THE REACTION OP PROPYLENE OXIDE WITH
ORGANOMAGNESIUM BROMIDES

fey
Charles Overlie Bostwick

A THESIS
Submitted to the School of Graduate Studies of
Michigan State College of Agriculture
and Applied Science
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Chemistry
1947

ProQuest Number: 10008466

All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

uest
ProQuest 10008466
Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author.
All rights reserved.
This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1346

ACKNOWLEDGEMENT
To Dr. Ralph C. Huston I wish to
express my gratitude for the many helpful
suggestions and for the never failing
encouragement that have made possible this
work.

C. 0. Bostwick

204822

CONTENTS
page
Introduction

1

Historical

2

Theoretical

6

Experimental
Reagents

12

Apparatus

13

Reaction of organomagnesfum bromides with
one or two moles of propylene oxide.
Preparation of Grignard reagents.

14

Analysis of the Grignard reagent.

15

Reaction with propylene oxide.

15

Hydrolysis

16

Analysis for bromohydrin yield.

16

Analysis for alcohol yield.

17

Calculation of percentage yields.

17

Derivatives

18

Preparation of propylene bromohydrin and
bromohydrin magnesium alcoholate from
magnesium bromide etherate.
Preparation of magnesium bromide
etherate.

20

Reaction of magnesium bromide etherate
with two mole equivalents of propylene
oxide.

20

Analysis of the bromohydrin magnesium
alcoholate.

20

Hydrolysis of the precipitate and
distillation of the bromohydrin.

21

Proof of structure of the bromohydrin.

22

page
Reaction products of diethylmagneslum
and propylene bromohydrin.
Preparation of diethylmagneslum.

23

Reaction of diethylmagneslum with two
mole equivalents ofbromohydrin.

23

Reaction of diethylmagneslum with one
mole equivalent of propylene
bromohydrin.

24

Tables

25

Discussion

33

Summary

36

References

38

INTRODUCTION
In 1907, Louis Henry reported his studies
on the reaction of propylene oxide and ethylmagneslum bromide.

Since that date, little has

been published on studies of the reaction itself
although the method has been used occasionally
to prepare the alcohol products.
The purpose of this investigation was to
study the reaction of propylene oxide with several
organomagnesium bromides.

This was done by

varying the conditions of reaction and amounts
of reagents, Investigating intermediate compounds
present before hydrolysis, and carrying out other
reactions that would throw light on the reaction
between propylene oxide and Grignard reagents.

2.
HISTORICAL
In 1907 Louis Henry (1) pointed out that
several different alcohols were possible as
products when propylene oxide reacted with ethylmagnesium bromide.

It depended on how the epoxide

split and whether or not rearrangement occured
before the product was obtained.
Hydrolysis Product Type
RMgBr + CH~CHCH 0 +
3V
2

CH^CHCH^6_ 2

CH,CCH,
3 0- 3
-> CH 5 CHCH2 0-

4

-

‘

ch3 ch2 cho-

CH^CHCHoR
3 0H
2

P

CH,CCH,
OH 3
CH3 CHCH2 0H

IA

IB
II A

k

CH3 CH2 CH0H

II b

With ethylmagneslum bromide, Henry obtained 60%
2-pentanol (Type I A reaction) on hydrolysis.
Later, Levene and Walti (2) obtained Type I A
optically active alcohols with no Walden inversion
when optically active propylene oxide was reacted
with n-propyl, iso-propyl, and phenyl magnesium
bromides•
The action of propylene oxide on ethylmagnesium bromide was studied again in 1936 by
Norton and Hass (3).

Although they obtained about

3.
11

$ alcohol (using

1

mole of epoxide to

1

mole

of Grignard reagent and then heating), it appeared
to be mainly 3-pentanol (compared to commercial
3-pentanol) indicating a Type II B reaction.

With

diethylmagneslum, however, they obtained 23%
2-pentanol (Type I A reaction).
1-Phenyl-2-propanol was prepared by Newman (4)
in 1940 in a 60$ yield by refluxing, for 20 hours,
a slight molar excess of propylene oxide with
pheny lmagne slum bromide.
A survey of the literature showed that the
majority of aliphatic monosubstituted ethylene
oxides appear to give alcohols resulting from Type
I A reaction (5).

Occasional conflicts of data

may be due to different conditions of reaction and
subsequent treatment.

Styrene oxide exhibits both

Type II A and B reactions (6 ).
In their Grignard reactions with 2,3-epoxybutane, Cottle and Powell (7) found, besides Type
I A alcohol, some of the alcohol formed by the
addition of the Grignard reagent to methyl ethyl
ketone (Type I B alcohol).

Upon further investigation,

they found that they could obtain this ketone in
large yields from the magnesium alcoholate of
3-bromo-2-butanol when It was allowed to warm to
room temperature.

Decomposition and rearrangement

4.
of* bromohydrin intermediates were previously
noted by Godchot and Gauquil (8 ) and Bedos (9).
Ketones and aldehydes formed by decomposition
and rearrangement of the magnesium alcoholates
of the halohydrins account for the alcohols
found when the Grignard reactions appear to go
by Type I B or II B addition.

This explanation

appears reasonable since all the reported reactions
of epoxides with dialkylmagneslum compounds
indicate only Type I A or II A addition even
when the reaction mixture is refluxed (3)(b)(10)(11).
Most of the published data seems to indicate
greater reactivity of epoxides with the bromomagnesium bond than with the alkyl-magnesium bond
If the temperature is kept low during addition of
the epoxide and the reaction mixture is not heated
before hydrolysis.

This greater reactivity is

Indicated by large yields of bromohydrin and small
yields of alcohol when the ratio of reactants
is

1:1

(1 2 ).

The aryl-magnesium bond reacts with greater
rapidity and the yield of alcohol may even be
greater than the bromohydrin yield with a

1:1

ratio of reactants.
The magnesium alcoholate of ethylene bromohydrin

reacts with, dialkylmagneslum to give the primary
alcoholate formed by replacement of the bromine
by the alkyl (11).

If any rearrangement of the

bromohydrin intermediate to acetaldehyde occurs,
it occurs at a much slower rate than the formation
of the primary alcohol intermediate.

With many of

the substituted ethylene oxides, rearrangement of the
magnesium alcoholate of the bromohydrin to the
aldehyde or ketone followed by other reactions,
such as addition of the alkyl-magnesium bond,
seems to occur more easily than replacement of the
bromine by alkyl on the bromohydrin intermediate

6,
THEORETICAL
Blaise (12), Ribas and Tapia (13), and
Huston and Agett (11) successively developed the
theoretical aspects of the reaction of ethylene
oxide with organomagnesium bromides,
A
(2RMgBr £ RgMg + MgBi£>) + 2 CHgCBg
(BrCH2 CH 2 0)2Mg t RgMg

(BrCHgCHg 0)j?Mg + RgMg

> (RCH2 CH2 0)2Mg t MgBr 2

(2RMgBr5? R ^ g + MgBr2) f 4CH2 CH2-» (BrCHgC^O^Mg + (HBI^CHgO^Mg
On hydrolysis, the first reaction yields
mainly bromohydrin, the second (after refluxing
with benzene) mainly alcohol, and the third large
amounts of bromohydrin and alcohol.

analysis of

the precipitated intermediate indicated formation
of (BrCH2 CH2 0)2Mg (11).

While the oxonium type

intermediates, as proposed by u-rignard (14) and
supported by Meisenheimer (15), may play some
part in the reaction, they probably exist but
for a brief instant.
For the reaction of propylene oxide with
magnesium bromide etherate, an intermediate
similar to that formed with ethylene oxide, has
been found to exist.

A

MgBrg + 2 CHgCHCHg

ch3
-> (BrCHgCHO)gMg

The magnesium alcoholate of the hromohydrin

seemed to be rather soluble in the presence of
excess magnesium bromide as evidenced by the
necessity of having to add over half of the
theoretical amount of propylene oxide to the
magnesium bromide before precipitation occured.
With all of the organomagnesium bromides, no
precipitation took place when the ratio of reactants
was 1:1.

When the bromohydrin intermediate was

prepared by adding one or two moles of propylene
bromohydrin to one mole of diethylmagneslum in
ether and benzene, precipitation occured immediately.
Both heat and excess magnesium bromide (in
ether - benzene solutions) caused decomposition of
the magnesium alcoholate of the bromohydrin (Table I).
Hydrolysis of the dry precipitate, by acid
and strong and weak bases, was also studied and
compared with pure distilled bromohydrin (Tables
II and III).

When the dried bromohydrin magnesium

alcoholate was heated in a dry nitrogen atmosphere,
it first melted and then acetone, HBr, and what
appeared to be CH^CH^CHBr (from its boiling point
and refractive index), distilled from the flask.
Some tar was also formed.
air gave pure MgO.

heating to 500°C in

The following mechanism is

proposed for the decomposition of the propylene
bromohydrin magnesium alcoholate on heating.

Polymerization of* the intermediate carbonium ions
may occur.
9h 3
<?h3
CH,
9 H3
BrGH 2 CH 2 0Mg0CH2 CH2Br * BrCH 2 CH2 0Mg0CH 2 CH2+
CH 3
ch 3
^
BrCHgCHgOMg :0-C-CH2
BrCHgCHg*

“f

^

:Br:

ch 3
ch 3
Vo-C-GHg * 0=CCH

OMg

<?H3

Ir BrCH=CH
4*
H* +
:Br:
_

* •

HBr

The mechanism proposed for the reaction of
propylene oxide with organomagnesium bromides
follows.

It is felt that a large shift in the

Schlenk equilibrium is not necessary.
ch3
(RMgBr * R2Mg + MgBr2 ) + CH3 CvHGH 2 * (BrCHgCHO) 2Mg
ch3
(RCH2 CH0)2Mg
gH3
ch 3
BrCH 2 CHOMgOGHCH2R
ch 3
BrCHgCHOMgR
ch 3
RCHgCHOMgR
RMgBr
R2Mg
MgBr 2
When the ratio of reactants is 1:1, for alkylmagnesium bromides, the first, fourth, and seventh
Intermediates are ^resent in large amounts.

The

9.
unreacted magnesium bromide present is equivalent
to the amount of alkyl which reacted.
ratio is

1:2

When the

for the more reactive alkylmagnesium

bromides, the first three intermediates are
present in relatively large amounts.
Upon standing for a length of time, the
intermediates, present in the 1:1 ratio Grignard
reaction mixture, tend to rearrange and produce
more secondary alcohol at the expense of the
bromohydrin.

It may be due to reaction of the

dialkylmagnesium and the bromohydrin magnesium
alcoholate or more likely, the rearrangement
of one of the other intermediates such as the
following.

CH3
BrCHgGHOMgR

CH3
-► RGHgdHOMgBr

Evidence of this was found in the reaction of
one mole of propylene bromohydrin with one mole
of diethylmagnesium over a period of four days
to give some 2-pentanol on hydrolysis.
increase In secondary alcohol yield,

A slight
at the

expense of bromohydrin yield, was noticable
when the

1:1

reaction mixture of ethylmagnesium

bromide and propylene oxide was allowed to stand
2

days at room temperature as compared to 5-g hours

cooled In Ice (Table IV).
Before the reaction mixture of one mole of

10.
ethylmagnesium bromide and two moles of propylene
oxide could set to a hard solid (after

20

hours),

it was separated and dried to a white powdery
solid.

On analysis it appeared to be composed

of either of two mixtures, the latter being
favored by the Schlenk equilibrium.
ch 3
2 (BrCH2 GH0)2Mg

ch3
ch 3
2 BrCH 2 CH0Mg0CHCH 2 C2 H 5
or

CH,
1 (C2 H 5 CH 2 CH0)2Mg

ch3
1 (BrCH2 CH0)2Mg

Analysis of dry precipitate. Calc*d: Br, 39.1, Mg, 9.23
Found : Br, 36.0, Mg,10.07
Analysis of hydrolyzed dry precipitate.
Moles calcTd: Bromohydrin, 4 ; 2-pentanol, 2
Moles found : Bromohydrin, 3.5; 2-pentanol, 2
However, when the reaction of the magnesium
bromide alcoholate of

2 -pentanol

(prepared from

the alcohol and ethylmagneslum bromide) with

*

propylene oxide was carried out, no precipitate
was obtained, but a stiff gel.

On hydrolysis,

a 60% yield of bromohydrin was obtained.
Evolution of gases, among which was identified
the unsaturated hydrocarbon corresponding to the
Grignard reagent used, were noticable with all of
the

1:2

ratio reactions and some of the

reactions.

1;1

In most cases considerable amounts

were still being evolved after 2 days.

Cooling

11.
with Ice stopped the evolution of gases.

Where

these gases were observed in greatest amounts,
the yield of bromohydrin was usually abnormally
low.

This evidence indicates some unknown

reaction between the unreacted organomagnesium
and perhaps the bromohydrin magnesium alcoholate.
This reaction does not increase the alcohol yield
as the organic radical is lost in the gaseous
products♦

12.
EXPERIMENTAL
Reagents;
Propylene Oxide (B.P. 33.4-34.4°C)
Dow Chemical Co.

Furnished by

Used without further

purification.
Organic Bromides.

Eastman Kodak Co. and Columbia

Chemicals Co.

Dried over anhydrous CaClg for

at least one week and fractionated slowly
through a glass helix packed column with a
total reflux, partial take-off,head.

Purified

bromides distilled over a ^ to 1°C range.
Stored in glass stoppered amber bottles in
the dark.
Magnesium.

Dow magnesium turnings for Grignard

reactions•
Anhydrous Diethyl ether.

Dried over sodium wire

for at least one week.
Benzene.

Thiophene free and anhydrous.

Distilled

to remove moisture for Grignard reactions.
Bromine.

Dried by shaking with concentrated HgS04 .

DIoxane.

Dried over sodium.

Sodium Sulfate.
Ammonium Bromide.

Anhydrous, C.P.
C.P.

Silver Nitrate Solution,

.1 N.

Analytical Grade AgNO^.

Prepared from

13.
Potassium Thiocyanate Solution, about .1 N.
Standardized against ,1 N. AgNO^.
Hydrochloric Acid Solution, about .2 N.
Standardized against .2 N. NaOH.
Sodium Hydroxide Solution, about .2 N.
Standardized against sulfamic acid.
Apparatus:
Por the preparation of Grignard reagents and
magnesium bromide etherate, reactions with propylene
oxide, and NaOH reaction with the bromohydrin, a
2

liter, three necked, round bottom flask, fitted

with a condenser, mercury seal stirrer, and a
Hershberg dropping funnel or separatory funnel, was
used.

A nitrogen gas inlet was provided for all

reactions except reaction with NaOH solution.

All

Grignard reagents and magnesium bromide etherates
were protected from atmospheric moisture and COg
by CaClg and soda-lime tubes.
Por all distillations a heated glass helix
packed column 12-| in. long was used.

The head was

of the total reflux, partial take-off, type.
pot was heated with a Glas-col mantle.

The

14.
REACTION OP ORGANOMAGNESIUM BROMIDES WITH
ONE OR TWO MOLES OP PROPYLENE OXIDE.
Preparation of Grignard reagents:
One mole of* redistilled organic bromide was
mixed with 150 ml. of* anhydrous ether.
grams of* magnesium and
In the reaction flask.

100

Twenty-six

ml. of ether were placed

Ten to fifteen milliliters

of the bromide - ether solution were added and the
magnesium stirred until reaction started.

The

aromatic bromides sometimes required refluxing to
start the reaction.

Two hundred and twenty-five

milliliters of ether were then added to the reaction
and the remainder of the bromide - ether solution
added dropwise.

The Grignard reagent was stirred

for over an hour after all of the bromide was In
and then allowed to stand under a nitrogen atmosphere
over-night.
The aryl and normal alkyl Grignard reagents
were prepared by adding the bromide mixture to the
uncooled magnesium and ether as fast as the condenser,
and ice placed on top of the flask, would allow.
The yield of Grignard reagent was over 90%.
Secondary and iso-alkyl Grignard reagents
were prepared by adding the ether - bromide mixture
more slowly and not allowing the reaction mixture

15.
to get above 35°C.

The yield of Grignard reagent

with Iso-propyl, sec-butyl, and iso-butyl bromide
was between 85 and 90$.
Tert-butylmagnesium bromide was prepared in
50 to 55$ yields from the same proportion and
quality of reagents.

The reaction mixture was

cooled in a mixture of salt and ice and the
ether - bromide solution added very slowly.
Analysis of the Grignard reagent:
The Grignard reagent was forced through a
glass wool plug from the unreacted magnesium by
nitrogen pressure.

Its volume was measured in a

500 ml. graduated cylinder.
Five milliliters were pipeted out and analyzed
for the organomagnesium content by Gilman!s
procedure (17).
One milliliter was pipeted out and analyzed
by the Volhard method for bromide ion (18).
Reaction with propylene oxide:
The Grignard reagent was poured, under a stream
of nitrogen, into the reaction flask filled with
nitrogen.
When the ratio of reactants was to be 1:1,
the amount of propylene oxide was based on the
bromide ion titration.

16.
The sum of the number of moles of bromide ion
and organic radical was used for calculation of
the grams of propylene oxide to be used where the
desired ratio was

1

:2 .

The Grignard reagent was cooled in salt and
Ice and a mixture of propylene oxide and an equal
volume of ether added slowly.

The salt and ice

bath was allowed to warm up to room temperature
and removed the next day.
When the ratio of reactants was 1:1, the
reaction was allowed to stand for 2 days.
the

1:2

With

ratio reactions, the reaction mixture

was allowed to stand until the test with Michlerfs
ketone was faint or negative.
Hydrolysis:
The reaction mixture was cooled with ice and
then 150-200 ml. of saturated ammonium bromide
solution was added dropwise (19).

The ether

solution was decanted from the precipitated
magnesium salts and dried over anhydrous Na^SO^
Water was added to the magnesium salts until they
were pasty.

This paste was extracted with benzene

to obtain any remaining product.
Analysis for bromohydrin yield:
The dried hydrolysis products were refluxed

17.
with a mixture of 40 g. NaOH (stirred i hour)
and 150 ml. of water (added slowly) for 1 hour,
with vigerous stirring.
and the layers separated.
extracted with benzene.

The mixture was cooled
The water layer was
The benzene and ether

extracts were dried over anhydrous Na^SO^..

The

NaOH layer was heated to boiling to expell excess
benzene, cooled, and diluted to 1 liter.

A 5 ml.

aliquot portion was titrated for bromide ion by
the Volhard method.
Analysis for alcohol yield:
The ether - benzene solution of products was
distilled at atmospheric pressure until all of the
ether and nearly all of the benzene was removed.
Por most of the alcohols, reduced pressure was used
to prevent decomposition.
Calculation of percentage yields:
The bromohydrin yield was found by taking the
ratio of the moles of bromide ion found in the
analysis of the hydrolyzed reaction mixture to the
moles of bromide ion found in the unreacted Grignard
reagent•
The alcohol yield was calculated by taking
the ratio of the number of moles of alcohol
distilled to the number of moles of organic radical

18.
present in the unreacted Grignard reagent.
Derivatives:
The 3,5-dinitrohenzoates were made by using
pyridine and 3,5-dinitrobenzoyl chloride (2 0 ).
Another method was developed for any type of alcohol
(primary, secondary, tertiary) or phenol which made
use of ethylmagnesium bromide.

The alcohol must

not contain any contaminant that forms other
alcoholates with the Grignard reagent.

Care must

also be exercised not to oxidize the Grignard
reagent before reaction with the alcohol.

After

the slight excess of alcohol had been added to the
cooled Grignard reagent (evolution of ethane ceased),
i to 1 g. of 3,5-dinitrobenzoyl chloride (or other
acid chloride) was added.

The tube was stoppered

with a CaClg tube and allowed to stand over-night
(or longer - if the alcoholate was slow in reacting).
ROH

♦

ROMgBr

CgHgMgBr
+

R'COCl

■»
*

ROMgBr
MgBrCl

+

CgH^*
+

R'COOR

Water was added to precipitate the magnesium salts
and the ether layer filtered off and evaporated.
If the derivative was not too soluble in ether,
the residue was extracted with some other hot solvent.
The ester was then handled In the usual manner.
Alpha-naphthylurethanes were prepared from

19.
alpha-naphthyl Isocyanate using a drop of trimethylamine - ether solution

as catalyst (20).

As further proof, the secondary alcohols were
oxidized to ketones with a saturated solution of
KMnO^ in 1 to
2

6

N. HgSO^ and filtered.

The

.4-dinitrophenylhydrazones were made by adding

this filtrate to 3 ml. of a solution of 2.4 g. of
2

.4-dinItrophenylhydrazine in a mixture of 80 ml.

of water and 40 ml. of 72% perchloric acid (2 1 ).
The semicarbazones were made by separating the
ketone from the water and using the method for
water insoluble ketones (2 0 ).

20.
PREPARATION OP PROPYLENE BROMOHYDRIN AND
BROMOHYDRIN MAGNESIUM ALCOHOLATE PROM MAGNESIUM
BROMIDE ETHERATE.
Preparation of* magnesium bromide etherate;
Twenty-nine grams of magnesium were placed in
the reaction flask and covered with 500 ml. of ether.
One hundred sixty grams (26 ml.) of bromine was added
dropwise.

After all of the bromine had been added,

the reaction mixture was refluxed for an hour and
allowed to stand over-night.

The magnesium bromide

solution was removed from the excess magnesium
by nitrogen pressure through a glass wool plug,
measured, and titrated for bromide ion by the
Volhard method as described under the analysis of
the Grignard reagents.
Reaction of magnesium bromide etherate with 2 mole
equivalents of propylene oxidei
The magnesium bromide solution was poured into
a flask filled with nitrogen and then cooled in
a salt and ice bath.

A mixture of the propylene

oxide and an equal volume of ether was then added
slowly.
Analysis of the bromohydrin magnesium alcoholate;
Two hours after addition of all the epoxide,

21.
the precipitate was placed in centrifuge bottles,
stirred with anhydrous ether, and centrifuged.
The ether layer was decanted and the process of
washing repeated.

The precipitate was dried in a

vacuum desiccator over anhydrous CaClg for several
days at 12 mm. or less.

The amount of bromine

was determined by the Parr Bomb method (22), and
by NaOH reaction and Volhard titration.

Ignition

of the precipitate in a muffle furnace at 500°G
left MgO, from which the percent of magnesium was
calculated.
Calc'd for CgHgBrgMgOg : Br, 53.2; Mg, 8.1
Pound
: Br, 51.7; Mg, 8.5
Hydrolysis of the precipitate and distillation of
the bromohydrin:
Two hours after addition of the propylene
oxide to the magnesium bromide, the reaction mixture
was cooled in ice and 150 to 200 ml. saturated
ammonium bromide solution added slowly.

The ether -

benzene solution was decanted from the precipitated
magnesium salts and dried over anhydrous Na 2 S 0 4 .
The ether and fractions up to the boiling point
of the benzene were removed.

The remainder of the

benzene was taken off at reduced pressure and the
propylene bromohydrin fractionated at 12 to 15 mm.

22.
Proof of structure of the bromohydrin:
Oxidation, with a saturated KMnO^ solution in
6
2

N# HgSO^, gave a derivative which formed a
,4-dinitrophenylhydrazone•
Conversion of the bromohydrin to acetol,

by the method of Levene and Walti

(

structure to be 1-bromo-2-propanol#

16

) >

proved the

Acetol

derivatives were made and checked the literature
values for their melting points#

23#
REACTION PRODUCTS OF DIETHYLMAGNESIUM AND
PROPYLENE BROMOHYDRIN.
Preparation of diethylmagneslum:
Two moles of ethylmagnesium bromide were
prepared in the usual way.

The bromine containing

compounds were precipitated by slowly adding 175 ml#
dioxane mixed with

200

Grignard reagent (23).

ml. of benzene to the cooled
The heavy precipitate was

separated from the solution of diethylmagnesium by
centrifuging.

The strength was determined by the

Gilman procedure (17).
Reaction of diethylmagnesium with two mole
equivalents of bromohydrin:
The calculated amount of bromohydrin was mixed
with

200

ml# of benzene, placed in the reaction

flask, and cooled with ice.

The diethylmagnesium

solution was added from the dropping funnel.

The

resulting white solid (which formed immediately)
was washed with ether by centrifuging and dried in
a vacuum desiccator under reduced pressure.

The

analyses for bromine and magnesium were carried
out as before.
Calc'd for CJH^Br^MgOp : Br, 53.2; Mg, 8.1
Found
: Br, 51.2; Mg, 8.5

24.
Reaction of diethylmagnesium with one mole
equivalent of propylene bromohydrin:
The diethylmagnesium solution was placed in
the reaction flask and the bromohydrin, mixed
with an equal volume of benzene, was added slowly.
A white precipitate formed.

The reaction mixture

was allowed to stand 4 days, refluxed
and hydrolyzed.

hours,

Part of the bromohydrin was recovered

along with some alcohol.
alcohol gave

6

2 -pentanone

On oxidation, the
which was checked by

a mixed melting point of its

2

,4 -dinitrophenyl-

hydrazone with a known derivative.

25.
Table I
0
Moles MgBz*o Moles CH^C/HCHo Temperature Time
1

ii

1

2

1

2

1

2

Reflux,CgHg 7 hours
Reflux,C 6 H 6 7 hours
Reflux,CgHg 4 hours
Ice bath
2 hours

/^Bromohydrin
24
58
64
80

Table II
CH 3
Stability of dried (BrCI^CHOjgMg towards various bases.
Treatment before titration by Volhard method
Dissolved in HNCU
Dissolved in
water and HN0 3 added
Dissolved in
NaOH, boiled, cooled, HNOgadded
Decomposed by Parr Bomb method
Theoretical

^Bromine
8.0

13.0
51.2
51.1
53.2

Table III
Stability of propylene bromohydrin towards various bases.
Treatment before titration by Volhard method
Dissolved
Dissolved
Dissolved
Dissolved

in
HN0~
inwater, heated, cooled, HNO^ added
in water and MgO,
HNOg added
in NaOH,heated, cooled, HNO,, added
Theoretical

^Bromine
7.0
12.0
56.8
57.0

26.
Table IV
Reaction of 1 mole of ethylmagnesium bromide
and

1

mole of propylene oxide under various conditions.

Conditions

Time

1.Grignard reagent mixed with equal
volume of mesitylene and cooled in
ice and salt during addition.
Refluxed.
Ether distilled off.
Temperature rose rapidly from
80 to 150°C.
Cooled quickly.

2

^Bromo%
hydrin Alcohol

1g-

hour
hours
1 ^ hour
0

23.*

10.

22.*

4 hours

59,

9 .*

4.Grignard reagent heated to reflux
with equal volume of benzene added
during addition of epoxide.
2 hours
Kept warm.
3 hours

56.

14 •*

5.Grignard reagent cooled in ice
and salt during addition.
Kept in ice bath.

68

2.Grignard reagent cooled in ice
during addition.
Ether distilled off, replaced
with benzene and refluxed.
Room temperature.
3.Grignard reagent cooled in ice
and salt during addition.
Refluxed with an equal volume
of benzene added.

6.Grignard reagent cooled in ice
and salt during addition.
Allowed to come slowly to room
temperature and remain.

2

hours

6 hours
17 hours

2

hours

lj hour
4 hours
2

hours

2

days

.

5.

62.

13.

* Includes tertiary alcohol

27
Table V
Reaction of 2 moles of propylene oxide with 1
mole of ethylmagnesium bromide under various conditions.
Conditions

Time

^Bromo%
hydrin Alcohol

l.Grignard reagent cooled in ice
and salt during addition,
1 ^ hour
Ice removed and mixture allowed
to warm up and reflux gently
li hour
without applied heat.

80•

45.

2.Grignard reagent cooled in ice
and salt during addition.
Kept in ice.
Allowed to warm to room
temperature and refluxed gently
with applied heat,

4i hour

74.

55.*

3.Grignard reagent cooled in ice
and salt during addition.
Kept in salt and ice.

li hour
2 hours

76.

15

76.

54.

4.Grignard reagent cooled In ice
and salt during addition.
Cooling bath allowed to thaw
and come slowly to room
temperature•

1-g hour
2 hours

li-

.

hour

li days

* Includes some tertiary alcohol.

28.
Table VI
Yields of alcohols formed by the reaction of
propylene oxide with organomagnesium bromides.

Grignard reagent
prepared from

A

,0s

2 CH 3 CHCH 2
RMgBr + CHgCHCHg RMgBr
2 days
% yield % yield Time % yield % yield
Bromo­ Alcohol days Bromo­ Alcohol
hydrin
hydrin

Ethyl bromide

62

15

2

76

54

n-Propyl bromide

69

4

6

74

51

iso-Propyl bromide

50

7

7

76

58

n-Butyl bromide

67

5

8

70

56

sec-Butyl bromide

62

4

21

62

51

iso-Butyl bromide

64

4

25

28

15

tert-Butyl bromide

62

4

44

52

15

Phenyl bromide

59

47

1

74

67

55

58

Mesltyl bromide

1

29.
Table VII
Physical constants of the alcohols.
Alcohol

r,20
nD

B.P. °C

Ref.

1-Bromo-2-propanol

1.4801

49.6 (1 2 m m . )

2-Pentanol

1.4068

118.8 (745mm.)

2-Hexanol

1.4155

139.5 (740mm.) 25

4-Methyl-2-pentanol

1.4120

68

(52mm. )

26, 27

2-Heptanol

1.4214

77

(24mm.)

27

4-Methyl-2-hexanol

1.4223

85.5 (44mm. )

28

5-Methyl-2-hexanol

1.4227

73

(32mm. )

29

4

,4 -Dimethyl- 2 -pentanol 1.4248

65

(40mm.)

30

1

-Phenyl- 2 -prop anol

1.5221

95

(7mm.)

4,

1-Mesityl-2-propanol

1.5282

137

(9mm.)

16, 24
3

6

, 27

30.
Table VIII
Derivatives of the alcohols and corresponding ketones.
M.P. °0

Ref.

1-Bromo-2-propanol
3,5-dinitrobenzoate
alpha-naphthylurethane

85-87
116-116.5

24
a

2.4-dinitrophenylhydrazone 128

a

2-Pentanol
3.5-dinitrobenzoate
alpha-naphthy lure thane

59-61
75

2.4-dinitrophenylhydrazone 143-145

3

27
27 &

2-Hexanol
3.5-dinitrobenzoate

36-37

2.4-dinitrophenylhydrazone 106-108

31
27 •&

4-Methyl-2-pentanol
3.5-dinitrobenzoate
alpha-naphthy lure thane
2.4-dinitrophenylhydrazone

61-62
94-95.5

31
2 a

91-92

27 *

2-Heptanol
3

.5 -dinitrobenzoate

2

.4 -dinitrophenylhydrazone

47.5-48.5

27

72.5-73.5

-*

4-Me thy 1-2-hexanol
3

.5 -dinitrobenzoate

semicarbazone

62.5-63.5

28 a

128-129

28 #

a

5-Methyl-2-hexanol
3

.5 -dinitrobenzoate

34-36

2

,4 -dinitrophenylhydrazone

94-96

27

31.
M.P. °C

Ref.

48-50

30

4,4-Dimethyl-2-pentanol
3.5-dinitrobenzoate
1

-Phenyl- 2 -propanol
alpha-naphthylurethane
s emi c arb a zone

88-89.8
193

6

27

1-Mesityl-2-propanol
3.5-dinitrobenzoate
alpha-naphthylurethane

153.8-154.8
114.8-115.2

semi carbazone

206-206.5

a
a

■fr Checked by a mixed M.P# with a known derivative,
a Analyzed for nitrogen content (Table IX).

32.
Table IX
Analyses of new derivatives for nitrogen
toy the Micro Kjeldahl method (32).
1o N Calc *d

% N Found

1-Bromo-2-p ropanol
alpha-naphthylurethane
1

4.54

4.33

2.4-dinitrophenylhydrazone 17.66

17.56

-Bromo- 2 -p rop anone

4-Methyl-2-pentanol
alpha-naphthylurethane

5.16

4.96

9.03

9.09

9.03

8.97

7.42

7.36

3.97

3.84

4-Methyl-2-hexanol
3. 5 -dinitrotoenzoate
5-Methyl-2-hexanol
3. 5 -dinitrotoenzoate
1-Mesityl-2-propanol
3

.5 -dinitrobenzoate

alpha-naphthylur ethane

33.
DISCUSSION
Attempts to bring about the interchange of*
the organic Grignard radical for the bromine on
the bromohydrin intermediate by heating (Table IV),
caused only a slight Increase in alcohol percent,
this increase being largely made up of tertiary
alcohol (as evidenced by boiling point and refractive
Index).

Excessive heating caused violent reaction

with total loss of bromohydrin and only a slight
increase in total alcohol percent (Table IV,
reaction 1.).

The tertiary alcohol probably

resulted from the reaction of the Grignard reagent
with the acetone, formed from decomposition of the
bromohydrin Intermediates.
With a 1:2 ratio of reactants, many of the
alkylmagnesium bromides required considerable
time for complete reaction at room temperature,
as evidenced by a positive M ic hler^ ketone test
(17).

Tert-butylmagnesium bromide, for example,

gave a positive test after standing for a month.
On the other hand, phenylmagnesium bromide was
completely reacted in 20 hours.
the

1:2

In all cases,

ratio of reactants gave a very stiff gel

or a very hard solid, the time depending on the
reactivity of the Grignard reagent.

34.
Some uncertainty in the percent yield of
bromohydrin, reported for the various Grignard
and magnesium bromide reactions, is indicated in
the data of Table II.

As shown in the titration

of the bromohydrin magnesium alcoholate for
bromide ion, after placing it directly into acid,
a considerable amount of bromohydrin was lost to
the basic magnesium compounds even under the best
conditions of acid hydrolysis.

The use of a

saturated ammonium bromide solution, for hydrolysis
of the Grignard and magnesium bromide reactions,
prevented a greater loss of bromohydrin.
As some doubt may be expressed concerning the
addition of ammonium bromide during hydrolysis of
the Grignard reaction mixture, It should be pointed
out that In the case of the zero percent yield of
bromohydrin (Table IV), and the small percent yields
of bromohydrin from magnesium bromide under certain
conditions (Table I), the analysis was carried out
in the same manner without introducing any bromine
from this source.
In analyzing the hydrolyzed Grignard for
bromohydrin percent, sodium hydroxide reaction,
followed by Volhard titration, was used due to the
instability of the bromohydrin at elevated distillation

35.
temperatures♦
In all or the distillations, propylene oxide
and acetone were noticable.

The former was all

attributed to the hydrolysis of the bromohydrin
with sodium hydroxide and the latter to the same
source and also from the decomposition of the
bromohydrin magnesium alcoholate.

The acetone

was checked by a mixed melting point of its
2.4-dinitrophenylhydrazone.

There was also some

evidence of some unsaturated bromide.

All these,

and any water that remained, were removed by low
boiling azetropes formed with benzene.

Mesityl

oxide (identified by the melting point of its
2

.4 -dinitrophenylhydrazone addition compound)

and other unsaturated ketones caused difficulty
in separation of some of the alcohols.

With

several Grignard reactions, a wax-like product
was precipitated out of the residue by adding
petroleum ether.

It had a softening point

of 62 - 63°C.
The bromohydrin yield did not vary greatly
in the

1:1

ratio reactions of the alkylmagnesium

bromides, since they were of nearly the same order
of reactivity.

The few abnormally low yields of

bromohydrin were probably due to higher room

36.
temperatures and consequently greater decomposition
of bromohydrin intermediate (to acetone and

1 -bromo

propene?) by the unreacted organomagnesium (which
also decomposes).

Very little decomposition of

this type can occur if the organomagnesium reacts
quickly.

37*
SUMMARY
1.

Analysis of the precipitate formed from

magnesium bromide and propylene oxide or from
1

-bromo- 2 -propanol and di ethylmagnesium,

9h 3

indicated formation of (BrCHgCHO)gMg •
2*

The intermediate reaction products in the

reaction of

1

mole of propylene oxide with

1

mole

of alkylmagnesium bromide, were predominantly of
the bromohydrin-alcoholate type.
3*

Heat or time caused no appreciable reaction

of bromohydrin magnesium alcoholate with organo­
magnesium bond to give substitution of the organic
radical for the bromine on the bromohydrin
intermediate•
4.

Excessive heating of the bromohydrin

alcoholate in the presence of Grignard reagent,
excess magnesium bromide, or in the dry state,
caused formation of acetone.
1

In the dry state,

-bromo-propene, HBr, and MgO were also formed.

When Grignard reagent was present, tertiary
alcohol was obtained by reaction with the acetone.
5

.

All of the alkylmagnesium bromides gave

poor (4-15$) yields of methyl alkyl carbinol when
the ratio of reactants was

1

:1 , the major product

(50-70$) being 1-bromo-2-propanol.

With a 1:2 ratio

of reactants, the normal alkyl Grignard reagents
gave methyl alkyl carbinol yields of about 55%,
iso-propyl and sec-butyl yields of about 35%, and
iso- and tert- butyl yields of about 15%.

These

reactions also gave large yields of bromohydrin.
Arylmagnesium bromides gave carbinol yields of
about 50% or more, even with a 1:1 ratio of
reactants.
6

.

Esters of phenols and alcohols of all types

(primary, secondary, and tertiary) were made by
addition of a slight excess of the alcohol to cold
ethylmagnesium bromide, followed by reaction (in
the cold) with an acid chloride.

39.
REFERENCES
1.

Henry, Compt.rend.,145,453(1907)

2.

Levene and Walti, J.Biol.Chem.,90,81(1931);94,367(1931)

3.

Norton and Hass, J. Am.Chem.Soc.,58,2147(1956)

4.

Newman, J.Am.Chem.Soc.,62,2295(1940)

5.

Ribas, Anales soc.espan.fis.quim.,26,122(1928)
Tapia & Hernandez, Anales soc.espan.fis.quim.,28, 691(1930)
Fourneau and Tiffeneau, Bull.Soc.Chim. (4) 1,1227
Ribas & Tapia, Anales soc.espan.fis.quim. ,28,636(1930)
Mangrane and Cottle, J.Am.Chem.Soc.,64,484TT942)

6

.

7.
8

.

9.

Golumbic and Cottle, J.Am.Chem.Soc.,61,996(1939)
Kharasch and Clapp, J.Org.Chem.,3,35^X1938)
Tiffeneau and Fourneau, Compt.rend.,146,697(1908)
Cottle and Powell, J.Am.Chem.Soc.,58,2267(1936)
Godchot and Cauquil, Compt.rend.,186,575,955(1928)
Bedos, Compt.rend.,189,255(1929)

10. Bartlett and Berry, J.Am.Chem.Soc.,56,2683(1934)
11. Huston and Agett, J.Org.Chem.,6,123(1941)
12. Blaise, Compt.rend.,134,551(1902)
13. Ribas 8c Tapia, Anales soc.espan.fis.quim.,50,944(1932)
14. Grignard, Bull.Soc.Chim. ,29, 944(1903)
15. Meisenheimer, Ber.,54,1655(1921)
16. Levene and Walti, J.Biol.Chem.,68,415(1926)
17. Gilman, Wilkinson, Fishel, and Meyers,
J.Am.Chem.Soc.,45,150(1923)
18. Willard and Furman, "Elementary Quantitative Analysis"
3rd ed., D. van Nostrand Co.,Inc., New York,1940
19. Fieser, "Experiments in Organic Chemistry", Part II
2nd ed., D.C.Heath and Co., Boston, 1941

40.
20. Shriner and Fuson, "identification of Organic
Compounds", John Wiley and Sons, New York, 1940
21. Duke and Witman, Senior Problem Report,
Michigan State College, East Lansing, Mich., 1946
22. Fisher, "Laboratory Manual of Organic Chemistry"
John Wiley and Sons, New York, 1938
23. Noller, J.Am.Chem.Soc.,55,640(1931)
Noller and White, J.Am.Chem.Soc.,59,1354(1937)
24. Rust and Vaughan, J.0rg.Chem.,7,491(1942)
25. Whitmore, Rec.Trav.Chim. ,57, 566(1938)
Whitmore, Popkin, Whitaker, Mattil, Zech,
J.Am. Chem. Soc.,60,2458(1938)
26. Dupont and Darmon, Bull. Soc.Chim. , 6>,1208(1939)
27. Heilbron, "Dictionary of Organic Compounds"
Oxford University Press, New York, 1943
28. Cymerman, Heilbron, Jones, J.Chem.Soc.,page 90(1945)
29. Tuot, Compt.rend.,202,1339(1936)
30. Whitmore and Homeyer, J. Am.Chem. Soc. ,55,4194 (1935)
31. Sutter, Helv.Chim.Acta, 21,1266(1958)
32. Niederl and Niederl, "Micromethods of Quantitative
Organic Analysis", 2nd ed., John Wiley and Sons,
Inc., New York, 1942