl correspor.ing to tne righlt-hand peak of the doublet due to












L4 2 T 0 -- i










. .... ^^ ^- ^ T-r- --I-,- .. ~..

L.' .1 A... i_ .... _'.0_L I_ L


S5 6


77W


WAVEirNGH iN UMCRONS


6 7 7 5 8 P 10 11 12 14 16 16 1 2 J


.I i I ,








___1 1 / 1 7 '
I l-_ i_ 4- ---- -^i-
- L_|' ]l-l -__ -r- -- .




Wl nil
T~r^^Tn&'i'Trl1,---- ^---i-i-r"-' --
1600 1400 1200 100o S03 60 0 500 ) ,S J o00
WA*VENUMst CM'


"'i-;ue 5.Infrared speci.trui of (CH 3)2NHiBr3 .
























ME2NHBBr3


10 Hz


1
llgure 6. 1 nnmr srectc.i_ of (CT 2IJ 3Br.
3)2 R3r*


V- *.,'









(CH3)2NHBH2I. The left-hand peak would be masked by the doublet assigned

to (CH3)2NHBHI The spectrum taken after 5 hours showed only the

doublet assigned to (CH3)2NHBHI2'

The chemical shifts and coupling constants for these adducts are given in

Table II.

The 1B nmr spectrum of (CH3)2NHBHI2 showed a doublet (JBH = 158 Hz)

centered at 43.5 ppm upfield from trimethylborate.

Synthesis of dimethylamine-triiodoborane. -- (CH3)2NHBH3 (2.5 g, 42 mmol)

was dissolved in 40 ml of dry CH Cl2 and 12 was added piece by piece with

stirring until a permanent color was given to the solution (about 21 mmol of

12 was required). A large excess of 12 was then added (at least 100 mmol), and

the solution was refluxed for about 40 hours. After refluxing, the volatiles were

removed on a rotary evaporator. A black, fuming solid remained in the flask.

Fresh CH2C12 (40 ml) was added to this solid. To this solution about 3 ml of

mercury was added,and the mixture was vigorously stirred for 15 minutes. At

this time the solution was decolorized, and solid HgI2 was observed in the flask.

The contents of the flask were filtered* Hexane was added to the filtrate,

producing a light tan precipitate. This solid was filtered off, washed with two

5 ml portions of hexane, and pumped dry. The yield was 6.0 g (14 mmol; 33%,

based on (CH3)2NHBH3), mp = 157-158 (some evidence of decomposition at

83-85 ). Anal. Calcd. for C H NBI3: C, 5.50; H, 1.62; N, 3.21. Found:

C, 5.62; H, 1.64; N, 3.04.



Dimethylamine-trihaloboranes have a low solubility in CH2C12; therefore,
this is a yield-reducing step.









The infrared spectrum of (CH3)2NHBI3 (Figure 7) is characterized by a

broad band at 3300-2950 cm-1 (with peaks at 3210 and 3110 cm-1) due to N-H

stretch, and a doublet at 615 and 600 cm-1, probably due to B-I stretch.55

The entire infrared spectrum is given in Table I.

The proton nmr spectrum (Figure 8) showed a multiple in the aliphatic

region, which resembles a quintet of 1:2:2:2:1 intensity, but which shows eight

peaks under high resolution. The pattern is a doublet of quartets due to coupling

of the methyl protons with both the nitrogen-attached proton and boron. The

chemical shift and coupling constants are given in Table II.

Synthesis of dimethylamine-trifluoroborane. -- [This is the only dimethyl-

amine-trihaloborane which can be prepared pure by direct addition of amine and

borane. Two different methods exist for the preparation of dimethylamine-

trifluoroborane. The adduct may be prepared by direct addition of dimethylamine

to boron trifluoride57 (in the gas phase or in an inert solvent), or the adduct

may be prepared via a displacement reaction, typically the reaction of dimethyl-

amine with boron trifluoride-etherate. 58

Both methods were used in this study, and the best results were obtained

when the two gases were bubbled together in an inert solvent (CH 2C2). Care

should be taken to keep BF3 in slight excess. The product of this reaction,

isolated by pumping off the volatiles on a rotary evaporator, was a white

crystalline solid, mp = 40-44. Anal. Calcd. for C2H7NBF3: C, 21.28; H, 6.25;

N, 12.41. Found: C, 20.90;H, 5.70;N, 12.16.

The infrared spectrum of (CH3)2NHBF3 (Figure 9) was broad and poorly

resolved, probably because of the low melting point of the crystals. It is














VNV5ELNCT LtN A ;COMS0
55 S '5 3 65 / 75 8


'_0 11 12 1 16 .6 Is 7 5 2 0 40

-LJ- IillRLi! _.-.l JL


I -



: 2


Jo 1700

WAVENUMBER CM'


Figure 7. Infrared spectrum of (CH3)2NHBI3.





















ME2NHBI3


__________O__z__ -I


Figure 8. 1H nmr spectrum of (CH )2NHBi3











'NAVENGTH IN MHi(G
55 6 45 75 8 I 10 II '7 1 6 1 R4 i ; J31 aI





S, --



I I I
-I .. ...... ..












1- 00 1600 1 .00 100 .1000 Soo 0.o 6 : 0
WAVNM t CM


Figure 0. Infrared spectrum of (CH3)2NI1BF3.









-1
characterized by a band at 3260 cm assigned to N-H stretch, and a set of
-1
three broad, ill-defined bands at 1210-1150, 1120-1040, and 980-900 cm-1

probably due respectively to B-F stretch, CH3 wag, and a combination of B-F

stretch and C-N stretch.55 The sharp peak at 705 cm-1 is probably due to B-N

stretch.40 The entire infrared spectrum is given in Table I.

The proton nmr spectrum (Figure 10) showed a 1:1:1:2:1:1:1 septet in the

aliphatic region. This pattern arises from coupling of the methyl protons to

both the nitrogen-attached proton and boron;40 the coupling constants are such

as to cause the overlapping of two of the expected 8 peaks. The chemical shift

and coupling constants are given in Table II.

Reaction of Dimethylamine and Boron Trihalides

Reaction of dimethylamine and boron trichloride. -- Boron trichloride

(1. 50 ml, 18. 3 mmol) was dissolved in 50 ml of dry benzene on a vacuum line (the

pressure of the solution was 120 mm). Dimethylamine (1.20 ml, 18.3 mmol) was

trapped out on another part of the line and allowed to warm up. When the pressure

of dimethylamine reached 500 mm, the stopcock to the BC1 -benzene solution

was opened, allowing the dimethylamine to enter at such a rate as to hold the

pressure of dimethylamine in the external system steady at 500 mm; a white

solid precipitated during this procedure. The addition was facilitated by stirring.

After all the dimethylamine had been transferred into the BCI -benzene solution,

the volatiles were removed by evacuation, leaving a white solid in the flask.

After transfer to a tared vial, the product weighed 2.55 g (86% yield).

A small amount (about 0.2 g) of this product was stirred with 20 ml of

CH2C12 for three hours. The solution was filtered, giving a white solid (A). The























ME2NHBF3


10 Hz
i------------ -- -?


Figure 10. 1H nmr spectrum of (CIi3)9NIHBF,.









solvent was removed from the filtrate on a rotary evaporator,leaving a second

white solid (B). Each fraction consisted of about 0.1 g of material.

An nmr spectrum was taken in CH Cl2 solution of the CH2 Cl-soluble

material (B), which showed a seven-peak multiple in the aliphatic region.

This multiple was the result of an overlapping of two resonances: 1) a 1:1:2:2:1:1

sextet, centered at 172 Hz, due to the 1:1 adduct, (CH3)2NHBC13, and

2) a 1:2:1 triplet, centered at 167. 5 Hz, due to dimethylammonium ion,

(CH3)2NH2 (JHNCH = 5.7 Hz). The relative proportions of these two species

was determined (by estimating visually) to be about 5:1, the adduct in excess.

The infrared spectrum of this fraction (B) was similar to that of authentic

(CH3)2NHBC13, but there were some small differences. A small peak was

observed at 650 cm1, which was probably due to BC14 ion. 5A broad band
-1 -1
at 3050-2920 cm two sharp peaks at 2680 and 2600 cm ,and two sharp
-1 The
peaks at 2680 and 2600 cm were probably caused by [(CH3)2NHI2BC12 The

absence of strong peaks at 2440 and 1025 cm-1 indicates the absence of
+
(CH3)2NH2 in fraction (B). The ammonium ion observed in the nmr spectrum

of fraction (B) must then have come from the hydrolysis of [(CH3)2NH2 BC12

Since unhydrolyzed [(CH3)2NH]2BC12 ion would not have been distinguishable

in the nmr (6 = 167 Hz in CH2 Cl2, about 15 Hz broad), one cannot judge the

relative proportions of adduct and boronium salt from the ratio of adduct to

ammonium ion as determined by nmr integration. However, the relative

intensities of the peaks in the infrared spectrum imply a ratio on the order of

3:1. [That is, the fraction is 60% adduct and 40% boronium salt, by weight. ]










The infrared spectrum of the CH2Cl2 insoluble material (A) (Figure 11)

differed from that of authentic (CH3)2NHBC13 in several ways. There was a
-1 59
peak of medium intensity at 650 cm probably due to BC14 ion, 5and
-1 +
another peak of medium intensity at 540 cm probably due to amine2BC1 .
-1
The pattern of peaks between 700-850 cm- was markedly different from the

pattern in the spectrum of (CH3)2NHBC13, probably reflecting both the absence

of coordinated BC13 groups and the presence of amine2BC12 groups. There
-1
was a marked difference in the 2500-3300 cm region also. The spectrum

showed two strong,broad bands centered at 3200 cm-1 and 2960 cm-1, and

two sharp bands at 2780 cm and 2680 cm .. These bands are probably due
+
to the amine vibrations (N-H and C-H stretch) in amine2BC1 The nmr

spectrum of this fraction(A) in CD CN showed a broad, featureless absorption

in the aliphatic region (157 Hz). [Exposure to atmosphere during sample

preparation resulted in the appearance of a 1:2:1 triplet in the spectrum
+
(6 = 160 Hz, J = 5. 5 Hz). It would be expected that [(CH3)2NH]2BC12 ion should

be quite sensitive to hydrolysis.]

As conclusive proof of these assignments, an authentic sample of

[(CH3)2NH]2BC12 BC14 was prepared and compared with this fraction (A).

Synthesis of bis-(dimethylamine)-boronium chloride. -- [(CH3)2NH]2BH2 C1

was prepared in a manner similar to that described by Miller and Ryschkewitsch60
+-
for the preparation of [(CH3)2NH]2BH2 I (CH3)2NHBH3 (1.18 g, 20. 0 mmol)

was dissolved in 30 ml of C6H6 and a 0.40 N solution of HCl in C6H 6(65 ml,

26 mmol) was added slowly with stirring. The volatiles were removed by












SA-VEIfNGI, IN MCWONS
S3 1 15 3 33 6 5










7i i i r















3000 000 6010
. A WAV[NME_ C'--



41 I -__..





W*VYLNLMBOO COC


Figure 1,. Infrared spectrum of fraction (A) from the reaction of (CH )2NH and BCl3.
*32&


10 1. 12 K 1 o iS ;& 1, 16 0









pumping, and 30 ml of fresh benzene was added to the white solid. A 1. 71-N

solution of (CH3)2NH in C6H6 (12. 0 ml, 20. 5 mmol) was added with stirring. The

volatiles were removed by pumping, and the white solid which remained was

transferred to a tared vial. Yield of [(CH3)2NH12BH2 Cl was 2. 62 g (94. 7%,

based on (CH3 )NHBH ), mp= 161-3 The infrared (Figure 12) and nmr spectra

were nearly identical to those observed by Miller and Rys chkewitsch60 for
+-
[(CH3)2NH12BH2 I Anal. Calcd. for C4 H N BC1: C, 34.70; H, 11.65;

N, 20.23. Found: C, 33.50; H, 11.70;N, 19.73.

Synthesis of bis-(dimethylamine)-dichloroboronium chloride. -- [(CH3)2NH]2-
+ -
BH2 CI (0.47 g, 3.40 mmol) was dissolved in 30 ml of CH2C12,and Cl2 was

bubbled in at a moderate rate for 1-1/2 hours. During the addition of C12, a

white solid precipitated out and the solution turned yellow. The volatiles were

removed by pumping, and the white solid which remained was transferred (in

the dry box) to a tared flask. The yield of [(CH3)2NH]2BC12 Cl was 0.70 g

(98%, based on [(CH3)2NH]2BH2 C), mp = 123.5-124 Anal. Calcd. for

C4 H 14N2BC13: C, 23.17; H, 6.81; N, 13.51. Found: C, 22.72; H, 7.35;
+ -
N, 12.06. The nmr spectrum of [(CH )2NH] BC12 C1 in CH3CN was a broad,

featureless resonance centered at 155 Hz. The infrared spectrum (Figure 13)
-1
was very similar to the spectrum of fraction (A) in the region 900-4000 cm ,
-1
but there were several differences in the region below 900 cm notably the

absence of the peak at 650 cm and a lower intensity of the pattern at 750-850 cm1

Reaction of bis-(dimethylamine)-dichloroboronium chloride and boron
+ -
trichloride.-- [(CH3)2NH]2BCl2 C1 (0.20 g, 1.0 mmol) was taken in a flask,

attached to a vacuum line,and evacuated. BC13 (2. 6 mmol) was introduced into













2 E 5 5 6 75 7 5


S t2 14 611 1 8 ;0 37 40
"I .r ij] i i n n n 1 l- IT -T !i


100 1200
WAVYNUMbER CM'


Figure 12. Infrared spectrum of [(0C1H,) NH11 B- Cl .
Figure 12. Infrared spectrum of [(Cl, ,) NHIT.I~ 1 C+ l














0 11 12 14 16 16 ~B :' 30

'I J- J ] [ ,11


100 120S
WAVLNUMiRg CM'


Fi:gace 13. Infrared spectrum of [(CH ,NI. BC12 C .








the flask very slowly so that no heat buildup occurred. After the solid was

exposed to all of the BCl3, a large excess (about 20 mmol) of BC13 was

introduced into the flask, and after one hour's exposure, the volatiles were

pumped off. The white solid which remained was transferred to a tared vial.

The weight of the solid was 0.275 g, corresponding to 75% conversion of

Cl to BC4 (Since there was some material loss during transfer, the actual

percentage may be as high as 80%,) Thc mpwas 98-98.5:.The infrared spectrum

of this solid (Figure 14) was identical to that of fraction (A). The new peaks

in the spectrum, which are thus due to BCI are at 650(w), 735(m), and

780(s) cm"1

In summary, the direct reaction of dimethylamine and boron trichloride

produces two major products as shoon in the following equation:



6 (CH3 )NHBC13

20 (CH3)2NH + 20 BC13- +

7 [(CH3)2NI BCl Cl (or Cl)


-, .- . -., i i.r,, *r, . r.,ht. ., . -- Boron tribromide

(3. !0 ml, 3,. 0 mmol) was dissolved in 50 ml of dry benzene on the vacuum line

(the 'ress.u re o( the solution was about 1)rl rn.m), lnmethylan.in (2.37 ml,

36. 0 mmoi) v:is trapped out in another part of the line and allowed to warm up.

iWhen the pressure of dimethylamine r-acl.ed 500 mmn, the stopcock to the BBr3-

benz:i;o c;-'lri,- was ov p(, '-d, allowing the diJliet_!'%.y i 1 ), o ;i.ee u( such a

ratc as i.'- ',. thp pressure of dimebh, in"ia'd in 'de cxteyra: system steady at











s 5 1'S 5 75 2 14 '6 '6 25 4
I-












-4.-








S 300 2*0 0 Ii 0 1600 H 1 10X 0 800 1000o500 'o 4 CC
I I iWVENUMI CMI -
I": I I I il, -. .i-






., _ __ -_: _L I

-$\ : 4 ~ --,. . .
,---\ / I . .. i: -! --- --= .. . . ./ --- .. ..!. -: i---, ...... I i- .... .. .. .






WAVENUUOLI'. CMI


Fi'iure 14. Infrared spcc'rum of [(CHi3) NH2 BCl BCl2 -
j 2 2 2 '









500 mm; a white solid precipitated during this procedure. The addition was

facilitated by stirring. After all the dimethylamine had been added, a sample

of the reaction solution was taken for nmr. The spectrum showed a pattern in

the aliphatic region (about 125 Hz) which appeared to be the 8-peak multiple

due to (CH3)2NHBBr3 overlapped by a singlet possibly due to (CH3)2NBBr2

(or some other aminoborane). The volatiles were distilled out into a liquid

nitrogen trap, leaving a white solid in the flask. [An nmr spectrum of the

liquid which was trapped out showed only benzene absorptions, but the liquid

fumed, implying the presence of BBr3.] After transfer to a tared vial, the

product weighed 9.37 g (88% yield).

A small amount (1. 0 g) of this product was stirred with 40 ml of CH 2C2

for ten hours. The solution was filtered, giving a white solid (A). The solvent

was removed from the filtrate on a rotary evaporator, leaving a gray powdery

solid (B). Each fraction was about half of the original material.

An nmr spectrum was taken in CH2Cl2 solution of the CH C12-soluble

fraction (B), which showed a complex pattern in the aliphatic region. This

multiple was the result of the overlapping of three sets of resonance: 1) an

8-peak multiple, centered at 180 Hz, due to (CH3)2NHBBr3, 2) a 1:2:1 triplet,
+
centered at 171 Hz due to (CH3)2NH2 (JHNCH = 5.6 Hz), and 3) a singlet at

181 Hz due to (CH3)2NBBr2. The ratio of intensities of these three resonances

(determined by estimating visually) is roughly 8:1:0. 3 [(CH3)2NHBBr3:

(CH3)2NH2 : (CH3)2NBBr2]. The infrared spectrum of this fraction (B)

was identical to that of authentic (CH )2NHBBr3.









The infrared spectrum of the CH2Cl2 insoluble material (A) (Figure 15)

was quite different from that of (CH3)2NNBBr 3. In fact, above 900 cm-1 the

spectrum of fraction (A) is very similar to that of [(CH3)2NH] BC1 Cl ,
+
implying that the material is a salt of [(CHI) 2NH 2BBr The region below

900 cm-1 shows a series of bands at 650-820 cm-1 (maxima at 665, 685, 720,
-1 +
785, and 815 cm ) due to amine BBr2 vibrations. In addition, there are

two bands at 450 and 470 cm and a band at 550 cm due to BBr2 and BBr4

respectively.61 [The two bands at 450 and 470 cm- are related to two similar
-1 + -
bands at 505 and 540 cm in the spectrum of [(CH3 ) NH2 BCI2 Cl An nmr

spectrum of this fraction (A) was taken in CH CN, which showed two

resonances in the aliphatic region: 1) a broad featureless resonance (about

1 Hz wide) at 171 Hz, due to [(CIH )2NHJ2BBr2 and 2) a 1:2:1 triplet at
+
159. 5 Hz due to (CH.)2NH2 (J = 5.6 Hz). However, the absence of bands in
4-
the infrared spectrum characteristic of (CH3 ) 2NH2 shows that the ammonium

ion comes from hydrolysis of [(C3) 2NH] BBr .

In summary, the direct reaction of dimethylamine and boron tribroinide

produces two major products and two minor products as shown in the following

equation:


2 (C1 )N.HBBr3

1(CH '.n + 4BBr3 1[(CIf3 )NHI 2BBr2 BBr (or Br )
+
<- 1 [(C I 2NH BB r + (C- 2N B r
I [(Cl!3)2 NH BB',4 + (CjI.12NJ3BrI


























' i f I I I













Figure 15. Infrared specter


5VAVEiNGH IN MICRONS
S~ 6 f65 7 7S &


1 4 5 5 5


WAVENULMBEl CM'



um of fraction (A) from the reaction of (CH ) NH and BBr3'
32&


Ibhu 'C0 140j 1200


SI 6 16 I la 33 0
i' i






I






.-.-L .-i- I JL .A






. .i aco ,sc jo ~ o -.. o









Preparation of Diethylamine-Haloboranes

Synthesis of diethylamine-borane. -- Diethylamine-borane was prepared

via a transamination reaction of diethylamine with trimethylamine-borane.

Trimethylamine-borane (21.4 g) was dissolved in a large excess of

diethylamine (60 ml) and refluxed for about three hours. The excess

diethylamine was pumped off, and the remaining diethylamine-borane, an oil,

was purified by shaking with water and extracting with methylene chloride,

the methylene chloride being evaporated to give a good yield (21. 6 g, 85%) of

liquid diethylamine-borane. Infrared and 1H nmr spectra showed the compound

to be free of impurities. The nmr spectrum showed a triplet in the methyl

region, a quintet in the methylene region (Figure 16),and a broad resonance

farther downfield, due to N-H. The chemical shifts and coupling constants

are given in Table IV. The infrared spectrum (Figure 17) was characterized

by a strong peak at about 3200 cm- due to N-H stretch, and a strong doublet

centered at 2350 cm1, due to coordinated BH3 stretch.

The 1B nmr spectrum of (CH3CH2)2NHBH3 showed a quartet (JBH

97.5 Hz) centered at 34.6 ppm upfield from trimethylborate.

Synthesis of diethylamine-monochloroborane. -- A sample of diethylamine-

borane (about 1.0 g) was dissolved in 20 ml of CH2Cl2, and a solution of HC1

in CH2Cl2 was added. The nmr spectrum was monitored as the HC1 solution

was added, and the triplet in the methyl region due to starting material was

observed to diminish in intensity as a new triplet just downfield increased in

intensity. Concomitantly, a complex multiple grew downfield from the 5-peak

multiple of starting material in the methylene region. When the triplet due to








ET2NHBH3


15 lHz


Figure 16. Methylene region of the 1H nmr spectrum of
(CH3CH2)2NIBai3.















i/a~flE,.Q~h IN MCONS~
53 I as 'I 3 s e


10 i 2 1 16 16 I ): 25


W00 .2AV

W*AVENUMia CM


iguare 1'7. Infrared spectrum of (CH3C' 112)2NIUB3 .









starting material had disappeared completely, addition of HC1 solution was

stopped and the volatiles removed from the solution by pumping. The oil

which remained in the flask was transferred to a dry box for storage. A sample

of this oil was dissolved in CH2Cl2, and a fresh nmr was taken. This spectrum

showed a triplet in the methyl region, a complex multiple in the methylene

region (Figure 18),and a broad absorption farther downfield, due to N-H

resonance. The chemical shifts and coupling constants are given in Table IV.

The infrared spectrum of this oil, taken between KBr plates, was consistent

with the formulation (CH3CH2)2NHBH2Cl (Figure 19). There was a strong

peak at 3190 cm due to N-H stretch, and a doublet at [2450, 2370 cm-],

due to BH2 stretch. The entire infrared spectrum is given in Table I.

Synthesis of diethylamine-dichloroborane. -- A sample of diethylamine-

borane (about 1. 0 g) was dissolved in CH2C12 and HCI gas bubbled in at a

moderate rate. The nmr spectrum of the solution taken after adding HC1 for

1/2 hour showed two triplets in the methyl region and a complex multiple in

the methylene region. One of the triplets, the one farther upfield, corresponded

to (CH3CH2)2NHBH2C1. The other triplet was assigned to (CH3CH2)2NHBHC12.

The ratio of the two compounds as evidenced by the ratio of intensities of the

two triplets was about 4:1, the monochloroborane adduct in excess. Addition

of HC1 was continued,and the nmr spectra taken at various times showed the

triplet due to monochloroborane adduct diminish as the triplet due to dichloro-

borane adduct increased (no new triplet appeared during the course of this

reaction). After about 4 hours, only the triplet due to dichloroborane adduct














ET2NHBH2Cl







16? Hz

o. -bs.


calc._


__i-.-^ /i~~ u iU ^i; ju, U l h



ig;: A. Methylenc region of the : r.r spectrum cf
(CiI3CH2) ;; 2C


21C H












9 '10 i 1 1i lo 16 i 8 r

i: : I,


1400 1200
WAVEMNUMSR CM


Figure 19. Infrared spectrum of (CH3CII2)2NITBI 2Cl.
22 2









remained. The volatiles were removed from the solution by pumping and

the oil which remained was taken into the dry box for storage.

A sample of this oil was dissolved in CH2C12, and a fresh nmr was taken.

This spectrum showed a triplet in the methyl region, a complex multiple in the

methylene region (Figure 20), and a broad absorption farther downfield, due

to N-H resonance. The chemical shifts and coupling constants are given in

Table IV. The infrared spectrum of this oil, taken between KBr plates, was

consistent with the formulation (CH3CH2)2NHBHC12 (Figure 21). There was

a strong peak at 3200 cm-1, due to N-H stretch, and a single strong peak at

2500 cm-1, due to B-H stretch. The entire infrared spectrum is given in

Table I.

Synthesis of diethylamine-trichloroborane. -- A sample of diethylamine-

borane (about 0. 5 g) was dissolved in 25 ml of CH C12 and chlorine gas bubbled

in until a permanent yellow color was imparted to the solution. The volatiles

were removed by a stream of nitrogen, leaving a white solid, mp = 129-130 .

Anal. Calcd. for C4H 1NBC13: C, 25.25; H, 5.83; N, 7.36. Found: C, 25.46;

H, 6.01; N, 7.22.

The nmr spectrum of (CH3CH2)2NHBC13 in CH2C12 consisted of a triplet

in the methyl region and a complex multiple in the methylene region (Figure 22).

The chemical shifts and coupling constants are given in Table IV. The infrared

spectrum of (CH3CH2)2NHBC13 (Figure 23) was characterized by a strong peak

at 3190 cm-1, due to N-H stretch, and by the absence of B-H stretching

absorptions in the region 2300-2500 cm-1. The entire infrared spectrum is

given in Table I.












ET2NHBHCI2




obs.


S165 Hz
i \ I


cac.


Figur; 20. Metlylei.v ? rgio(n oF *hic: H nt,,, etrun o1 (CHI-3CP T 2NH HCl2


225 Hz
I










35 4 5 I


WAVFIENO.: ;!N PVlCMO6S
55 6 7 7.5 p 10 11 14 16 16 10 20 25
Si j --i -- i -i I j 'i i 1 111i



f- 1 Mf1 \
,It I~ i I ,t! Zl l 11 I li l


400 1200
WAVENUM3ER CM'


Figure 21. Infrared spectrum of (CH3 CIl)2NI1BIIC12.


7 1








220 Hz
t


, I I I II I I I I


ET2NHBC,

170Hz
I


I II


Figure 22. Methylene region of the 1H nmr spectrum of (CII3CH2) NHBCI3.
3 22 3















Fure I-nfrared spectm of (i li )r
-:i 1 I I I I 2 3'
.. . .. ... .I


WAVENUMIER CM


Figure 23. Infrared spectrum of (CH3CH2)2NIi]lCI3.










Reaction of diethylamine-borane and hydrogen bromide. -- Diethylamine-

borane (0. 577 g, 6. 64 mmol) was dissolved in CH 2Cl and a 0.33 M solution

of HBr in CH2C12 (20.0 ml, 6.60 mmol) was added with stirring. The nmr

spectrum of the solution consisted of a triplet in the methyl region, just

downfield from the position of the triplet of the starting material, a complex

multiple in the methylene region (Figure 24),and a broad absorption farther

downfield, due to N-H resonance. This pattern was assigned to (CH3CH2)2NHBH2Br.

Addition of excess HBr caused the peaks due to (CH CH2) NHBH2Br to diminish,

being replaced by a new triplet and complex multiple downfield from those of

(CH3CH2)2NHBH2Br. The new peaks were assigned to (CH3CH2)2NHBHBr2.

Reaction of diethylamine-borane and bromine. -- Diethylamine-borane

(0.432 g, 4.97 mmol) was dissolved in 25 ml of CH Cl Bromine (0.791 g,

4.95 mmol) dissolved in 2 ml of CH C12 was added to the amine-borane solution

dropwise with stirring. The nmr spectrum of the resulting solution showed

two sets of triplets in the methyl region, of relative intensity 5:1, the positions

of which correspond to the previously assigned (CH3CH2)2NHBHBr2 and

(CH3CH2)2NHBH2Br, respectively. More bromine (0.132 g, 0. 83 mmol)

was added as a CH2C12 solution and another nmr spectrum taken. The spectrum

showed only the triplet due to (CH3CH2)2NHBHBr2 in the methyl region. In

addition, there was a complex multiple in the methylene region (Figure 25)

and a broad absorption due to N-H resonance farther downfield. The chemical

shifts and coupling constants for these two adducts are given in Table IV.

The 11B nmr spectrum of (CH3CH2)2NHBHBr2 showed a broad (400 Hz

wide) featureless resonance at 22. 8 ppm upfield from trimethylborate.














ET2NHBH2Br



160Hz





/ obs.


calc.


-- --4


1
F 'r 2 I. Methylene region of the 1 nmr spectrumr of (CH3 CII2)2NIBH2Br.


210Hz
I













ET2NHBHBr2





170Hz


\ obs.


calc.


Figurif 23. Mcthylene region of the H nmr. spectrum of (CHICH 2)~HHBHBr,,.


220 Hz
I









Synthesis of diethylamine-tribromoborane. -- Diethylamine-borane (about

0. 8 g) was dissolved in CH Cl2 and bromine added slowly with stirring until

a permanent color was imparted to the solution. The volatiles were evaporated

by a stream of nitrogen, leaving a light yellow solid, mp = 155-157 Anal.

Calcd. for C4H11NBBr3: C, 14.84; H, 3.43;N, 4.33. Found: C, 15.28;

H, 3.60;N, 4.27.

The nmr spectrum of (CH3CH2)2NHBBr3 in CH2C12 consisted of a triplet

in the methyl region and a complex multiple in the methylene region (Figure 26).

The chemical shifts and coupling constants are given in Table IV. The infrared

spectrum of (CH3CH2)2NHBBr3 (Figure 27) was characterized by a strong

peak at 3180 cm-1, due to N-H stretch, and by the absence of B-H stretching

absorptions in the region 2300-2500 cm-1. The entire infrared spectrum is

given in Table I.

The 11B nmr spectrum showed a single peak at 24.9 ppm upfield from

trimethylborate.

Synthesis of diethylamine-monoiodoborane. -- Diethylamine-borane

(0.985 g, 11.3 mmol) was dissolved in 30 ml of CH2C1l, and iodine (1.44 g,

5. 67 mmol) was added piecewise with stirring. The volatiles were removed

from the resulting solution by pumping, leaving a colorless, viscous oil, which was

taken into the dry box for storage. Yield was 1. 950 g (80. 5%, based on

(CH3CH2)2NHBH3).

A sample of this oil was dissolved in CH2C1l, and an nmr spectrum

obtained. The spectrum showed a triplet in the methyl region, a complex

multiple in the methylene region (Figure 28), and a broad absorption due to








ET2NHBBr

170 Hz
I


\ obs,


calc.


ighylee rgion of te 1 I
S ,1 ,fii i l ili li


Figure 26. Methylene region of the 1H nmr spectrum of (CH3CH )2NHBBr3.


240 Hz
i



















; v '-; : ;- J-' n ,- *- . .. -- l'-i - ,,,h 1.,0
-l I (Ii L ,-





-i I


-,-l- ....----t-- -- H

..... I I ../ A i

3100 305 5 2OW 00 I000 160 00 100 1N00 '000 800 600 OO 100 4Y O 300
WAVENJUMER CM'



figuree 27. Infr:ired spectrum of (CH CH 2)2NHBDr ;
















ET2NHBH2I









155 Hz

obs.


calc.


Fig-urC 28 Methylene region of thf IFT ninr spectrum of (CH CH) NHBHI.
3 2


215 Hz
I










N-H resonance farther downfield. The chemical shifts and coupling constants

are given in Table IV. The infrared spectrum of (CH 3CH2)2NHBH2I (Figure 29)

was taken, between KBr plates, and was characterized by a strong peak at

3140 cm-1, due to N-H stretch, and a doublet at [2450, 2500 cm-1], due to B-H

stretch. The entire infrared spectrum is given in Table I.

Reaction of diethylamine-borane and iodine. -- Diethylamine-borane

(0. 840 g, 9. 66 mmol) was dissolved in 30 ml of CH Cl2. Iodine (2.46 g,

9.70 mmol) was added piecewise with stirring. The nmr spectrum of the

solution taken 3 hours after all the iodine had been added showed two sets of

triplets in the methyl region (in a ratio of about 6:1). The triplet of lower

intensity and farther upfield was in the same position as the triplet of

(CH3CH2)2NHBH2I; therefore, the other triplet was assigned to (CH3CH2)2NHBHI2.

The nmr spectrum taken after 20 hours showed, in the methyl region, the same

two triplets in a ratio of 20:1, (CH3CH2)2NHBHI2in excess. The nmr spectrum

after 60 hours showed, in the methyl region, only the triplet due to

(CH3CH2)2NHBHI2. The complete nmr spectrum of (CH3CH2)NHBHI2 consisted

of a triplet in the methyl region, a complex multiple in the methylene region

(Figure 30), and a broad absorption farther downfield, due to N-H resonance.

The chemical shifts and coupling constants are given in Table IV.

In another experiment, diethylamine-borane (0. 812 g, 9.34 mmol) was

dissolved in 75 ml of C6H in a 125 ml flask fitted with a condenser. The system

was protected from the atmosphere by a stream of N2 through a T-tube above

the condenser. Iodine (7. 00 g, 27. 6 mmol) was added piecewise with stirring

until a permanent color was imparted to the solution, then the rest of the iodine














W\AVELENCTH IN MICRONS
55 5 7 7.5 8


10 II 17 14 16 16 IC 2 25 33 40


WAVENUMIER CM


Figure 29. Infrared spectrum of (CH3CH2)2NHBII2I.


35 4 45 5





ET2NHBHI2



170 Hz


obs.


Figure 30. Methylene region of the H nmr spectrum of (CH3CH2)2NHBBI2.


220 Hz
I









was added and the flask heated until the solution was brought to a gentle reflux.

After 15 hours of reflux, the nmr spectrum of the solution showed two triplets

in the methyl region in a ratio of about 3:1, and in the methylene region, a

quartet overlapping an unresolved absorption. After 40 hours of reflux, the nmr

spectrum showed only a triplet in the position of the triplet which was more

intense in the previous spectrum, and a quartet in the methylene region, in

the same position in the quartet in the previous spectrum. The ratio of intensity

was 39:62 (or about 4:6) and the pattern was assigned to (CH3CH2)2NH2 ion,

the coupling of NH2 to CH2 being lost because of rapid exchange of the

N-H protons.

Reaction of diethylamine and boron trifluoride-etherate. -- A solution of

diethylamine in ether was added dropwise to an ether solution of boron

trifluoride-etherate. Evaporation of the volatiles left low-melting, colorless

crystals of diethylamine-trifluoroborane. On standing in air these crystals

decomposed to a brown oil. An nmr spectrum of this material in CH2 C solution

showed a triplet in the methyl region, a complex, unanalyzable multiple in the

methylene region (Figure 31),and a broad absorption farther downfield, due to

N-H resonance. The chemical shifts and coupling constants, where assignable,

are given in Table IV.

Reaction of N-Deuterodimethylamine-Borane with Various Halogenating Agents

Synthesis of N-deuterodimethylamine-borane. -- A sample of dimethyl-

amine-borane (about 5 g) was added to 40 ml of D20 and stirred for 2 hours,

during which time most of the borane dissolved. The borane was extracted

from the D20 with two 40-ml portions of methylene chloride, using a 125-ml




















ET2NHBF






160Hz


Figure 31. Methylene region of the H nmr spectrum of
(CHI3CH2)2NHBF3.


200Hz
I









separatory funnel. The resulting solution was dried over Na2CO3 overnight,

then filtered and pumped to dryness. A white crystalline solid remained,

mp = 34. 5-35. 5 [literature 62 for (CH3)2NHBH3 = 36 ]. The nmr spectrum

in CH2C12 showed a single peak at 151.5 Hz, downfield from internal

tetramethylsilane (TMIS). The infrared spectrum was identical to that of

authentic (CH 3)2NDBH3 prepared from N-deuterodimethylammonium chloride

and lithium borohydride.3

Reaction of N-deuterodimetiylamine-borane and chlorine. -- A sample of

(CH3)2NDBH3 (about 0.05 g) was dissolved in 20 ml of CH2C12,and Cl2 gas

was bubbled in until the solution turned yellow. (The rate of addition was such

that the color change occurred within 10 minutes.) Volatiles were removed

by pumping, and a white solid remained. The nmr spectrum of this solid in

CHC2 l solution showed a six-peak multiple centered at 171. 5 Hz, dovwfield

from internal TMS, with an intensity ratio of about 1:3:4:4:3:1. The infrared

spectrum of this solid showed peaks at 3200 cm- (N-H stretch) and at 2400 cm-1

N-D stretch) in a ratio of about 1:1, in addition to ihe pattern for (CH3)2NHBCI3.

(Thrc.e was an extra peak at 970 cm-1.) Thus this solid was a mixture of

S-' 2NIIBC and (CH3)2NDBC3 in a 1:1 ratio.
2i 3 3 2 3
Note that the nmr pattern expected for (CH3 )NDBC13 is a 1:1:1:1 quartet,

wilh peaks coinciden with the 2nd, 3rd, 4th,and 5th peaks of the (CII )2NHBC13

sextet.

bt:action of N-deuteridimethvlamine-boran and bromine.-- A sample of

CT3 ) 2NDBH3 (abuut 0. 05 g) was issol'ed in 25 ml of CH2Cl, and Br2 was

added dropwise with :1, i.. iuril 'he solution turned yellow. The time of









addition was about 15 minutes. Volatiles were removed by pumping, leaving

a white solid. The nmr spectrum of this solid in CII 2C2 solution showed a

quartet (1:1:1:1) with two small peaks, one just upfield and one just downfield

of the quartet. This pattern was centered at 182. 5 Hz, downfield from internal

TMS. The infrared spectrum of this solid showed peaks at 3150 cm-1 (N-H

stretch) and 2350 cm- (N-D stretch) in a ratio of about 1:9, in addition to

the pattern for (CH3)2NHBBr3. (There were extra peaks at 870 cm-1 and

960 cm1.) Thus the solid was a mixture of (CH3)2NHBBr3 and (CH 3)2NDBBr3

in 1:9 ratio.

Reaction of N-deuterodimethylaimine-borane and iodine. -- A sample of

(CI- .oi'.,l3 (0. 0255 g, 0.432 mmol) was dissolved in 5 nm of CH2C12 and 12

(0 0553 g, 0.218 mmol) added piecewise with stirring. The nmr spectrum of

the resulting solution showed one peak at 165 liz, downfield from internal TAIS,

with no fine structure. Thus the product was (CH3)2NDBH2I.

A sample of (CH3)2NDBH3 (0. 1098 g, 1. 83 mmol) was dissolved in 15 ml

of CH2C 2. 12 (0.7003 g, 2.73 mmol) was added and the solution stirred. The

nmin spectrum of the solution after 7 hours of stirring showed one peak, with

no fine structure, at 173 Hz, downfield from internal TMS. Thus the product

was (Ci3)2NDBlHI2

Reaction of N-deuterodimethvllamine-borane and hydrogen chloride.-- A

sample of (CIH3)2NDBH3 (about 0.05 g) was dissolved in 25 ml of CH 2Cl, and

iHC gas w- s bubbled in for 30 minutes. The HC1 .w's thus present in large excess.

The voeitiles wcre removed immediately thertaft',r by pumping, leaving a

gummny solid. The nmr spectrum of this soin in Cti,Cl showed a 3-peak









multiple centered at 157 Hz, downfield from internal TAIS, which appeared

to be a strong singlet overlapping a doublet. The singlet is thus assigned

to (CH ) NDBH Cl and the doublet to (CH ) NHBHI Cl. The ratio of these two

compounds, as estimated visually. was about 1:3, the deuterated compound in

excess.

Reaction of N-deuterodimethvlamine-borane with hydrogen chloride and

chlorine -- A sample of (CH3)2NDBH3 (about 0. 05 g) was dissolved in 20 ml

of CH2Cl2. Both HCI gas and C12 gas were bubbled into the solution, the rate

of addition being such that the HCI was added in about a 10-fold excess over the

C12. After about 7 minutes of addition, the solution turned yellow, whereupon

the addition of gases was stopped and the volatile removed bh pumping,

leaving a white solid. The nmr and infrared spectra of this solid were nearly

identical in pattern and intensity to that of the solid formed by reaction of

(CH3)2NDBH3 with C12 alone. Thus, the incorporation of hydrogen on nitrogen

as the result of reaction of (CH )2NDBH3 with Cl2 is not affected by the

presence of IICI in excess.

PRaclion of N-dcuterodimethylamine-borane with deuterium chloride and

chlorine e. -- A sample of (CII3)2NDBri3 (about 0. 05 g) was dissolved in 25 ml

of CIH.2C12, Bth DC1 gas and Cl gas were bubbled into the solution, the rate

of addition being sucl 1itat the DCI was added in about a 10-fold excess over the

Cil. (The DC1 wa. poroluced from the hydrolysis of benzoyl chloride by D 0 as
2
dcc'-ribed in reference 63. The benzoyl chloride was heated for about 45 minutes

to ri.ive off any HCI produced as result of hydrolvysis by 20 impurity. D20

(99.5%) vaas then, aded,and the resulting DC1 was usPea only after a good stream










of gas was produced. Thus the purity of DC1 was 99% or better ) After about

20 minutes of addition, the solution turned yellow, whereupon the addition of

gases was stopped and the volatiles removed by pumping, leaving a white solid.

The nmr spectrum of this solid in CH Cl2 solution showed a six-peak multiple

in the same position as that observed for the product of the reaction of

(CH3)2NDBH3 with Cl2 alone, but the intensity ratios were quite different.

The outside two peaks had lost intensity while the 2nd and 5th peaks had gained

intensity. The infrared spectrum of this solid showed peaks for N-H stretch

(3200 cm- ) and N-D stretch (2400 cm- ), but the relative intensity was about

1:3, the N-D peak being the stronger. Thus, incorporation of hydrogen on

nitrogen as the result of reaction of (CH3)2NDBH3 with C12 is greatly retarded

by the presence of DC1 in excess.

Reaction of N-deuterodimethylamine-borane and hydrogen chloride at

low temperature and the synthesis of N-deuterodimethylamine-mono- and

dichloroborane. -- A sample of (CH3)2NDBH3 (about 0. 05 g) was dissolved in

8 ml of CH C12 and the solution cooled to about -78 by immersion in a dry

ice/acetone bath. HCI gas was bubbled into this solution for 10 minutes, then

nitrogen gas was bubbled in for 30 minutes to remove excess HC1. After warming

to room temperature, the nmr spectrum of this solution was obtained, which

showed one peak with no fine structure at 157 Hz,downfield from internal TMS.

An infrared spectrum of the solute was obtained by using matched NaCl liquid

cells with CH Cl2 as a standard. This infrared spectrum showed no peak at

3200 cm indicating the absence of N-H. Thus the product was (CH3 NDBH2C1.










In another experiment, the solution of (CH3),NDBIHCl and HCI at -73

was allowed to warm up to room temperature without removing HC1 beforehand.

Thus (CH3)2NDBH2CI was exposed to HCI at room temperature. The nmr

spectrum of this solution after 20 minutes at room temperature showed no fine

structure on the singlet due to (CH3)2NDBII2C1. Thus there was no exchange

between (CH 3)NDBH2CI and HCI at room temperature within 20 minutes.

In another experiment, a sample of (CH 3)2NDBH3 (about 0.05 g) was

dissolved in 8 ml of CH2Cl2 and cooled to -78 by immersion in a dry ice/acetone

bath. HCI was bubbled in at a moderate rate for 50 minutes. The resulting

solution was then allowed to stand, in a dry ice/acetone bath. and the nmr

spectrum vas taken from time to time. The nmr spectrum after S-1/2 hours

showed two singlets, one at 157 Hz [(CH ) NDBH Cl] and one at 163 Hz

[(CH3)2NDBHC12]. downfield from internal TMIS. The ratio of products was

about 3:1, the dichloro adduct in excess. After 21 hours the spectrum showed

a ratio of about 9:1 for (CH3) NDBHC12: (CH3)2NDBH Cl. There was still no

fine structure on the peak due to (Cl r i"'. 2 After 32 hours the spectrum

showed only (CHl) NDBHCi2. An infrared spectrum was run on this solution

in matched NaCl liquid cells with CH2Cl2 as a standard. This infrared spectrum
-1
showed no peak at 3200 cm1 Thus the product was (C )2NDBHICI .

A portion of the solution of (Ci3) NDBIICl2 and HCI at -78 as prepared

above was allowed to warm up to roonm temperature without removing excess

HCI beforehand. The nmr spectrum of th:; lution after 20 minutes at room

temperature showed no fine structure on the single due to (CH3)2NDBHCI2.









Thus there was no exchange between (CIH i.DF i it, and HCI at room
2
temperature within 20 minutes.

Reaction of N-deuterodimethvla mine-monochloroborane and chlorine. --

A solution of (CH3)2NDBI2 Cl was prepared as described above, the excess

HC1 gas having been removed by bubbling in 2 gas and by a short period of

pumping. Cl2 gas was then bubbled in until the solution turned yellow. (Time

required was about 15 minutes.) The volatiles were removed by pumping,

leaving a white solid. The nmr spectrum of this solid,dissolved in CH2 Cl2

showed a six-peak multiple centered at 175 Hz,downfield from internal TMS,

with intensity ratios of approximately 1:12:15:15:12:1. The infrared spectrum

showed both N-IH stretch (3200 cm-1) and N-D stretch (2400 cmn- ), but the

N-D stretch was five to six times as intense. Thus the product was a mixture

of (CH 3)2NIBCl3 and (CH3)2NDBCl3 in a ratio of 1:G.

Reaction of N-deuterodimethvlamine-dichloroborane and chlorine. -- A

solution of (CH3)2NDBBC12 was prepared as described above, the excess HCI

gas having been removed by bubbling in N2 gas and by a short period of pumping.

Cl2 gas was then bubbled in until the solution turned yellow. (Time required

was about 5 minutes.) The volatiles were removed by pumping, leaving a

white solid. The nmr spectrum of this solid,dissolved in CH 2Cl2 showed a six-

peak multiplet centered at 175 Hz,downfield from internal TMS, with intensity

ratios of approximately 1:20:25:25:20:1. The infrared spectrum showed both

N-tH stretch (3200 cm- ) and N-D stretch (2400 em- ), but the N-D stretch was

seven to eight times as intense. Thus the product was a mixture of

(CiP : i' Ir i3 :nd (CHR) NDBCI3 in a ratio of 1:8.












temperature. -- A sample of (CHI3)2NDBH3 (about 0. 05 g) was dissolved in

10 ml of CH2Cl2 and the solution cooled to about -78 by immersion in a dry

ice/acetone bath. Cl2 gas was bubbled into the solution for 10 minutes at

which time the solution was yellow. The volatiles were removed by pumping,

leaving a white solid. The nmr spectrum of this solution showed a quartet,

centered at 175. 5 Hz,downfield from internal TMS, with intensity ratio of

approximately 1:1.2:1.2:1. Two very low-intensity satellite peaks were visible,

one upfield and one downfield of the quartet, in positions corresponding to the

Ist and 6th peaks of the (CH3)2NHBC13 sextet. The infrared spectrum of this

solid showed both N-H- stretch (3200 cm- ) and N-D stretch (2400 cm ), but

the N-D peak was fifteen to twenty times as intense as the N-H peak. Thus the

product is a mixture of (CII3)2NHBC13 and (CH3)2NDBC13 in a ratio of

approximately 1:20.

Reaction of N-deuterodimethylamine-trichloroborane and hydrogen

chloride. -- A mixture of (CH3)2NHBCI3 and (CH3)2NDBC13 in a ratio of 1:1

was prepared, as described above, by the reaction of (CH3)2 NDBH3 and Cl 2

A sample of this mixture (about 0. 05 g) was dissolved in 25 ml of CH2CI2, and

HCI ga, was gbJobled into the solution for 2 hours. (During this time, the CII Cl

lot by evaporation was replaced to keep a nearly constant volume of 25 ml.)

After the addition of HC1, the volatiles were removed by pumping, leaving a

white solid. The nmr spectrum of thi; solid was supcrimposable on the nmr

spectrum of the starling mixture. The infrared spectrum of this solid was the









same as the infrared spectrum of the starting mixture within the limits of

variation expected from variation in pellet composition. Thus (CH 2NDBCI3

and HCI do not exchange to a measurable extent at room temperature within 2

hours.

A mixture of (CH3)2NHBCI3 and (CH3)2NDBC13 in a ratio of 1:20 was

prepared,as described above, by the reaction of (CH3)2NDBII3 and Cl2 at

-78 A sample of this mixture (0. 0170 g) was dissolved in 3 ml of 0.12 M

IICI in CH2Cl2. The ratio of HCl:adduct was thus approximately 3:1. About

1 ml of the resulting solution was transferred into an nmr tube, and the nmr

spectrum was monitored as a function of time. There was no observable change

in Jhe spectrum, even after 24 days. Therefore. within the limits of

detcctability of nmr, HCl does not exchange with (CH, i '." 13 within 24 days.

Reaction of N-deuterodimethvlamine-trichloroborane with hydrogen

chloride and chlorine. -- A sample (about 0.02 g) of nearly pure (CH3)2NDBC13

prepared, as described above, by the reaction of (CH3)2NDBH3 and C12 at

-78 was dissolved in 15 ml of CH ClI, and both HCI gas and CI2 gas were

bubbled into the solution for 30 minutes. The volatiles were then pumped off,

leaving a white solid. The nmr spectrum of this solid was superimposable

on the nmr spectrum of the starting material. The infrared spectrum of this

solid vas the same as the infrared spectrum ot the Fta rating material within the

limits of variation expected from variation in pellet composition. Thus

(CiH3) NDBCl and HICI do not exchange to a measturable extent at room tempera-

ture in the presence of Cl, within 30 minutes.




86




Reaction of d.imethlamir.e-trichlo)roberene and deuterium chloride. --

Dimethylamine-tri'chloroborane (0. 0333 g. 0 218 mmol) was dissolved in

1 ml of a 0.22 M solution of DCI in CH Cl2 in an mnr tube, and the nmr

spectrum was monitored as a function of time. There was no observable change

in the spectrum, even after 11 days. Therefore, within the limits of

detectability of nmtr, DC1 does not exchange with (CH )2NtHBC13 within 11 days.










Table I

Infrared Data for Some Amine-Haloboranes


(CI3) 2NHBF3:




(CH3)2NHBCI3:









(CH)2 NHBBr3:







(CH3)2 NHI3:






(CH3 : 2 :3I 2Cl:


3260(m), 3080-2960(m). 1630-1600(w), 1465(m), 1340(m),
[1150, 1040, 940, 910(s)], 705(w), 560(w), 480(w).


3180(s), 3020(w), 2970(w), 2780(w), 2700(w), 2660(w),
2440(w), [1475, 1465, 1450, 1440(m)], 1410(m), 1380(m),
1340(m), 1230(w), 1195(w), [1150, 1140, 1130(m)],
1050(w), 1010(m), 900(s), 840(m), 810(s), 780-700(s),
510(m), 460(w), 375(w), 360(m).


3220(m), 3150(s), 2780(w). 2680(w). [1470, 1460, 1445,
1435(m)], 1410(m), 1370(m). 13:35(i(). 1200(w), [1140,
1130(m)], 1000(m), S95(s), 825(m), 790(s), [710, 680,
665(s)], 455(m).


3300-2950(s), 2770(m), 2640(v), 1500-1430(s), 1405(m),
1360(m), 1330(m), 1200(m), [1135, 1125(m)], 1040(w),
1015(w), 990(m), S95(m), 820-790(m), 775(m), [645,
620, 600(s)], 550(w), 420(m).


3190(s), 3040(sh), 2990(s), 2945(w), 2910(w), 2890(w),
2840(w), 2450(s). 2373(m), 1470(m), 1445(m), 1395(m),
1370(w), 1315(w), 1260(m), 1190(m), 1150(s), 1130(s),
l100(mi. 1065(m), 1030(m), 1020(sh), 980(w), 870(m),
790(m). 710(e), 615(n)).