SYR‘TEESIS 0F a-BALGGEGARGEORGK CGBEPOQEBS

Thesis Eor fin Degree of M 5.
EECEESAB STATE UHEVERSX'EY
Esther Yu-hsuau Chase

W70

.‘rn‘g

{(33-0. A " l
5

“I“; "a" "1"" - . . y
L I u A A x 3.
M id“) 1mm 3 t a
U

Univcrsicy

 
 

w—uwj-

4“..',. n -.'_4,<:‘-’. 6‘ :1,

 

 

 

 

ABSTRACT
SYNTHESIS OF a-HALOORGANOBORON COMPOUNDS
By

Esther Yuehsuan Chao

Since a-haloorganoboron compounds are useful in organic
synthesis and mechanistic studies, convenient methods to
prepare these compounds are desirable. It is found that
reactions between phenyl(trichloromethyl)mercury and boron
trihalides do not produce the expected a-haloorganoboron
compounds. Similar reactions of sodium trichloroacetate
with boron trihalidesare also unsuccessful. However, the
reaction of trimethyl borate with dichloromethyllithium
gives a salt with the structure ClZHagiOCH3)3Li. Hydrolysis
of the salt affords the dichloromethaneboronic acid,
CleCB(OH)2, in very good yields, 90-95%. Subsequently,
the acid derivative of diethanolamine, CleCB(OCH2CH2)2NH,
is made very readily in 70—75% yield. In addition, di-flf
propyl dichloromethaneboronate, C12HCB(OCH2CH2CH3)2, is

obtained in 50-60% yield.

SYNTHESIS OF a-HALOORGANOBORON COMPOUNDS

BY

Esther Yu-hsuan Chao

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Chemistry

1970

To

My Parents

ii

ACKNOWLEDGMENTS

The author wishes to express her appreciation to Dr.
Michael W. Rathke for his guidance and encouragement
throughout this study.

Thanks are given to the members of her research
group who have all contributed in one way or another to
the successful completion of this thesis research.

Thanks are also given to Yen Chao, the author's

husband, for his encouragement and understanding.

iii

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . . .

LITERATURE REVIEW . . . . . . . . . . . . . . .

RESULTS

0 O O O O C O O O O O O O O O O 0 O 0 0

DISCUSSION . . . . . . . . . . . . . . . . o . .

EXPERIMENTAL O 0 O O O O O O O O O O O O O O 0 0

A.

B.

C.

D.
E.

Phenyl(trichloromethyl)mercury. Method
Schweiger and O'Neill25 . . . . . . . .

Attempted Reaction of Phenyl(trichloro-
methyl)mercury and Boron Trichloride .

Attempted Reaction of Phenyl(trichloro-

methyl)mercury and Boron Trifluoride Etherate

Sodium Trichloroacetate . . . . .*. . .

Attempted Reaction of Sodium Trichloroacetate

and Boron Trichloride . . . . . . . . .

Attempted Reaction of Sodium Trichloroacetate

and Boron Trifluoride Etherate . . . .

Trichloromethyllithium Solution. Method of

Kobrich and Flory16'17 . . . . . . . .

Reaction of Trichloromethyllithium and
Benzophenone" . . . . . . . . . . . . .

Attempted Reaction of Trichloromethyllithium

and Boron Trifluoride Etherate . . . .
Dichloromethyllithium Solution. Method
Kobrich and Flory18 . . . . . . . . . .

Reaction of Dichloromethyllithium and
Trimethyl Borate . . . . . . . . . . .

0

of

Preparation of Dichloromethaneboronic Acid

(x)

iv

Page

20
29
29
3O

31
31

32
32
33
33
34

34

35

35

TABLE OF CONTENTS (Continued)

M. Derivatives of Dichloromethaneboronic
Acid (X) . . . . o . . . . . . . . o
a. Attempted Preparation of Boroxine
b, Acid Derivative of Catechol (XII)
c. Acid Derivative of Diethanolamine
N° Esters of Dichloromethaneboronic Abid
a. Ethylene Glycol Ester (XIV) . o
b. Ethanol Ester (XV) . . . . . .
c. .g—Butanol Ester (XVI) . . o . . .
d. anropanol Ester (XVII) . o .
REFERENCES . . o . o o . . o . . . o o . . o

Page

36
36
36
36
37
37
38
38
39

4O

INTRODUCTION

The reactivity of a1halogens to nucleophilic displace—
ment is known to be greatly enhanced by a neighboring boron
atom. a—Haloorganoboron compounds, especially the boronic
esters which are relatively quite stable, are therefore
potentially useful for synthetic applications and mechan-
istic studies. Consequently, it was decided to develop
methods for the convenient preparation of a-haloorganoboron
compounds. The reactions of a—haloorganometallic compounds
with boron halides or borate esters were chosen for study.

This thesis reports the results of this study.

LITERATURE REVIEW

Recently, the ability of the neighboring boron atom
to increase the reactivity of Q-halogen as a leaving group
in nucleophilic displacements has been studied in synthetic
and mechanistic aspects. Work has been done mostly by D.S.
Matteson on the formation of halogen—carbon—boron bonds
during the last ten years.

In 1960, D. S. Matteson obtained the first a-halogen
aliphatic boron compound by radical-catalyzed addition of

polyhalomethane to o,5—unsaturated boronic esters.1 For

 

example,
radical
CH2:CHB(OBu)2 + Cl3CBr initiator > C13CCH2CHBrB(OBu)2
b.p. 95-960/o.08 mm
94%

The intermediate radical of the reaction is resonance-

stabilized. The canonical structures are expressed as:
(5 <3
Cl3CCH2CH-B(0Bu)2 <—————> C13CCH2CH-B(0Bu)2

If the a—halo boronic ester is contacted with base,
hydrolysis of both the ester and a—halo group occurs, but
with cold water only the corresponding boronic acid is ob—
tained. The acid polymerizes with evolution of hydrogen

halide either on storage or on heating.1
2

3
In 1961, Zeldin and Girardot described the synthesis
of chloromethyldimethylborane (I) by direct chlorination
of trimethylborane at -95°2. They found this derivative

(I) to be reasonably stable at 0°.

-950

 

(CH3 ) 3B + C12 > H2CC1B (CH3 ) 2 (vapor tension
0
(I) 36.5 mm at O )
34%

The ionic addition of hydrogen bromide or iodide to
a,5-unsaturated boron compounds was carried out by D. S.
Matteson and his coworker in 1963 when the first simple

a-haloorganoboron compounds became available.3

CH3
_ 0 o
CH3CH=C13(OBu)2 + HBr(liquid) §—%%§> CH3CH2CB(OBu)2
l I
CH3 Br
(11)

The dibutyl 2—bromobutane-2-boronate (II) (boiling point
64~65°/O.1 mm) was obtained in 61% yield. The addition
of hydrogen halide across the double bond is directed by
the electron-donating inductive effect of the boryl group

which stabilizee form (A) over form (B).3

GD GE
CH3CH2CB(OBu)2 CH3CHCHB(OBu)2

I I

CH3 CH3

(A) (B)

The displacement reaction of a—halo boron compounds
with sodium iodide in acetone was studied by D. S. Matteson

in 1966.4 Comparison with literature values for simple

4
alkyl halide gives the following relative rate constants:
isopropyl bromide 1.0; ethyl bromide, 40; dibutyl 2—bromo-
propane—Z-boronate, 1600; allyl bromide, 4000; dibutyl
l—bromoethaneboronate, 6000.4 It was concluded that the
neighboring boron atom participates in and greatly acceler-
ates nucleophilic displacement of the a-halide ion.5
Attack of the nucleophile was assumed to occur initially
on the boron atom, followed by migration of the nucleophile
to the a-Carbon atom with expulsion of the a—halide ion.

The following mechanism is representative:6

 

 

x }<CD f C)
' I Nu: , ' / \ :._ I ' /
-C-B\ >- (3-1.3‘ > I C: P\ > -C B\ + x
Nu 3 ‘ Nu

Nu

The preparation of halomethaneboron compounds has
proved difficult. A very small yield of chloromethane-
boronic ester was made by Chlorination Of methaneboronic
acid derivative with tfbutyl hypochlorite.7 Irradiation
of tfbutyl hypochlorite and di-Efbutyl methaneboronate at
00 did produce some chloromethylboron compound. The
product which was isolated was obtained in only 9% yield

as the di-g-butyl ester (III).8

 

CH3B(OEfBu)2 + EfBuOCl > ClCH2B(OEfBU)2 + CH3BOC(CH3)2
I
CHZCl

+ other products QCEEQE—> ClCH2B(OEfBu)2
(III)

The problem is that chlorination of the B-methyl group is

5
barely favored energetically over attack on C-methyl groups,
which outnumber the former nine to one.8
However, by starting from boron tribromide and iodo—
methylmercuric iodide, useful quantities of dibutyl iodo-

methaneboronate (IV) were obtained.9

 

 

O +
ICHZHgI + BBr3 temgeggture > ICHzBBrZ BUOH NaI> ICH2B(OBu)2
(IV)
b.p. 60°/O.1 mm
35-40%

Ethyleneboronic ester (V) at first was found to be
inert to liquid hydrogen bromide at —70°.3 By refluxing
with hydrogen iodide, the corresponding a-iodo compound

(VI) is obtained as the predominant product as follows:4

CH2:CHB(OBu)2 + HI £§£l3§> CH3CHB(OBu)2 + ICH2CH2B(OBu)2
I
(V) I (v1)
60% 40%

The pure a-iodoethaneboronic ester (VI) is light sensitive
and unstable on storage. Decomposition products include
iodine and butyl borate.4

D. Seyferth and B. Prokai, in 1966, proposed that a-
haloalkaneboron compounds (VII) are formed as intermedi-
ates in the reaction of trialkylboranes with bromodichloro-

methyl phenyl mercury.10

hydroboration C6H5HgCClzBr

 

3RCH=CH2 > (RCH2CH2 ) 3B

 

GKHhCH2)2BCC12CH2CH2R
(VII)

> RCHZCHZBCIZ + RCHZCHZCH=CHCH2R

Attempts to isolate the intermediate (VII) were not success-
ful. In the meantime, D. J. Pasto and his coworker found

that hydroboration of vinyl halides leads to the formation

of highly reactive intermediate adducts which undergo
further reactions.11 The addition of the boron occurs

predominantly to the carbon bearing the halogen.12

 

\. Cl
\\ ..... 3”
. \.

hydrobpration ‘ . + .

60% 40%

Furthermore, the stereochemistry of the rearrangement of

o-haloorganoboranes was shown to be as follows:13

 

 

  

 

Cl H23 ”Cl H \ _‘..Cl
CH3 ‘ J$CH3 ‘ naCHa
hydroboration H + \‘BHZ
H xfiHCl e//;:3:::;;sfer ////////é-elimina-
‘ H inversion L tion
“CHE,
‘\ CH3
+ BH2C1
oxidation
1 o BH3
OH OH 2. Oxidation
" \§\ H
.s C 3 CH3
~3/I L +
cis
41% trans

In 1968, D. J. Pasto prepared 1-chloro-2-methyl-
propylboronic acid (VIII) by hydrolysis of the correspond-

ing borane in excellent yields, 2.84%.14 The reaction
is shown as:
BH3 H20

tetrahydrofuran> (CH3)2CHCHC1BH2 >
room temperature

(CH3)gc=CHCl

(CH3)2CHCHClB(OH)2
(VIII)
m.p. 63-640

7
At present, efforts are still being made to find con-
venient synthetic methods for preparing a-haloorganoboron
compounds. This thesis describes the successful synthesis

of dichloromethaneboronic acid and its esters.

RESULTS

The reaction between a-haloalkylmetallic derivatives

and boron compounds is a reasonable synthetic approach to

d-haloorganoboron compounds. The reaction could be gener—

alized as:

X X
I I
RCHM + BX; > RCHBX; + MX'

 

X = halogen, X' = halogen, methoxide, etc.

Phenyl(trichloromethyl)mercury is a stable compound.

Therefore it was, initially, used as the a-haloalkylmetallic

intermediate in this synthetic work. The reaction between

phenyl(trichloromethyl)mercury and boron trihalide would

be expected to produce compounds of the type C13CBX2 as

follows:

/«

+ O

A mixture of phenyl(trichloromethyl)mercury and excess

boron trichloride was refluxed under a dry-ice condenser
overnight at 00 under nitrogen. Simple distillation of
the reaction mixture led to decomposition. A dark solu-

tion was obtained from which no definite products could

be isolated.

9
Since it is known that the B-0 bond is stronger than
the B-X bond, the ethylene glycol ester of the boron com-
pound might be expected to be more stable than organoboron
dihalide. The ethylene glycol ester of the Cl3CBX2 might

be prepared as follows:

 

C13CBX2 + (CHZOH)2 > Cl3CB(OCH2)2 + 2HX

Reaction of stoichiometric quantities of phenyl(trichloro—
methyl)mercury and boron trichloride was performed in
chlorobenzene solution. The reaction mixture was stirred
for one hour at 0° under nitrogen, followed by esterifica-
tion with ethylene glycol. Attempts to distill the cor-
responding ethylene glycol ester under reduced pressure
afforded only boric acid which melted at 180° with de-
composition. Similar attempts to react phenyl(trichloro—
methyl)mercury with boron trifluoride etherate were also
unsuccessful.

The thermal decomposition of sodium trichloroacetate
produces NaCCl3. The trichloromethyl anion is unstable
in the reaction media and decomposes to give dichlorocar-
bene by prolonged reaction time.15

An attempt was made to intercept the trichloromethyl
anion with boron trihalide to obtain a-haloorganoboron com—
pounds of the type Cl3CBX2. The ethylene glycol ester of
Cl3CBX2, which might be relatively stable, was desired.

The overall reaction could be shown as follows:

10

O
NaOCCCl3

 

 

A > C0.) + NaCCl3 > C13CBX2 + NaX
) (CHZOH)2

C13CB(OCH2)2 + 2HX

A solution of stoichiometric amounts of sodium trichloro-
acetate and boron trichloride was refluxed under a dry-ice
condenser for one hour at 00 under nitrogen. The evolu-
tion of carbon dioxide gas was observed. Esterification
with ethylene glycol and distillation of the reaction
mixture under reduced pressure afforded only boric acid.
Similar reactions between sodium trichloroacetate and boron
trifluoride etherate gave identical results.

In 1964, KObrich and Flory described the preparation
of trichloromethyllithium solutions”:17 by slow addition
of an equivalent amount of nfbutyllithium to a tetrahydro-
furan etherIpentane (4:1:1) solution of chloroform at -108°

under nitrogen, as shown by the following equation:

tetrahydrofuranI

HCCl3 + CH3CH2CH2CH2Li ethfiggintane > LiCCl3 + CH3CH2CH2CH3

 

It was thought that trichloromethyllithium might re-
act with the boron compound to give a product of the struc-

ture Cl3CBX2 as shown in the following equation.

tetrahydrofuran]

etherlpentane > Cl3CBX2 + LiX

LiCC13 + BX3 -1080

In order to verify the formation of trichloromethyllithium,

the reaction between trichloromethyllithium solution and

11

benzophenone was performed as follows:

 

 

_ +
- +
:iizifiyzizixza“ 9“
LiCCl3 + ¢CO¢ “ 'p 0 ¢C~CC13 0 >
-108 $ -108

¢2C(OH)CCl3

It was found that hydrolysis of the lithium salt with con~
centrated sulfuric acid and methanol at -108° instead of
at room temperature afforded ¢2C(OH)CCl3 in a higher yield
(30%) than reported by KObrich (20%).17

The reaction between trichloromethyllithium and boron
trihalide was performed. An equivalent amount of
BF3°O(CH2CH3)2 was added to a trichloromethyllithium solu-
tion at -108° under nitrogen to produce a clear solution.

The reaction mixture turned a black color as the tempera-
ture was raised to room temperature, indicating decomposi-
tion. No definite product Could be isolated. Use of tri~
methyl borate in place of boron trifluoride etherate gave
similar results. Attempts to isolate any trihalomethylboron
product were unsuccessful.

Dichloromethyllithiunn'was first prepared by KObrich
and Flory in 1964.18 Dichloromethyllithium, which is more
stable than trichloromethyllithium, was prepared by slow
addition of an equivalent amount of nfbutyllithium to methylene
chloride in tetrahydrofuran solution at -1000 under nitrogen.

The reaction is shown as:

tetrahydro-
furan

-1oo°

 

CH2C12 + CH3CH2CH2CH2Li > LiCHClZ + CH3CH2CH2CH3

12
It was proposed that dichloromethyllithium would re-
act with trimethyl borate to produce a compound like

C12HCB(OCH3)2 as follows:

tetrahydro-

LiCHCl2 + B(OCH3)3 furan > CleCB(OCH3)2 + LiOCH3

-1oo°

Addition of trimethyl borate to a dichloromethyllithium
solution at -100°, followed by removal of solvent afforded
an oil-like material with a positive flame test for boron.
The nmr spectrum showed a singlet at O 5.3 (1H) and a sing-
let at O 3.3 (9-1OH). The results suggest the structure

€> .69 . .
of the product as CleCB(OCH3)3L1 (IX) which 13 formed

according to the following equation:

tetrahydrofuran >

6) .
O C12HCB(OCH3)3L1
-100

(IX)
The compound (IX) is soluble in ethyl ether and chloroform.

LiCHCl2 + B(OCH3)3

However, attempts to crystallize it from these solvents
failed.
Hydrolysis of the lithium salt (IX) should give the

corresponding boronic acid (X) as follows:

 

9 .69 dilute HC1 -
C12HCB(OCH3)3L1 tetrahydrofuran> C12HCB(OCH3)2 + CH30H + L1Cl
_ 0
(Ix) 100 1 H20
room temperature
C12HCB(OH)2

(X)
It was found that hydrolysis of the lithium salt (IX) with

dilute hydrochloric acid at -100°, followed by extraction

13
with ethyl ether, and concentration of the organic phase
afforded a white solid with a positive flame test for boron.
After recrystallization from ethyl ether. the material
melted at 97-98°. The nmr spectrum showed a singlet at
O 5.25 (1H) and a singlet at O 4.88 (2-3H) in (CD3)ZSO.
The nmr spectrum indicated the presence of boric acid,
B(OH)3, together with the product (X). The overall yield
of (X) was 90-95%. The material is soluble in alcohols and
tetrahydrofuran, but insoluble in diethyl ether, chloroform,
carbon tetrachloride, and benzene. Although it is quite
stable in air, the a-halo boronic acid (X) might lose
water readily and reversibly to form the corresponding

boroxine (XI):19.20

 

/' ‘\
_3H20 CleC-B E-CHClZ
3C12HCB(OH)2 :f——————$ O O
‘ 3H20
(X) \‘B’/
CHCl2
(XI)

The presence of boric acid and boroxine (XI) could explain
the variable ratios of protons in the nmr spectra.

Since the a-halo boronic acid was not fully character-
ized, derivatives of the acid were desired. An attempt
was made to synthesize the corresponding boroxine (XI) by
dissolving the acid in benzene and distilling the benzene—
water azeotrope at 70°. However, the reaction mixture
turned brown and then gave a black residue of carbon under

the reaction conditions.

14
It is known that the boronic acid derivatives of
catechol and diethanolamine are solids in many cases,
which makes them useful derivatives for the characteriza-
tion of boronic acids. The preparation of dichloromethane—
boronic acid (X) derivatives of catechol and diethanol-
amine was therefore attempted.

The reaction of the acid and catechol can be shown as:

0H
tetrahydrofuran /
ClZHCB<OH>2 + @ OH room temperature> ClZHCon + ZHZO

 

(X)
(XII)

A mixture of stoichiometric amounts of the acid and catechol
in tetrahydrofuran was stirred for one hour at room tempera-
ture under nitrogen. A white solid with a positive flame
test for boron was obtained, but it turned to a deep purple
color soon after it was isolated.

The reaction of the acid and diethanolamine is shown

as:

tetrahydrofuran
room temperature

 

C12HCB(OH)2 + NH(CH2CH20H)2
(X)
CleCB (OCHzCHz ) zNH + 2H20

(XIII)

An equivalent amount of diethanolamine was added dropwise
to the tetrahydrofuran solution of dichloromethaneboronic
acid (X) at room temperature under nitrogen. A white solid
formed immediately. Filtration of the reaction mixture un-

der nitrogen afforded a solid with a positive flame test

15

for boron. Recrystallization from acetonitrile gave white
needles which melted at 134° with decomposition. The
derivative (XIII) is hydrOSCOpic, but quite stable in air
since its melting point stays constant over a period of
one week when exposed to air. It was found that compound
(XIII) is insoluble in many common organic solvents such
as ether, chloroform, carbon tetrachloride and benzene, but
soluble in alcohols, pyridine and hot acetonitrile. The
nmr spectrum showedam.H-C-B and H-N-C singlet at O 5.25 (1H),
an O-CHz-C triplet at O 4.35 (2H), and a C-CHz-N triplet
at 5 3.9 (2H) in D20. The results are consistent with the
structure of the dichloromethaneboronic acid derivative of
diethaneolamine (XIII). The yield of the pure compound was
70-75%.

In general, boronic esters of the type RB(OR')2 can
be made by removal of water formed in the reaction of the
boronic acid with an alcohol. The reaction of dichloro-
methaneboronic acid (X) with various alcohols can be

generalized as:
C12HCB(OH)2 + 2ROH dlStlllatlon > C12HCB(OR)2 + 2H2O
(X)

Usually, a solution of a-halo boronic acid containing ex-

 

cess alcohol and benzene was distilled under nitrogen at
atmospheric pressure to remove the water—alcohol-benzene
azeotrope. The product was collected by distillation under
reduced pressure. The results of esterifications of the
acid (X) with ethylene glycol, ethanol, gfbutanol, and

'n-propanol are summarized as follows:

16
The reaction between a-halo boronic acid (X) and

ethylene glycol can be shown as:

—2H20

 

C12HCB(OH)2 + (CHZOH)2 > C12HCB(OCH2)2
(X) (XIV)

The ethylene glycol ester obtained after removal of sol-
vent, was recrystallized from ethyl ether to give a white
solid, mp 97-990, which gave a positive flame test for
boron. The nmr spectrum showed an H-C-B singlet at O
5.3 (1H), an O-H singlet at O 4.8 (3-5H), and an O-CHz-C
singlet at 5 3.8 (4-7H) in (CH3)ZSO with variable ratios of
the respective protons. The mass spectrum proved the forma-
tion of the ethylene glycol ester(flIV) by the parent peak
at m/e 154 and an isotope peak at m/e 156 in the ratio of
intensity 1.7:1 identical to the theoretical value. Fur-
ther attempts at purification failed. It is possible that

monomer, dimer and polymer by-products are formed as follows:

 

ClZHCB(OH)2 + (CHZOH)2 diftéiéation> CleCB(OCH2)2 +
(x) CHCl2 CHC12 (XIV)
HOCHZCHZOEOCH2CH20H + H(OCH2CH20EOCH2CH20)2H

monomer dimer
CHCl2
+ H(OCH2CHZOEOCH2CH20)nH
polymer

Esterification Of the a-halo boronic acid with ethanol

was attempted as follows:

17

distillation '
ClZHCB(OH)2 + 2CH3CH20H _2H20 > CleCB(OCH2CHa)2

(X) (XV)

 

The product was Collected at 26—320/0.5-0.6 mm and showed
a positive flame test for boron. The nmr Spectrum showed
a singlet at O 5.25 (1H), a quartet at o 3.5 (2-5H), a
triplet at O 1.0 (3—7H) (neat). It was concluded that the
unreacted ethanol distilled together with the expected
ethanol ester, CleCB(OCH2CH3)2, under reduced pressure.
Attempts to isolate the ester product in good yields were
unsuccessful. However, 10-15% yield of the pure ethanol
ester which gave an nmr spectrum with correct proton ratio
(1:4:6) was obtained.

The nfbutanol ester of the a-halo boronic acid (X) was

prepared by the following equation:

distillation
-2H20

C12HCB(OCH2CH2CH2CH3)2

 

ClZHCB(OH)2 + 2CH3CH2CH2CH20H
(X)

(XVI)

The ester (XVI) was obtained at 103-104°/5—6 mm and showed
a positive flame test for boron. The nmr Spectrum showed
an H-C-B singlet at a 5.3, an O-CHz-C (of boronate (XVI))
triplet at a 4.1, an O-CHz-C (of tri-gfbqu borate) trip-
let at 5 3.8, a C-CHz-C multiplet at O 1.9 and a C-CH3
triplet at O 1.0 with variable proton ratios. The boiling
point of trienfbutyl borate, B(OCH2CH2CH2CH3)3, is found

to be 103-105°/8 mm or 114-115°/15 mm. Evidently,

18
tri-nfbutyl borate was present together with the expected
product (XVI). Further attempts to isolate the desired
ester (XVI) failed.

An ester of the boronic acid (X) with a sufficient
boiling point difference under reduced pressure from the
corresponding borate ester or the alcohol was desired.
The gfpropanol ester of the boronic acid (X) was prepared

as follows:

d's 'l at'
C12HCB(OH)2 + 2CH3CH2CH20H :2;:Ol lon> ClZHCB(OCH2CH2CH3)2

(X) (XVII)
Esterification of the boronic acid with nfpropanol gave
the nfpropanol ester (XVII) in 40% yield. The Efpropanol
ester (XVII), collected at 1030/15 mm, showed N23D 1.4230
and a positive flame test for boron. The nmr spectrum
showed an H-C-B singlet at O 5.32 (1H), an O—CHz-C trip-
let at O 4.0 (4H), a C—CHg—C sextet at O 1.56 (4H) and a
C-CH3 triplet at 5 0.95 (6H) (neat). These results are
consistent with the structure of dijg-propyl dichloro-
methaneboronate (XVII) which is soluble in almost all
organic solvents including alcohols, ethers, chloroform
and benzene. It was found that moisture in the air could
slowly hydrolyze the neprOpanol ester (XVII) to the cor-

responding boronic acid (X) as follows:

 

CleCB(OCH2CH2CH3)2 moizggre . C12HCB(OH)2 + 2CH3CH2CH20H
temperature

(XVII) (X)

Therefore, the nfpropanol ester (XVII) should be stored

under a nitrogen atmosphere.

19

Finally, preparation of di—

chloromethaneboronic acid (X), directly followed by esteri-

fication with nfpropanol afforded the ester (XVII) in an

overall yield of 55-60%.

The results of elemental analyses of boronic acid (X),

and boronate (XIII), (XVII) are not acceptable.

are probably due to:

first,

the formation of an inert

boron carbide under analytic operation conditions;

the highly hydroscopic quality of boronate (XIII) and

moisture-sensitivity of boronate (XVII).

The reasons

second,

These make the

operation of elemental analyses determination difficult.

The analytic data obtained is listed in Table (A).

 

 

 

Table (A). Results of Elemental Analyses
Formula C%’ H% Cl% N%
CleCB(OH)2 Calcd 9.32 2.33 55.12 --
(x) Found 5.52 2.84 32.21 --
ClZHCB(OCH2CH2)2NH Calcd 30.40 5.05 35.89 7.08
(XIII) Found 29.86 6.69 28.51 11.21
30.94 5.06 34.28 7.93
28.83 4.84 -- 6.11
CleCB(OCH2CH2CH3)2 Calcd 39.46 7.05 33.35 —-
(XVII) Found 38.80 6.89 34.19 —-
24.28 4.08 —- —-

 

DISCUSSION

The reactions between phenyl(trichloromethyl)mercury
and boron trihalide are unsuccessful for preparing boron
compounds of the type Cl3CBX2, according to the proposed

scheme:

[::LHgCC13 + BX3 > Cl3CBX2 + HgX

The difficulties of synthesizing C13CBX2 by this method

 

may be due to the low reactivity of the organomercury com-
pound or the instability of the product C13CBX2. Further-
more, the side-product, phenylmercuric halide is partially
soluble under the reaction conditions, thereby making work-
up difficult.

Consequently, the relatively more reactive NaCC13
Species, which can be generated by thermal decomposition
of sodium trichloroacetate was attempted. The reaction
between NaCC13 and the boron trihalide possibly produces
trichloromethylboron dihalide, Cl3CBX2, as follows:

0
II BX
NaOCCCl3 —L> C02T+ NaCCl3 —§—~> C13CBX2 + NaX

The evolution of carbon dioxide gas is observed from the

reaction, but the desired product is not obtained. It is

20

21
. . . . C3 '
assumed that carbon diox1de might react With CC13 or
:CClz to produce other compounds under the reaction
conditions.
Thus, milder reaction conditions to generate the tri-
halomethyl anion are desirable. .KObrich and Flory describe

the preparation of chloromethyllithium solution as follows:

16,17

CH3CH2CH2CH2Li >

- 1
HC XC 2 -100°,ether

LiC-XCl2 + CH3CH2CH2CH3

X=Cl, H.

The reaction between trichloromethyllithium, LiCCl3, and
boron trihalide leads to decomposition products presumably
due to the instabilities of both the C13CBX2 product and
LiCCl3. In fact, it is known that LiCCl3 is stable at
-115° but decomposes exothermally at -80° to form a mix-
ture of tetrahaloethylene.21

Dichloromethyllithium is relatively more stable than
trichloromethyllithium.16v18 Therefore, dichloromethyl-
lithium can be handled easily in preparation of d-halo-
organoboron compounds. An oil-like product (IX) is obtained
from the reaction of dichloromethyllithium and trimethyl

borate at -100° under a nitrogen atmosphere.

tetrahydrofuran>

. GD
_1000 CleCEIOCH3)3Li

 

LiCHCl2 + B(OCH3)3
(IX)

The nmr Spectrum probably suggests the reasonable structure

22
of (IX). Crystallization of the lithium salt (IX) fails
possibly because the material is highly air-sensitive or
the material may actually be an oil.
Usually, the reaction between organolithium compounds
and boron derivatives gives a mixture of products“?2 as

follows:

C>€)
RLi + BX3 > RBXZ + RzBX + R3B + R4BLi

 

However, the reaction between dichloromethyllithium and tri-
methyl borate produces mainly monoalkylation of the boronic
ester. The electron withdrawing effects of the chlorine
atoms attached to the a-carbon make the boron atom highly
electron-deficient; thus the product is isolated as a

salt with the structure ClzflCgkOCH3)3%& (IX). This salt

is presumably resistant to further alkylation by the
organolithium reagent.

It is known that the boron-carbon bond is very strong
and is not easily broken by strong acids such as hydrochloric
acid and sulfuric acid, or by sodium hydroxide at room
temperature. Therefore, hydrolysis of C12HC§XOCH3)3%E (IX)
with dilute hydrochloric acid at -100° under nitrogen
produces the desired compound, dichloromethaneboronic acid

(X), in very good yields (90-95%).

 

e 6? dil HCl H O
CleCB(OCH3)3L1 W> ClZHCB(OCH3)2 roam >
(IX) temperature
CleCB(OH)2

(X)

mp 97-980

23

The acid (X) is found to be soluble in alcohols and tetra-
hydrofuran, but insoluble in diethyl ether, chloroform,
carbon tetrachloride and benzene. The nmr spectrum of the
acid (X) shows an H-C-B Singlet at O 5.25 (1H) and B-O-H
Singlet at 6 4.88 (2-3H) in (CD3)ZSO. It indicates the
presence of boric acid, B(OH)3, together with the boronic
acid (X). In addition, the a-halo boronic acid (X) might
lose water to form the corresponding boroxine (XI) readily

and reversiblym'2° as follows:

 

 

O
-3H20 / \
3C12HCB (OH) 2 < 4 > C12HC-B B"’CHC:L§_)l
( ) 3H20 0 6
X
\ B /
CHCl2
(XI)

The boronic acid (X) in the presence of boric acid and
boroxine (XI) could explain the variable proton ratios of
the nmr Spectra.

Attempts to prepare the pure boroxine (XI) by dehydra-
tion of (X) in refluxing benzene are unsuccessful, possibly
because of the heat-sensitivity of the a-haloorganoboron
compound under the reaction conditions at 70-110°.

Most known a-halo boronic acid derivatives of catechol
and diethanolamine are solids. In order to Characterize
the dichloromethaneboronic acid (X): the catechol and di‘

ethanolamine derivatives were made.

24
,A sample of the catechol ester (XII) was prepared ac-

cording to the following equation:

CleCB(OH)2 +

OH })
tetrahydrofuran .
OH room temperaturg ClZHCQb + 2H20

(X)
(XII)

It is known that compounds such as catechyl 2-bromopropane-
2-boronate and catechyl 1-bromoethaneboronate are unstable
on storage, turning to a black liquid.4 Likewise, a pure
sample of the type (XII) could not be obtained.

However, the boronic acid derivative of diethanol-
amine (XIII) is obtained very readily. The reaction is
shown as:

tetrahydrofuran,

CleCB(OH)2 + NH(CH2CH20H)2 room temperaturg

(X)

C12HCB(OCH2CH2)2NH + 2H20.

(XIII)

70-75%
The product (XIII) melts at 134° with decomposition. It
is hydroscopic but otherwise stable in air as indicated by
its constant melting point when exposed to air. The
boronic acid derivative of diethanolamine (XIII) is insol-
uble in many common organic solvents such as ethers, Chloro-
form, carbon tetrachloride and benzene, but soluble in
alcohols, pyridine and hot acetonitrile. The easy synthesis
and high stability of the acid derivative of diethanolamine

(XIII) is probably due to the presence of the coordinatively

25

saturated boron atom as shown below:23:24

H2

CleC-B<~--NH

H2
0 \cn/
(XIII)

The nmr Spectrum of the acid derivative (XIII) shows an
H-C-B and N—H Singlet at O 5.25 (1H), an O-CHz-C triplet
at 5 4.35 (2H), and a C-CHz-N triplet at O 3.9 (2H) in
D20. These results are consistent with the structure of
the compound (XIII).

-Efforts have been made to convert the a-halo boronic
acid (X) to the corresponding esters.

distillation>

Cl HCB OH + 2ROH .
2 ( )2 -2H20

(X)

CleCB(OR)2

This reaction is reversible and water is usually removed
as the benzene azeotrope.
A solid product is obtained from the esterification

of the a-halo boronic acid (X) with ethylene glycol.

-2H20
CleCB(OH)2 +-(CH20H)2 -——————9 CleCB(OCH2)2

(x) (XIv)

Both mass and nmr Spectra Show that materials other than

the desired ethylene glycol ester (XIV) are present. At-
tempts to purify the ester (XIV) failed. The impurities

could be the unreacted a-halo boronic acid (X) which is

26
hard to remove from the ethylene glycol ester (XIV) pre-
sumably because of similar solubility. It is also likely
that monomer, dimer and polymers of the ethylene glycol

ester form under the reaction conditions.

 

CleCB(OH)2 + (CHZOH)2 d13t111at10n> ClZHCB(OCH2)2 +

- H20
(X) (XIV)
CHCl2 CHCl2
I I
HOCHZCHZOBOCHZCHZOH + H(OCH2CH20BOCH2CH20)2H +
monomer CHClz dimer
I
H(OCH2CH20BOCH2CH20)nH
polymer

Esterification of the acid with ethanol according to

the following equation

CleCB(OH)2 + 2CH3CH20H dlfgéiéatlon> CleCB(OCH2CH3)2

(X) (XV)

was attempted. Quantitative yield of the product was not
obtained. It is assumed that unreacted ethanol distilled
together with the ethanol ester, ClZHCB(OCH2CH3)2 (XV),

at 26-32°/O.5-0.6 mm. Further isolation gave only 10-15%
yield of pure diethyl dichloromethaneboronate which showed

the correct nmr spectrum.

The nfbutanol ester of the acid was prepared as follows:

 

C12HCB(OH)2 + 2CH3CH2CH2CH20H dlfgéléatlon >
2
(X)

CleCB(OCH2CH2CH2CH3)2
(xv: )

27

A product was collected at 103-1040/5-6 mm. The nmr spec-
trum indicated the presence of a mixture of difin-butyl di-
chloromethaneboronate (XVI) and trifgfbutyl borate in
variable proportions. The boiling point of tri-nfbutyl
borate is 114-115°/15 mm, which is so close to that of the
nfbutanol ester (XVI) under reduced pressure that it is
impossible to separate the mixture by simple distillation.

However, useful quantities of pure di-nfpropyl di—
chloromethaneboronate (XVII) were collected at 103°/15 mm

according to the following equation:

d'st' la '
C12HCB(OH)2 + 2CH3CH2CH20H l_2;:0 tlon >C12HCB(OCH2CH2CH3)2

(x) (XVII)
40%

Since tri—nfpropyl borate boils at 900/15 mm, separation
by distillation is more facile in this case than for the
butyl esters. The nmr Spectrum of the ester (XVII) shows
an H-C-B singlet at O 5.32 (1H), an O-CHz-C triplet at O
4.0 (4H), a C-CHz-C sextet at 5 1.56 (4H), and a C-CH3
triplet at O 0.95 (6H) (neat). These results are consistent
with the structure of the ester (XVII). The ester (XVII),
N23D 1.4230, is soluble in many common organic solvents such
as ethers, alcohols, chloroform, and benzene. Exposure of
the ester (XVII) to air does not result in oxidation but

rather in the formation of the boronic acid (X).

CleCB(OCH2CH2CH3)2 ngggfi£§> CleCB(OH)2 + 2CH3CH2CH20H

(XVII) temperature (X)

28

It is observed that a-haloorganoboronic acid (X) gives
black residue of carbon as the reaction temperature exceeds
110°. Possibly the solid a-haloorganoboron compounds are
sensitive to heat, therefore the mildest reaction condition
would be optimum for the synthesis of the a-haloorganoboron
compounds. Starting from methylene Chloride and nfbutyl-
lithium reaction, directly followed by hydrolysis and esteri—
fication with nfpropanol, the stable di-nfpropyl dichloro-
methane boronate (XVII) is obtained in 55-60% yield. This
ester may be useful for further studies. Subsequent nucleo-
philic substitution reactions of ester (XVII) could be

proposed as follows:

1. Reaction with Grignard reagents,

CleCB(OCH2CH2CH3)2 + RMgX ——s ClHCB(OCH2CH2CH3)2‘+ MgXCl
I
(XVII) R

2. Reaction with NaH,

CleCB(OCH2CH2CH3)2 + NaH ——> C1H2CB(OCH2CH2CH3)2 + NaCl
(XVII)

3. Formation of chlorocarbene,

 

C12HCB(OCH2CH2CH3)2 A > ClCH + ClB(OCH2CH2CH3)2 .

()CVII)

EXPERIMENTAL

Melting points were taken on a Thomas Hoover Capillary
melting point apparatus and are uncorrected. Nuclear mag-
netic resonance (nmr) Spectra were obtained using a Varian
A-60 spectrometer. All spectra are recorded in O units
relative to tetramethylsilane (TMS).

Elemental analyses were carried out by Crobaugh

Laboratories, Cleveland, Ohio.

A. Phenyl(trichloromethyl)mercury. Method of Schweizer

 

and O'Neill.25

 

In a 1-1 three-necked, round bottomed flask equipped
with a reflux condenser and an efficient mechanical stir-
rer were placed 400 ml of dry benzene, 48 g (0.36 mole) of
ethyl trichloroacetate and 26.4 g (0.074 mole) of phenyl-
mercuric bromide. The mixture was stirred and cooled in
an ice—water bath for 15 min. Sodium methoxide (16.8 g,
0.308 mole) was added all at once. The mixture was stirred
for 1.5 hr with cooling, then quenched with an equal vol-
ume of water. After thorough mixing, the benzene layer was
decanted and filtered. The aqueous mixture was extracted
with three 100-ml portions of benzene after which the
organic layers were combined and evaporated to dryness under

29

30
a stream of air. Phenyl(trichloromethyl)mercury (7.3 g,
50% yield), a white solid, mp 102-107°, was obtained.
After one washing with 20 ml of cold ethanol, 4.8 g (40%)
of the desired product was obtained, mp 107-1080 (lit.

mp 114-115°).25

B. Attempted Reaction of Phenyl(trichloromethyl)mercury

and Boron Trichloride

 

In a 100-ml three-necked, round bottomed flask equip—
ped with a dry-ice condenser was placed 7.3 g (18 mmole)
of phenyl(trichloromethyl)mercury. Around 20 ml (25 mmole)
of boron trichloride was allowed to evaporate into the
flask at 0° under nitrogen. The reaction mixture was stir-
red overnight at 0° and then warmed gradually to room temper-
ature. At this time, evolution of boron trichloride gas
was observed. A white solid precipitated. After addition
of 10 ml of ethylene glycol, the esterified mixture was
stirred for 1 hr at room temperature and then distilled
under reduced pressure. The distillate collected showed
the negative flame test for boron. The white solid left
in the flask melted >2500.

A Similar experiment using stoichiometric amounts of
both reagents in chlorobenzene solution was also unsuccess-

ful.

31
C. Attempted Reaction of Phenyl(trichloromethyl)mercury

and Boron Trifluoride Etherate

To a solution of 0.8 g (2 mmole) of phenyl(trichloro-
methyl)merucry in 10 ml of anhydrous ethyl ether at room
temperature was added, under nitrogen, 0.25 ml (2 mmole)
of boron trifluoride etherate. The reaction mixture was
stirred for one hour at room temperature, hydrolyzed with
5 ml of water and concentrated. Only phenylmercuric fluor-
ide with a melting point >250O was obtained.

A similar experiment in the presence of sodium iodide

as catalyst was also unsuccessful.
D. Sodium Trichloroacetate.26

Trichloroacetic acid (163.4 g, 1 mole) was put in a
2-1 Erlenmeyer flask with a side arm and cooled in an ice—
bath. After the addition of 9/10 of 150 ml of 0.15 g,
ice-cold sodium hydroxide solution, the mixture was ti—
trated with the remaining solution to a methyl orange end—
point. The flask was then covered with a rubber stopper,
connected to an aspirator and mounted on a steam bath in
order to remove water. Finally, quantitative amount of
white solids was collected and dried in a dessicator under

vacuum overnight.

32
E. Attempted Reaction of Sodium Trichloroacetate and

Boron Trichloride

In a 100-ml three—necked, round bottomed flask equip-
ped with a dry-ice condenser was placed 4.63 g (25 mmole)
of sodium trichloroacetate under nitrogen. Boron trichlor-
ide initially was collected in a trap which was immersed
in a dry-ice and acetone bath to prevent the compound
from evaporating. Around 15-20 ml (25 mmole) of boron
trichloride was evaporated into the round bottomed flask
at 0°. The reaction mixture was stirred for one—half
hour at 0° and then warmed to room temperature gradually.
A gas was evolved and identified as C02 by a positive
Ba(OH)2 test. However, the unreacted boron trichloride
evolved from the reaction mixture spontaneously at a
temperature higher than 12°. Esterification of the re-
sulting solution with 5 ml of ethylene glycol and distil-
lation under reduced pressure gave a distillate with a
negative flame test for boron. The solid left in the
flask was shown to be boric acid which melts at 180° with

decomposition, in 80% yield.

F. Attempted Reaction of Sodium Trichloroacetate and Boron

Trifluoride Etherate

In a 50-ml round bottomed flask was placed 9.27 g
(50 mmole) of sodium trichloroacetate under nitrogen. Soon
after injection of 6.3 ml (50 mmole) of boron trifluoride

etherate, the evolution of gas which is identified as

33
carbon dioxide by a positive Ba(OH)2 test, was observed.
When the evolution of C02 gas stopped, 20 ml of ethylene
glycol was added, then the mixture was stirred overnight
at 25°. Distillation under reduced pressure gave only
boric acid which melts at 180° with decomposition, in 80%

yield.

G. Trichloromethyllithium Solution. Method of Kabrich

 

and Flory16I17

 

In a 100—ml three—necked, round bottomed flask equip—

ped with a mechanical stirrer were placed 1.60 ml (20

mmole) of chloroform and 48 ml of the combined solvents

tetrahydrofuran/ethyl ether/pentane in 42121 ratio at -108°
by immersing the flask in an absolute alcohol-liquid nitro-
gen mixture in a dewar flask under nitrogen. To the solu-
tion was added dropwise with stirring at -108°, 12.6 ml

(20 mmole) of nfbutyllithium over a 15-minute period. Stir-
ring at this temperature for an additional 20 minutes rem

sulted in a colorless suspension of trichloromethyllithium.

H. Reaction of Trichloromethyllithium and Benzophenone17

‘The trichloromethyllithium solution (20 mmole) ob-
tained from the previous procedure, was treated with 3.5 g
(< 20 mmole) of benzophenone in 20 ml of ethyl ether. The
reaction mixture was stirred for 20 min, then hydrolyzed
with concentrated sulfuric acid and methanol at —108°.

Evaporation of solvents and crystallization from ice-cold

34
pentane afforded 1.65 g (30%), (lit 1.33 g, 20%), of

¢2C(OH)CCl3, mp 64-650 (lit 64.5—65.5°).

I. Attempted Reaction of Trichloromethyllithium and Boron
Trifluoride Etherate

 

 

Trichloromethyllithium solution (50 mmole) obtained
from the method of KObrich and Flory, was treated dropwise
during 15 min with stirring with 6.3 ml (50 mmole) of boron
trifluoride etherate. The colorless reaction mixture was
stirred for 30 min at —108°, then warmed gradually to room
temperature. However, a dark colored solution resulted.

It possibly indicates the decomposition of trichloromethyl-
lithium reagent or the expected product, C13CBF2.

A similar reaction was performed using trimethyl
borate instead of boron trifluoride etherate. It also

resulted in the formation of a dark colored solution.

J. Dichloromethyllithium Solution. Method of KObriCh and
Flory18

The dichloromethyllithium solution was prepared by

adding 15.75 ml (25 mmole) of nfbutyllithium dropwise over

a period of 30 min, under nitrogen, to a vigorously stirred
solution of methylene chloride (1.59 ml, 25 mmole) in 50 ml
of tetrahydrofuran at -100° obtained by immersing the con—
tainer in a dewar flask containing a mixture of absolute
alcohol-liquid nitrogen. The reaction mixture was stirred
for one—half hour at this temperature. A colorless solution

of dichloromethyllithium resulted which was stable even at

—74°.

35

K. Reaction of Dichloromethyllithium and Trimethyl Borate

An equivalent amount of trimethyl borate (5.7 ml, 50
mmole) was added to the colorless solution of dichloromethylw
lithium (50 mmole) obtained from the previous preparation,
at -100° under nitrogen. The reaction mixture was stirred
for one-half hour at -100°, then warmed gradually to room
temperature. Removal of the solvents afforded an oil—like
material. The nmr spectrum showed singlets at 5 5.25 (1H),
and O 3.3 (9-10H) which suggested the formation of a com-
pound having a reasonable structure as CleC§(OCH3)3L3.5 (IX).
However, crystallization of the oil-like material from

ethyl ether or chloroform failed.

L. Preparation of Dichloromethaneboronic Acid (X)

 

The CleC§(OCH3)3L? (IX) solution obtained from the
previous reaction, was hydrolyzed with dilute hydrochloric
acid at -100° under nitrogen. The reaction mixture was
stirred vigorously for 30 min at -100°, then warmed to room
temperature gradually. After extraction with ethyl ether
and removal of solvents, a white solid, mp 92-95°, which
gave a positive flame test for boron was collected from ice-
cold pentane. It melted at 97-98° after recrystallization
from anhydrous ethyl ether. The acid (X) is soluble in
alcohols and tetrahydrofuran but insoluble in diethyl ether,
chloroform and benzene. The nmr Spectrum showed a singlet

at O 5.25 (1H) and a Singlet at 5 4.88 (2-3H) in (CD3)2SO.

36
The crude product, dichloromethaneboronic acid (X), was ob-
tained in 90-95% yield. The results of elemental analysis
are not acceptable possibly due to the presence of the

corresponding boroxine (XI) and boric acid.
M. Derivatives of Dichloromethaneboronic Acid (X)

a. Attempted Preparation of Boroxine (XI). A mixture
of 1.29 g (10 mmole) of dichloromethaneboronic acid (X) and
a5 ml of dry benzene was distilled at atmospheric pressure
under nitrogen. The benzene-water azeotrope was collected
at 70°. The reaction mixture turned brown and then gave a
black residue of carbon as the temperature exceeded 110°.

No boroxine (XI) was obtained.

b. Acid Derivative of Catechol (XII). A solution
of 0.79 g (5 mmole) of dichloromethaneboronic acid and
0.55 g (5 mmole) of catechol in 10 ml of tetrahydrofuran
was stirred at room temperature for two hours under nitro-
gen. The reaction mixture, after drying (M9804) and re—
moval of the solvents gave a white solid which turned to
a deep purple color immediately after it was isolated.

Attempts to obtain the pure sample were unsuccessful.

c. Acid Derivative of Diethanolamine (XIII). Di-
ethanolamine (0.96 ml, 10 mmole) was added dropwise to a
clear solution of dichloromethaneboronic acid (1.29 g, 10
mmole) in 10 ml of tetrahydrofuran at room temperature
under nitrogen. A white solid formed immediately. The
reaction mixture was stirred gently for 10 min. Filtration
under nitrogen afforded 1.48 g of C12HCB(OCH2CH2)2NH (XIII)
in 70-75% yield. Recrystallization from acetonitrile gave

white needles which melted at 134° with decomposition. The

37
derivative (XIII) is insoluble in many common organic sol-
vents such as ethers, chloroform and benzene but soluble
in alcohols, pyridine and hot acetonitrile. It is hydro-
scopic but rather stable in air by its constant melting
point over a period when exposed to air. The nmr spectrum
Showed an H-C-B and H-N-C singlet at O 5.25 (1H), an
O-CHz-C triplet at 5 4.35 (2H), and a C-CHz-N triplet at
O 3.9 (2H) in 020.

N. Esters of Dichloromethaneboronic Acid (X)

a. Ethylene Glycol Ester (XIV), A solution of 1.29 g

 

(10 mmole) of dichloromethaneboronic acid and 0.73 ml (11
mmole) of ethylene glycol in 10 ml of tetrahydrofuran con-
taining anhydrous M9804 (1 teaspoon) as dehydrating agent
was refluxed for two hours under nitrogen. Removal of the
MgSO4 by filtration and evaporation of the solvent yielded
an off-white compound which melted at 90-99°. Recrystalli—
zation from anhydrous ethyl ether gave a white solid, mp
97-990.

The mass spectrum proved the formation of ethylene
glycol ester (XV) by the parent peak at m/e 154 and an
isotope peak at m/e 156 with a ratio of intensity 1.7:1,
identical to the theoretical value. The nmr spectrum
showed singlets at 5 5.3 (1H), 5 4.8 (3-5H), 5 3.8 (4-7H)
in variable proton ratios. Both Spectra indicated the un-
expected compounds present in the product. vFurther puri—

fication was unsuccessful.

38

b. Ethanol Ester (XV). A solution of 3.87 g (30
mmole) of dichloromethaneboronic acid in 20 ml of ethanol
and 15 ml of benzene was distilled at atmospheric pressure
under nitrogen. The water-ethanol-benzene azeotrope dis-
tilled at 70°. The solvents were removed by distillation
under reduced pressure and the main product, which gave a
positive flame test for boron, was collected at 26-320/
0.5-0.6 mm. The nmr spectrum showed a singlet at 5 5.25
(1H), a quartet at O 3.5 (2-5H), and a triplet at 5 1.0
(3-7H). Further isolation only gave 10-15% of the pure
diethyl dichloromethaneboronate which showed an nmr spec-

trum with correct proton ratios, (1:4:6).

c. nfButanol Ester (XVI). In a 100-ml round bottomed

 

flask equipped with a fractionating column were placed 3.87 g
(30 mmole) of dichloromethaneboronic acid and 30 ml of n:
butanol. The reaction mixture was distilled at atmospheric
pressure under a nitrogen atmosphere to give the water—g7
butanol azeotrope at 90-91°. The unreacted nfbutanol was
removed by distillation under reduced pressure until the
thermometer registered a sudden rise in temperature. The
product distilled almost entirely at 103-104/5-6 mm. How-
ever, the product, difigfbutyl dichloromethaneboronate (XVI)
(H-C-B peak in the nmr at O 5.3) constituted a variable
proportion of the by-product. tri-nfbutyl borate, which
boiled too close to that of di-nfbutyl dichloromethane—

boronate (XVI) to be removed easily by simple distillation.

39

d. EfPropanol Ester (XVII). A solution of 3.23 g

 

(25 mmole) of dichloromethaneboronic acid in 25 ml of £7
propanol was heated to gentle boiling and the water—n7
propanol azeotrope was distilled at 70° at atmospheric
pressure under nitrogen. The unreacted nfpropanol was re-
moved by distillation under reduced pressure until the
thermometer registered a sudden rise in temperature.
Initially, small quantities of the distillate were collected
at 85-970/15-16 mm, which was proved to be a mixture of
tri-nfpropyl borate and di-nfpropyl dichloromethaneboronate
(XVII) by its nmr spectrum. The pure di-gfprOpyl dichloro—
methaneboronate (XVII) was obtained at 103°/15 mm, N23D
1.4230, in 40% yields. However, the overall preparation

of the ester (XVII), starting from the reaction of di-
chloromethyllithium and trimethyl borate, followed by hy-
drolysis and esterification, gave a yield of 55-60%

based on trimethylborate. The ester (XVII) is soluble in
many common organic solvents such as alcohols, ethers,
chloroform and benzene. It also found that ester (XVII)

is slowly hydrolyzed by the moisture in air to the corres-
ponding boronic acid (X). The nmr spectrum showed an

H-C-B singlet at O 5.32 (1H), an O-CHz-C triplet at O

4.0 (4H), a C-CHz-C sextet at O 1.56 (4H) and C-CH3 triplet

at O 0.95 (6H).

1.

10.

11.

12.

13.

14.

REFERENCES

D. S. Matteson, J. Amer. Chem. Soc., 82” 4228 (1960).
L. Zeldin and P. R. Girardot, Abstracts of the 140th
National Meeting of the AmeriCan Chemical Society,
Chicago, 111., Sept. 1961, P.15N.

D. S. Matteson and J. D. Liedtke, Chem. Ind. (London),
1241 (1963).

D. S. Matteson and G. D. Schaumberg, J. Org. Chem.,
21, 726 (1966).

D. S. Matteson and R. W. H. Mah, J. Amer. Chem. Soc.,
22. 2599 (1963).

D. S. Matteson, Organometal. Chem. Rev., 1 (1966).
D. S. Matteson, J. Org. Chem., 22, 3399 (1964).

R. J. Brotherton and H. Steinber , "Progress in Boron
Chemistry", Vol. 3, p. 142 (1970).

(a) D. S. Matteson and T. C. Cheng, J. Organometal.
Chem., 6” 100 (1966).

(b) D. S. Matteson and T. C. Cheng, J. Org. Chem.,
33, 3055 (1968).

D. S. Seyferth and B. Prokai, J. Amer. Chem. Soc.,
'88, 1834 (1966).

D. J. Pasto and Sr. R. Snyder, O.S.F., J. Org. Chem.,
3_1, 2773 (1966).

H. C. Brown and R. L. Sharp, J. Amer. Chem. Soc., 99”
2915 (1968).

D. J. Pasto and J. Hickman, J. Amer. Chem. Soc., 89;
5608 (1967).

D. J. Pasto, J. Hickman and T. C. Cheng, J. Amer. Chem.
Soc., 99, 6259 (1968).

40

15.

16.

17.

18.
19.

20.

21.

22.

23.

24.

25.

26.

41
A. Winston, J. P. M. Bederka, W. G. Isher, P. C.
Juliano and J. C. Sharp, J. Org. Chem., 39, 2784
(1965).

G. KObrich and K. Flory, Angew. Chem., 76, 536 (1964);
Angew. Chem. internat. Edit., 3, 445 (1964).

G. KObriCh, K. Flory and R. Fischer, Chem. Ber.,
99(h), 1793 (1966) (Ger); Chem. Abs., 99, 7205f (1966).

G. KObrich, Chem. Ber., 99, 1782 (1966).
A. B. Burg, J. Amer. Chem. Soc., 99, 2228 (1940).

P. A. McCusker, E. C. Ashby and H. S. Makowski
ibid., 79, 5179 (1957).

W. T. Miller, Jr. and D. M. Whalen, ibid., 86, 2089
(1964).

G. E. Coats, "Organo—metallic Compounds", N Y., John
Wiley & Sons, Inc., p. 89 (1956).

O. C. Musgrave and T. 0. Park, Chem. and Ind. 1552
(1955)-.

R. I. Letsinger and I. Skoog, J. Amer. Chem. Soc.,
77, 2491 (1955).

E. E. Schweizer and G. J. O' Neill, J. Org. Chem. ,99,
851 (1963).

L. F. Fieser and D. H. Sachs, J. Org. Chem., 29, 1113
(1964).