THE REACTION OF HGNOCIIJ.()ROl<ETHYL b'THER " WITIl ORGANOllETAl.LlC COMPOUNOS

by

Philip Jen-chlen T~g

Thea16 submitted to t he Faculty ot the

Virginia Polytechnic Institute

in candidacy for the degree ot

KASTER uF SCIENCE

in

Chemistry

APPfcOVUl , APPRlVED I

·bwkl.Y~UI Director of Graduate Studies

%,~.~~~~ Dean ot Appl1e Scienee and. jor Protanor _- 1----- --Bueineal Adm letration ~

June, 1953

Bl acksburg, Virginia

. . . .

LD s 6 :::: V~5 !:" I q,; 3 ~---I g(; 1> c.2

-

•

TABLE OF CONTENTS

I. I!lTR(.DUCTICN...

II. LITERAl'DRE ru;;VI1W

Monochloromethyl ~~her

Organometallic Compounds •

Organozinc COilipounOS

Grignard Reagents and Organocadmlum Compounds

III. EXPERIJ;r;NTAL . • • • .• •

Preparation of Reactants •

Konochloromethyl Ether

Grignard Heaeents ..

DialJcy1 Cadmium COIIIpound.

Dialk,yl Zinc Compounds .•

Reaction with ~fonochloraneth31 Lther

Grignard Reagents •. . .

Dlalky 1 Cadmium Compounds

DialJcy1 Zinc Cempound.

Results . • . . . . . . . .

•

•

•

•

•

•

P8ge

4

6

6

8

9

11

13

13

13

13

14

15

16

18

18

19

19

20

IV.

V.

VI.

VII.

VIII.

IX.

-3-

DISCUSSION • • · . . . . . . . . . • • •

Reactions of Honoehloromethyl t..'t.her with OrganOll\etall1c Canpounds

. . . . . . . . . . . . .

A Coroparison Different

of the Performance ot Zinc-Copper Couples . . . . . . .

Sill-W'.ARY • • •

ACKNO,iLI:IlGMENT • • • •

BIBLIOGRAPHY

VITA . . • . . • . • • • · • • · • •

APPENDIX . . . · . • · •

Page

24

26

29

30

3l

33

34

- 4-

1. INTRCDUCTIGN

The carbon-halogen bond in alpha-halogenated ethere 1s

easily ruptured. When this bond is subject to attaek by or­

ganometallic compounds, usually an ether with a longer caroon

chain 1. tormed. In the case of monoehloromethyl ether, the

following generalized equations may be used to describe the

reactions I

2CICH:2OCHJ + 2lIMX __ 2HCH2OCII:! + MC~ + ~

2CICH:2OCIIJ + Il2'I ~ 2HCH2OC~ + I!C~.

The purpose of this investigation 1s to study the relative

ease or reaction at monochloromethyl ether with dialkyl zinc and

cadmiuz compounds ae compared to that with the more familiar

Grignard reagent ••

The dialkyl cadmium canpounds are prepa.red through the

reaction of the corresponding Grlgnard reagent. with anhydrous

cadmium. chloride.

The d1a.llQrl zinc cOIIlpound.e are prepared trom the reaction

of the al.ql halide and & zinc-copper couple. As this method. ot

preparation otten results in low yield or failure due to the

ditticult7 in initiating the reaction between tho alk71 halide

-5-

and the zinc-copper couple generally used, an investigation 18

made tor a more satisfactory method ot preparing the couple.

The reaction!ot monochloramethyl ether with the organo­

metallic compoWlds are carried out under conditione a6 l5im.ilar

as po&dble 80 that the reactions can be favorably compared.

-6-

II. LITERATURE REVIt"'

The purpose ot this review is to present a necessary

historical background to the present investigation. It ia

not within the scope of the prese.nt inveetigation to review

in ita entirety the large field of organometallic compounds

and their m~ reactiona. Those publications which directly

relate to the reaction of monochloromethyl ether with organo-

aetalllc compounds are caretully reviewed. Special emphule

1s placed Uf~n t hoee investigations describing the preparation

ot dialJc;yl zinc compounds.

Monochlorometrrrl Ether

By virtue ot its very reactlye chlorine atOll, mono-

ehloromethyl ether hae been used to introduce the meth~­

methyl group in place ot the reactiYe hydrogen ataaa(21,30,31),

in certain organic COIII.pounds. Itf! reaction with organometallic

compounds turnishes a general method ot preparation ot various (20)

methyl 8there. As early as 1891, Henry had. reported the

reaction ot monoch1oromethy1 ether with dimethTl, alethy1, and

dl-n-propy1 zinc to tona reepectlTely methT1 eth;y1, methf1

n-propy1J and .et~1 n-butl1 ether. The tlret reaction waa

-7-

reported. to gl ... e almost 90 per cent yield. No ;Yields vere ro-

ported tor the latter two reactions.

Since 1936, reactions of monoehloromet~l ether with m&n7

ditterent Grignard reagents were reported. Those with n-butyl,

•• e-butyl, and tert-butyl magnesium bromides gave yields of

69, Jl, and 0 per cent, respectivelT(2).

No investigation of t he reactions of monoch1orom.t~l

ether with dlaJ..)q'l cadmium ccmpounde has been reported.

Monochloromethyl ether is gener~ prepared by the action (ll)

ot hydrogen chloride on a mixture of formaldehyde and methanol ,

-8-

Organometallic Compounds

Ever since the diecoverJ of dlethyl zinc by Frankland in

1849~ orsanometallic compounds haTe been increasingly valuable

1n the field of organic chemistry. They have been usea chiefly

in aTDtheais and in the study ot reaction mechan1ame(9).

The preparation of organometallic compounds can b. clasai-

tied. into three groups.

(A) The action of a metal or an alloy of the a.tal. on a

halide.

RI+M __ IIU

or 2RX + 21! ~ R:!K + IIX:z

(B) The action ot a metal on the organometallic compound

prepared trom another metal.

~II + II' ~ ~II' + II

M' is generally of a hlgh6r potential or above K in the electro-

aotive eeries.

(c) The action of a metal halide on the organomet&llic

compound prepared from another metal.

H' is generally below M in the electromotlTe seriee.

-9-

The orgilflomet C!.llic c (ltlpounds of the Group II elements,

It!pcc l al4' t hose ot magne5iwh (Gr i gns.rd r eagents), 7.1nc and

cadmium, are parti cularly 1nt~re5tins . Systematic studies of

the C Olll.pounc. ~ of t hese three li!e t ulr; by Grlgnard, Gi lman,

Blaise, :~oiler<27) J Cason( 4, 5,6) J and other workers, have

shown the cOQpounds t o be invaluable i n the synthesis of not

only other organic compounds but also the less known organo­

metallic compounds.

Ol'ganozinc Ganpounda

Frankland first reported the isolation or an organometallic

compound in 1849. Trying , to prepare the tree athyl radical, he

obtained diethyl zinc from the reaction of ethyl ioQide and

zinc(9). The orgallozinc compounds(8) were used 1n the prepara­

tion of m~ compounds until being l argely replaced by the

Grignard reagents. They are now used ohi.elly in the synthesis

ot ketones with acyl halides, or quaternar,y hydrocarbons with

tertlar;r halides and 1n the Relormateky reaction.

Dlalkyl i,1nc compouna s can be prepared by either or the

three JDethOG8 mentioned above, but are utiually prepared by the

action ot i,lnc metal on a halide(18) . A mixture ot one-halt

-10-

~l bromide and one-half alkyl iodide reacting with a

zinc-copper couple is found to be most aatistactor,y(27).

RX + Zn --+, RZoX

2RZnX heat, ~Zn + ZnXz

Different methods increaeine the reactivity of the zinc

were tried by sever,il investigators. Aside from one report of

(28) using a zinc-sodium alloy , most investigators favored the

use of a zinc-copper couple. The following methods for the

preparation of zinc-copper couplee have been reported.

(A) A solution of 1. 5 per cent cupric sulfate was poured

on zinc £0116(16). The foils were then washed with water/al-

cohol and ether, respectively, and dried in carbon dioxide.

(B) Copper was preCipitated from a cupric sulfate solu­

tion with zinc(29). The precipitate was mixed with zinc dust

and heated in an atmofphere of hydrogen .

(c) Cupric o~ide was mixed with zinc dust and heated

in an atmospher9 ot hydrogen until it wa~ completely reduced

to coppor(27).

(D) A zinc-copper elloy containing 5 to 10 per cent

copper was lathed into turnings and used(33).

-11-

Dlalkyl zinc compounds are generally prepared b7 using

ee) anc (D), above. However, the yield large~ depends upon

the teclmlque of the worker 1n the preparation of the zinc-

copper couple. The preparation ot a highly active zinc-copper

couple ha~ been described as "an art rather than a 6ciencen (29).

Experimenters with I1ttle experience will often find the yield

ot the dialkyl zinc ver,y poor or even void.

The spontaneously tlamnable nature ot dl&li<yl zinc COIl-

pounde in air is o'Yereome by the addition of an equal volume

of an inert eolvent(27). The compounds thus treated can be

handled 1n air with reasonable safety.

Grignard Reagents and vrganocadmium Compounds

Grignard reagents are the best known and moet studied of

all organometallic compounds. Their eaee of preparation and

reaction with many functional groups makes them ideal for the

IS)'nthesis ot other compounds.

Hallwachs and Shafaric first reported the isolation ot

what was thought to be diethyl magnesium(l3). Barbier and

Grignard in 1899 eflected the Grignard reaction by treating

methylheptenone with methyl iodide and Magnesium and got the

corresponding dlmethylheptenol. In 1900, Grignard divided

the synthe5is into two steps, the first of which bears hie

-12-

name, lind Made <! tls t.elfiettc study ot ite t'eact i one during the

f ollowing yeers.

Grlgll ... ro reagcr. t. t: ure utualJ.y prepared by t he reaction of

ax + Kg __ HMgX

2RIIgX

Before 1930 , little wa:s knOtro about organocad.'D1wn como.

pounaa other than t bei r prep~ration. In t hat year, Gilman and

Nelaon(14) recowmenoed t he re~ction between organocadmium cam-

pounds and acyl halides tor t he preparation 01 ketones. How-

eyer, the iaportance of thla methoe. war. not recognized until

seyeral years later. In 1941, ue Bennev1lle(l) reported the

preparation of ketones and keto-acids lI.'ith organoeadmlum com­

pound.. Subsequent investigations by Caaon(4,5,6) showed that.

this methoa compares favorably with other methode ot prepara-

ticn of ketonea. Since then, t he organocadmium compowlda hay.

aleo been usea for the preparation of a number ot other como.

. (22,)2) pounaa •

D1alkJ'1 eadlD.1um compoundl5 are usually prepared by the

reaction or anhydrous cadmium chloride with corre8ponding

Grignard roagent.(7).

2RIIgX + CdC12 ~ R2Cd + MgClz + Ksl2'

-13-

III. EXPERlMEIiTAL

Preliminary Remarke

In the preparation of monochloromethyl ether, the procedure

described by Marvel and Porter( ll) was used..

The preparations of the Grlgnard reagents and the dlalk,yl

cadmium canpounds followed that of Cason and Prout (7).

Dla~l zinc compounds were prepared by the modification ot

procedures described by Noller(27), and SOro08 and Morgana(33).

Under the headings, Pr epAration ot Reactante and Reaction

with Monochloromethyl Ether, only those procedure8 that were

found to give the beet results are described.

Preparation of Reactant.

Honochlorametbll Ether

Three hundred end fifty grame (10.9 molee) ot methanol and

900 g. of technical formalin containing 252 g. (8.4 mole.) or

formaldehyde were placed in a tva-liter round-bottom flask to

which were attached a reflux condenser and a glase tube reach-

ing to t he bottom ot the flask . Hydrogen chloride was run

into the solution, which was cooled with running water . Alter

a period of time, two distinct layers appeared. The tlow ot

-~-

hydrogen chloride was continued until it was no longer absorbed.

The two layers were separated. The water l~er was saturated

with calcium chloride. Again, two layers appeared. The mono-

chloromethyl ether layer was separated and added to the main

portion. It was then dried over calcium chloride and distilled.

The portion boiling from 57 to 60 °C. was considered sufficient-

17 pure and Va5 used in later reactions.

In generating hydrogen chloride, two methode were trled(12).

One was the addition ot concentrated hydrochloric acid to con-

centrated sulfuric acid; the other was the addition of concen-

trated sulfuric acid to a paste of sodium chloride and con-

centrated hydrochloric acid. Both were found to be un.atl.-

factor,y as the hydrogen chloride streams thus generated were

slow and uneven. Frequent refilling of the generator wae

necessary. The usual time needed to complete the reaction

was 40 hours. The yield was low (usually about 350 g. or 52%).

By using dry hydrogen chloride from a commercial cylinder, the

flow of ga5 could be controlled at much higher rates. The re­

action va8 completed in 10 hours. The yield was 596 g . or 88%.

qr1gnard Reagents

Twenty-four and three-tenths grams (1 g. atom) of magnesium

turnings were covered with 150 mi. of dry ether in a one-liter

-15-

three-necked flask to which were connected a mechanical stirrer,

4 reflux condenser, and a dropping funnel. One mole at alkyl

bromide in 250 ml. of dry ether was placed in the dropping

tunnel. A tew railliliters of the bronu de were ad.ded to the

flask which wae warmed with stirring. The reet of the bromide

was added dropwise during one and one-halt hours. Heating wu

discontinued as soon as the reaettan started. After the addi­

tion of the bromide, the mixture was refluxed for 15 more

-minutes to complete the reaction. Vurine the reaction, a pro­

tective nitrogen atmosphere was used.

Dlalkyl CadDllum Compound.

One mole of alkyl bromide in 250 ml. ot dry ether was

added to 24.3 r,. (1 g. atom) of magnesium turnings and the

Grignard reagent was prepared as above.

The rluK was ther, cooled in an ice bath and the dropping

runnel removed. Ninety-eight grams (0.54 mole ) of anhydrous

cadmium chloride were added ever a period or 10 minutes under

nitrogen atmosphere with ~tirrin~. The ice bath was removed

and the mixture was reflw:.ed for 45 minutes. The mixture va.

then tested tor the presence ot the Grlgnard reagent. It the

test was positive, the mixture vas retluxed until further

tests indicated the absence of the Grignard reagent.

-16-

The test for the presence of Grignard reagents (Gilman

test)(lO) was made as tollows: a small portion or the reaction

mixture was added to one-half milliliter of 1% Michlerls ketone

in dr,y benzene. After shaking the mixture, one mill iliter ot

water and a rew drops of 0 .2;( iodine in gl acial acetic aeld

were added. The development of a greenish-blue color shows the

presence of a Grlgnard reagent .

Dialky1 Zinc Compound.

1. Preparat ion of Zinc-copper Couples

Three different methode of making zinc-copper couples

had been used in the prepar ation of dlalkyl zinc. All of the

couples contained 5 to 8~ of copper by weight.

(a) One hundred and twenty grams of zinc duet and

10 g. of powdered cupric oxide in a 500 ml. three-neckecl flask

vere heated gent~ over a free flame in a current ot hydrogen

with shaking until the cupri c oxide was reduced. Care must be

taken not t o heat to t he point ot fusion. All moisture must

be driven off.

(b) Pure zmc and copper metals were m.1xed end melted

to ronn an alloy containing 5 to 8% of copper by weight. The

alloy wu then lathed to small turnings.

-17-

(c) A moditieci method of (a) wa~ employed as tollowl

one hundred and thirty grams of zinc dust were mixed thorough~

with 25 g. ot cupric citrate pow"der and placed in a 500 ml.

three-necked tluk. The contents of the tlll8k were heAted

with a tree flame under a protective current ot nitrogen until

all of the cupric citrate decomposed. This W8.a shown by the

tact that gases anu moisture ceaaed to be evolved. Furthermore,

the powder chnngod in appearance from a creen color to a rusty

color.

All couple. prepared, except in the case ot the alloy turn­

ings, must be used directly and. without exposure to the air.

2. Preparation ot Dialkyl Zinc

A 500 ml. three-necked flask containing approximatelY

130 g. of zinc-eopper couple was equipped with a reflux con­

den"er, a mercury .ealed stirrer, and a dropping funnel

(Figure 1). The apparatus was thorough~ flushod with nitro­

gen and. warmed over a water bath. Halt a mole each of the

alkyl broa1<ie and iodide were placed in the dropping tunnel.

A rew mill1liters ot the halide were added to the zinc-copper

couple and the rest added dropwiIBe over a I eclod ot 2 to 3

houre with stirring under an atmosphere ot nitrogen. The mix­

tu.re was allowed to etand overnight.

-18-

The flask waB then quickly connected with a vacuum dis­

tillation system (Figure 2) and distillation was attempted at

a pres5ure of 2 to 30 runt.

All of the cotulectiona that were exposed to the d.1a.lk.yl

zinc vapor had to be made with cork stoppers.

Whenever the reaction seemed t oo vigorous, heating was

discontinued and if needed, an ice bath was used t o cool the

flask until the reaction subsided.

Reaction with MonochloromethYl Ether

Grignard Reagent.

Thirty- six gr ams (0 .44 mol e) of monochloromethyl ether in

an equal volume of ether were added dropwiee to the previouely

prep&rea Grignard reagent. The reaction flask was cooled with

running water. Atter the addition the contents of the flask

were heated to reflux conditione for 30 minut ea. The flask waB

cooled t o room temperature and then placed 1n an ice bath. One

hundred milll1iter6 of water and subsequently 200 JIll. of ld

concentrated hydrochloriC acid were added slowly. The water

and. ether layers were separa.ted. The ether l a.yer was dried.

over Drierite and fractionally distilled.

-19-

DialJsy 1 Cadmium Canpound.

Solvent ether ..,&6 first partial~v removed by distillation

from the mixture contRinlng the d ialkyl cadmium. 'l'hirty-six

grams (0.44 mole) of monochloromethyl ether in an oqual volume

of ether vere added dropwise into the flask. The reaction flask

was cooled. with running water during the addition. The eontents

of the flask were then refluxed for 30 minutes, cooled to roam

temperaturs, and placed 1n an ice bath. One hundred milliliter.

ot water and then 200 ml. of 1:1 concentra.ted hydrochloric: acid

were added. slowly . The water and ether 1~er8 were separated..

The ether l~er wax dried over Drlerite and fractionally distilled.

DlaJ.k,yl Zinc Compound.

~ent¥-five grams (0.31 mole) of monochloromethyl ether in

an equal yolUUlo of n-butyl ether were added ciropwise to the

dlalkyl zinc all50 in an equal. volume of n-butyl ether. During

the addition, the flask was cooled in running water. The con­

tents vere then reluxed for 30 minutes, atter which time 100 ml.

ot water vere aQded to hydro~ze any unreacted dialkyl zinc.

The ~ther layer was separated frca the ~at.r l~cr, and dried

over Drierite. It was then fractionally distilled.

- 20-

Reeulte

Owing to the close boiling points and refractive indic ••

of methyl n-butyl ether and methyl iso-butyl ether with respee-

tively hexane and 2, 3-dimethyl butane, which may be produced

by the coupling reaction of the organometallic comPOund.(25)

(Table I), the fractions within 3· of the bolling points or the pure compounds were combined ana added slowly into a tlask

of concentrated GUlfurlc acid in an ice bath.

TABLE I

Compound B. P. H2O ~

Methyl n-butyl ether 71' 1.38

Hexane 68.7' 1.3749

Methyl iso-butyl ether 59'/741 mm --2,3-dimethyl butane 58' 1.3750

The insoluble product, if any, was separated and weighed.

The weight of the combined portions minus the weight of the

insoluble prodUCT wae considered to be the weight of the ether

obtained. The zul1'uric acid layer was then poured slowl,y into

a flask ot cracked ice. The ether recovered was dried over

Drierite and the boiling point and refractive index taken.

- 21-

From two t o fiv€ runs we r e made of each of t he magnesium,

cadmium, and zinc de r i vatives ot n- and l Bo- propyl halides . The

observat ions and the r esult s ar e summarized as follow! (Table I I) :

TABLE II

Number E&se ot Reaction with Y10~~ Compound of Runs Prepar a.t ion Ch1oromot hy1 RCH20C Ether

~-C3¥gBr 4 easi est vigoroue 68.C

lso-C3¥gBr 2 easiest vigorou. 42~

(n-C3&r)2CQ 5 easy l ess vigorous 32%

( i ao-CJ &r)2C<1 J oaay 18t16 vigorou. 16%

(n-CJ &r)2Zn 3 diffic ult Vi~OroU8 63%

( iso- CJIry)2Zn 2 difficult vieorous 32%

• Baaed on monoehl oromet hyl ether.

In the react ions of the magneBium and cadmium compowtde, no

attempt wa~ m~de t o i solat e the compounds befor e the addit ion of

monochloro~othyl e ther.

An investigation of the activit ies of the zinc- copper couple.

,-las made . lo!ethyl , ethyl , n-propyl and isopropyl balides were re-

acted with zinc-copper couples prepured by t hree differ ent meth-

ods. The results ar e shown in Tables III and IV .

-22-

T\BLE III

Couple Initiation Runs

Type Period Ccmpleted Lvicence ot ~Zn

lieact10n Dbt111ed

A 1. 5-2. 5 hr. II 6 2

B 2-10 min. 8 8 4

C 0 . 5- 5 min. 7 7 4

TABLE IV

Couple % Yield R2Zn

Type Methyl Ethyl n-Propyl bo-Propyl

A 80 10 0 -B - - 8 8

C - 38 13 12

A represents couples made from the reduction of cupric

oxide in the presenee of zinc.

B represents the zinc-copper alloy.

C represents couples made from the decomposition of

cupric citrate in the presence ot zinc.

-2)-

\oath the equlr;ment available, oxidation of the dl&lkyl can­

pounde could not be avoided during the vacuum distUlation. The

yields ot the dlal.lQrl zinc compound.. were much lower than the

percentage conversions ot the alkyl halides baaed on the amount.

of unreacted materials.

The reaction of the halides with the type C couples waa

much more vigorous in compari8on with the other couples.

The reaction of monochloromethyl ether with diethyl zinc

gave 82% )"ield. of methyl n .. propyl ether. The yield. ot methyl

ethyl ether from the reaction ot monoehloromethyl ether and

dimethyl zinc: wu greater than 70%. The low bolling point ot

metnyl ethyl ether (4·C.) made the recovery ot th16 material

extremelT difficult.

-24-

IV. DISCUSSION

Reactions of Honochloromethyl Ether with

Organometallic CompoWlde

The carbon to chlorine bond in some organic cOilpounds iB

easily ruptured. For instance, allyl chloride and benzyl

chloride possess carbon to chlorine bonds which are easily

cleaved. Such compounds have a tendency to diasoclate a

chloride ion when an outside force such as a nucleophilic

reagent 1& present. In the ca~e of allr1 chloride in a non-

ionizing solvent, the following reaction may occur when an

iodide 1s present I

Monochloromethyl ether under similar conditions, m., react

&8 tollows:

It haa been demonstrated that the carbon to chlorine bond

1n the ether ie more reactive 1n a nucleophilic displacement

reaction auch 8a the above than is the case for allyl chloride.

-25-

In recent yeare, some evidence haa been presented to shov

that alpha-halo ethyl alkyl ethers torm tautomerlc~ an

oxonium 10n and a halide ion.*

[CA2 • CH-9H J+ C1 R

Such eviQence offere additional proof of the reactivity of the

carbon to chlorine bond in alpha-huo ethere.

The over-all reaction of monochloromethyl ether with organo-

metallic compounds can be expressed in the following two equa-

tiona:

The yields ot methyl n-butyl ether 1n the n-prop,yl seriel

appeared in the order RMgBY> RzZn> RzCd. Tho yield. of methyl

ieo-butyl ether in the iso-propyl Beries also tollow the same

order. The prim8.l'1 organometallic compounds gave higher yleld..

of ethere than the corresponding secondary compounds. The

* Original article by Shostakovekl1 and BogdanoT& ab.tracted in C. A., ~, 45l9d (1948) v80 not seen.

-26-

lower yields of ethers from secondary organometallic compounds

may be due to competing side reactions.

Better yl~lds were obtained when the monochloromethyl

ether was slowly aQaed to t he organometallic compounds.

From the results of this investigation, Grignard reagents

and dlalkyl zinc compounds are recommended over the dialkyl

cadmium compounds in the preparation ot ethers from monochloro­

methyl ether. The best yield8 were obtained by using the or­

ganometallic compounds prepared from primar,y halides,

A Comparison of the Performance of

Different Zinc-Copper Couplea

In the following discussion, the zinc-copper couples pre­

pared from zinc dust and cupric oxide will be referred to ae

type A couples, the alloy a8 type B, and that prepared trom zinc

duat and cupric citrate aa type C.

The investigation ot the pertormance or the couples wae made

because ot the unsatistactory results otten obtainea with the

cammon~ used type A couples by the inve6tigator and by others.

The three types or couples were comparea on the basie ot

(1) ease of preparation or the active couple, (2) length or

the initiation period for the [ormation ot the alkyl zinc halla.,

and (3) the yield of the dialkyl zinc compound.

-27-

Type A 1e hazardous to prepare as i t involves heating the

mixture of zinc duet and cupric oxi de wit h a fre e flame under

an atmosphere of hydrogen near t he point of fusion of the metals .

Furthermore, the cupric oxide, when well mixed with zinc dust, is

hard to reduce, probably because of its snail chance of contact

with hydrogen. lbere is no w~ of knowing when the couple is

r eady for use . Heating for as long a period as one and one-halt

hours with constant shaking somet imes failed t o give reactive

eoupl es. Type B involves tedious wor k i n melting, caeting, and

lath.lug the alloy and i s, therefore , impractical fo r use i n just

a rew mall Beale experiments. Type C 16 made s imply by toating

with. free flame a mixture of cupric citrate ane,;, zinc dust be­

low the point of fus i on ldth occadonal shaking . It 1s ready

tor use in 10 to 30 minutes, depenaing upon t he amount of heat

applied .

The initiation period of the type A couple is usually

longer than one and one-halt hour_. Type B usually react.

within 10 minutes. The r eaction of t ype C couple in sa.e case.

has been almost instantaneous . The init iation period 18 u8ual..lJ'

lee. than five minute • •

The ,-iol d. of the dialJc;yl zinc compounds obtained by the

use of type A couple. are usually good, while thoBe frcm. type B

are coneiaerably lower in lIlost instances . Examination of the

-28-

residue trom t he type B couple often ~bOW8 unreacted zinc in

the center of the turnings, especial~ when the larger turnings

are used. The yields of the dia1.kyl zinc ecmpound~ from the

type C couples are superior to those obtained from the other

types of couples.

In view of the above facts, type C le considered to be

superior and is strongly reco~ended for l aboratory use 1n

place of the other two types.

The preparation ot a zinc-c opper couple by the decomposi­

tion of cupric tartrate or other cupric salts of organic acids

has not been investigated. The therm&l decomposition ot such

salts has been used to prepare copper catalysts. Theee salta

will probably give the S8me results as obtained with cupric

citrate.

The high reactivity of the couple was believed to be due

to the formation of extremely reactive minute particles of

copper, that 1s, particles ot high specific surface.

-29-

V. Sl~'-~AR!

1. The reactions of aonochloromethyl ether wi th the mag­

nesium cadmium and zinc derivatives of normal propyl and iso­

propyl halides were investigated.

2. In both the normal propyl and iso-propyl aeries, the

yields of the expected condensation products RCH2OCH) appeared

in tho ordor RMgX>RzZn>~Cd.

3. The pr1.mary organometallic compounds reacted. with

monochloromethyl ether in higher yields than did the corres­

ponding !Secondary eanpounds.

4. A neN" method was deyeloped for preparing a zinc-copper

couple by heating a mixture ot zinc duet and powdered cupric

citrate under an atmosphere of nitrogen.

5. This method gave a couple superior in laborator,y use

to the zinc-copper couples how generally employed.

-30-

VI. ACKNOWU;LGbENT

TIle author offer6 hlE sinc ere thanks tc Dr. Robert C. Krug,

whose A8sistance h&6 m&de this inv6stigation possible.

-31-

VII. BIBLIOGRAPHY

1. OeBonneville , J. Org. Cham., 6, 462-6 (1941) .

2. Brugh, ~Act1on of Mono-chloromethyl ~~her on Aliphatic

Grlgnard Reagents," B. S. Thesis, V. r . 1., 1950.

3. Buck, et 01., J. !nst. Petroleum, ~, 339-68 (1948).

C. A., 42, 70llh (1948).

4. Cason, Chemical Reviews, ~, 15-30 (1947).

5. Cason, J. Am. Chem. Soc., 66,1764 (1944).

6. Ca.on, J. Am. Cham. Soc., 68, 2078 (1946) .

7. Cason and Prout, J . Am. Chem. Soc., 66, 46 (1944).

8. Denni. and Hance, J. Am. Chem. Soc., !t1, 370 (1925).

9. Gilman, "Organic Chemistry," Vol. I, John Wiley an1 Sons,

New York, 1947, p. 490-524, 547-552.

10. Gilman, J. Am. Chem. Soc., !t1, 2002 (1925).

11. G1lm.an and Blatt, "Organic Synthesis," Collective Vol. I,

John Wiley and Sons, New York, 1948, p. 369.

12. Gilman and Blatt, "Organic Synthesh," Collective Vol. I,

John Wiley and Sons, New York, 1948, p. 287 .

13. Gilman and Brown, J. Am. Cham. Soc., zg, 5045 (1930).

14. Gilman and Nelson, Rec. traY. chim., 22, 518 (1936). 4 C. A., ~,5951 (1936) .

15. Gillun am I/oods, J. Am. Cham. Soc., m., 520 (1945).

-32-

16. Gladstone and Tribe, J. Ch .... Soc., 26. 445 (1873).

17. Gladstone and Tribe, J. Chem. Soc., 26. 678 (1873).

18. Gladstone and Tribe, J. Ch .... Soc., 26, 961 (1873).

19. Hatch, Gordon, Sutherland, and Ross, J. Org. Cham. J 1Jt,

1130 (1949).

20. H.nry. Compt. r.nd •• 113. 368 (1891).

21. Huntress, "Organic Chlorine Compounds," John Wiley and Sons,

lIe" York. 1948. p. 976-80.

22. Hurd end Iloly ... J. Am. Chem. Soc •• E. 2005 (1950).

23. Krauae. Ber •• 22> 1813 (1917).

24. Long ~~d Norrlsh, Trans. Roy. Soc. London, ~, 587 (1949),

c. A •• ~. 5242e (1949).

25. McCleary and Degaring, Froe. Indiana Acad. Sci., ~, 127

(1934). C. A •• ~. 72453 (1938).

26. Meyer. Chem. N ..... 131. 1 (1925).

27. Noller. J. Am. Chem. Soc •• 2!. 594 (1929).

28. Ragooin, Ber., 26 (Rer.), 380 (1893).

29. Renshaw and Greenlaw, J. Am. Chem. Soc., !i3., 1472 (1920).

JC. SiRon.en, J. Chem. Soc., 22. 1777 (1908).

31. SiRon. en, J. Chem. Soc •• 107. 783 (1915).

32. Sondheimer and Weedon, Nature, 165, 483-4 (1950).

33. Sor008 and Morgana, J. Am. Chem. Soc., 22, 893 (1944).

-33-

VIII. VITA

The author was born near Shanghai, China, in Mareh 1929.

He attended ~ t . John l ~ Univer sity in Sltanghai, China, in

September, 1946. He 1'1&S tran sferred to Indiana University in

September, 1947 , and was gr aduated in JWle , 1950, with a major

in chsmietry. In September, 1950, ho entered Virginia Fo~eeh­

nle Institute for gr aduate study in ehemistr,y.

-34-

IX. APPUlDIX

Figure 1. Frepar aticn or Clalkyl Zinc Compound.

Figure 2 . Vacuum Distillation anc Reaction Sys t.em

--------

I ! ! i I , , i i ,

I

!

I ,

I l I I

L . __

-_ ...... -. -_._---------- ---.---.--~!

L __ N2 FIG_l '

._-_.-

, , I , I

: i , I . i

I

! I I . I , , ; I

: j ; , i , I I 11 ' 'I II '! ! I

; 1 • !

t i

.--j ,

j

, " \ , ,

: ! I i

-, - ....

)~ /

-

HH ~

) i , , , I I ,I' ,

J -'

, ... " "..-J -------=== ---- -----------

-_._--------

1 ______ . ----- ._--

I ~