-v

.‘rn—vO—v- "

F}; H Colt—‘0 n ,
I.

Meats

’...
. r1 =0

(' ; .: L §

A, . ' I . ‘. .
3‘ '.h ‘J 11:. \l ‘- .

Y,

fin.- --.§."I r.“ r. ~~ ,.

 

A Study of
P-and O-Benzyl Phenol
and

Some of Their Derivatives

Thesis

Submitted to the Faculty of
Michigan State College as partial fulfill~
ment or the requirements or the Degree at

Master of‘Science.‘

By
Frank Hiram.Maxfield
June 1929

Appreciation

, The writer wishes to eXpreea his
appreciation to Dr. R. C. Huston for the
assistance and timely advice gnren throughout
this work, without which the work would not

have been possible.

331.651.-

In 1872, a. Paterno (Gaze. chim. ital..§, 1-6)
heated gently a mixture of bennyl chloride and phenol
in the gresenee of an. H01 was given of , and the li uid
entered into ebullition. Later, a brown liquid was sep-
areted frcn the 2111, ed distilled. the em: mbined benzyl
chloride and phenol were distilled off below 2600. ihe
remaining mass Wss distilled at 6 mm.. and the main fract-
ion thus distilling at 180~1900. :his solidified to a
mass of small needles. contaminated with sn.oil, which was
pressed out and discarded, while the needles were recry-
stallized trom.alcohol. lhis bensylated phenol crystallized
in white silky needles (Id. 84) which were soluble in el-
cohol, other, benzene and chlorofcu‘m. Ihe pure phaiol
boiled at 175-180 at 4-5 mm. It was soluble in alkaline
solutions, but was precipitated b; acids. It was insoluble
in ma. 23:03 formed substitutioz'i 1r oz‘ucts. {Egg-.30.. {five
a sulfonic acid with the phenol, the barium salt of which
was soluble in are.

Benzyleted anisole treated/Yijlfihzmd boiled fa'
eight hours at 170° gave CBsI,and the same benzyl phenol
as Just described.

In 1874, E.-Feterno and K. Eileti (Gaza. chin.
ital.Ԥ,121-129, 251-254) published an article in which
they describe more derivatives of benzyl phenol. Lcnzyl phenol
with acetyl chloride cave benzyl pheml acetate in the cold .
This compound could not be crystallized but remained a pale

yellow liquid, s.r. 317°, density 1.10é3 at 160. it was very

easily hydrolized, absorbing sufficient moisture from
the air to decompose the ester, benzyl phenol crystallizing
out.

Benzyl phenol and benzoyl chloride gave benzyl
phenol benzoate. This was very soluble in c.H., and cry-
stallized from.cefia in.friable crystals of the triclinio
systsn. It crystallized from alcohol in glistening needles,
M.P. 86°. This compound was not decomposed by alcohol, H30
or aqueous potassium.hydroxide.

.The same authors describe a di~brom derivative
prepared by adding excess Bra to a solution of benzyl phenol
in 03.. This compound melted at 175° and remained an amorphous
substance, soluble in CHCla and 052, but insoluble in el-
cohol and other. Later workers have questioned this compound.
(Zincks and Walter. Ann.'§§g, 367-385, (1904) ).

In 1875, Paterno and Fileti (Gazz. chim. ita1.'§,
381) again published an article in thich they describe the pre-
paration of’benzyl phenol by a different method. Benzyl.al-
cohol and phenol were condensed, using a mixture of H380. and
acetic acids to eliminate a molecule of FOB. They obtained a
mass of crystals, nhich on purification proved to be benzylated
phenol. A fluorescent oil was also obtained in oonsiderabla
quantity which did not crystallize, but appeared to be an
isomeric benzyl phenol.

In 1880, Perkin and Lodgkinson (Journ. Chem. Soc.
1880, P 724) boiled phenyl acetate with benzyl chloriie. A
beautiful fluorescent liquid was obtained, B. P. 810—3200.
Analysis shm ed a composition of 02511240.. This was readily

"

"V I.“ ‘

saponified with alcoholic KOH. An oil rose to the surface.
This oil on purification, yielded crystals, M.P. 390.
Analysis indicated formula of 0108100. This may be cinnamyl
methyl ketone, CoEs-GH - CH - CO - CHs. The alkaline solution
from the above saponification was acidified, and an oil sep-
arated out. On purification it melted at 80 - 810 and boiled
at 320-3220. analysis indicated formula of clsfllzo‘ They
concluded that this compound was identical to the benzyl
phenol prepared by Paterno by the action of benzyl chloride
on phenol in the presence of Zn dust. Tie dii‘fereme in
melting point was thought due to impurities in.their CORP
pound,

’ Liebmann (Ber. 1§, P. 152, 1882) (her..lg, 1842,
1881) prepared benzyl phenol from benzyl alcohol and phenol
in the presence of molten ZnClg. Prepyl and amyl phenols were
prepared by the same method. The benzyl phenol thus preps red
had e um. 84° and boiled at 325-3500. '

In 1882,3ennie made a rather extensive study of

benzyl phenol, and its derivatives. (3.0 hem. Soc. 1882, P 33).
is prepared his benzyl phenol in the same manner as Paterno, i.e.
benzyl chloride and phenol in the presence of Zn dust.
He prepared and described the mono sulfonic acid ani its salts L
with barium.and potassium, potassium.bromo-benzyl sulfonate, E
tri nitro benzyl phenol and the methyl ether of benzyl phaiol.
After a study of these derivatives,he states that benzyl
phenol resembles p-cresol in its action with HBSO‘.-HNO. or

Brfl s

Later in the same year Rennie (J. Chem. Soc.
1882 P 220) reports a further study of benzyl phenol der-
ivatives. This time he includes benzyl phenol-sulfonic acid,
mono nitro benzyl phenol, amide benzyl phenol, d1 nitro
benzyl phenol, tri nitro benzyl phenol oxidation products,
nitro bran benzyl phenol and its reaction with 11110.. Di
nitro benzyl phenol oxidized gave benzoic acid, indicating
that those two nitro groups were in the original phenol
ring. Tri nitro benzyl phenol, upon oxidation, gave p-
nitro benzoic acid, hence indicating that the third nitro
group went into benzyl group, rather than in the original
phenol ring.

After studying these compounds, all derivatives
of benzyl phenol, flannie concludes thatthe benzyl gnaup
must be in the para position in the phenol ring. This
view is substantiated by the fact that the same nitro-
brem.derivativo is obtained, whether HBO; acts on the mono
bran derivative cr'Brg acts on the mono nitro derivative..
Hence, the nitro brom.derivative, and consequently the nitro
sulphonate, and the brom.sulfonate must be symmetrical.
The only possible way to explain this fact, is to seem >
that the benzyl group is in the para position of'the ph6101
ring. Ala: nitro brom.benzyl phenol, with.HNO., gave orth>f
bromralpha-dinitrcphenol, the second no; group taking the p-
position left by the benzyl group.

Although the above reasoning seemingly makes it

certain, that the benzyl group is in the para position, other

\1

facts served to add an element of uncertainty to Ronnie's
theory. First, the melting points of the mono and di-
nitro derivatives differed by only 150, while the corres-
pending derivatives of para cresol differed by 40°.
Secondly, the yield of di nitro benzyl phenol was always
small, regardless of the method of preparation. Still
more important, however, was the fact that the potassium
salt was of an abnormal color for such.di-nitro compounds.
This salt gives orange needles in contrast to the depth
and beauty of the crystals of di-ortho-nitro derivatives of
other phenols. The fact, that addition of nitric acid.to
the di-nitrc-benzyl phenol gave tri-nitro phenol instead
of tri—nitro benzyl phenol, was also puzzling. Tri nitro
benzyl phenol is formed by direct nitration of di nitro
benzyl phenol. Its constitution.nill depend on that of tha
di nitro derivative. One N03 group must be in the benzyl
group; p-to the CH3, however, since it yields p~nitro benzoic
acid on oxidation, and a second one must be in.the position
ortho to the on group. The position of the third must b.
regarded as somewhat doubtful.

Rennie published again in 1886 (J. Chem. Soc. $3,
406) reporting some further examination of di nitro pire
benzyl phenol, and also a study of the oil which always seems
with the crystals of p-benzyl phenol. This oil was pressed
out, extracted with alcohol and studied. It was first all-
fonated and the barium.salt formed. The Ba salt of the p-
bsnzyl phenol crystallized out, but the remainder did not

promise to crystallize well, so the barium.was replaced with

LA

it

he:
be:
"it

ben

PU'D.
Pee:
hen:

Zinc

potassium, thereupon {good crystals were obtained with

2 1/2 molecules of H30 of crystallization. The nitro sul-
fonate formed pale yellow rosettes, M3 81-82. The sule
fonate treated with bromine, gave some brom sulfonate but
also some white flakes, which on recrystallizing from
alcohol and water, formed minute white crystals melting

at 919. This compound was soluble in alcohol, ether,
acetic acid or alkali. He remarks that this is probably
a di brom.derivative, but he makes no further study of it.
( This will be shown later in this paper, to be 2, 4, di brom
6 benzyl phenol, as Rennie suspected but had no definite
proof).

M. Backunin studied the condensation of benzyl
chloride and phenol in the presence of zinc (Gazzetta.§§_i,
495-496 (1803), ibid, 55 ii, 454-460 (1903) ) with tin idea
of determining the best conditions of Condensation. He
concludes that a neutral solvent is the best to use, since
it gives a finer division.of the substances reacting, and
hence more chance for contact and reaction. Be prepared
benzyl phenol benzoate (M.P. 86-87) by heating benzyl phenol
with benzoic acid, in the presence of phosphoric oxide in
benzene solution.

L. Zincke and W. Walter (Ann. ggg, 567-585,(1909) )
published some work:on bromine substitution.in phenols, es-
pecially in the case of p-benzyl phenol. They prepared the
benzyl phenol by condensing benzyl chloride and phenol with

zinc, as Paterno had done before. (10c. cit). He describes

the benzoyl derivative, as needles melting at 87°. P-benzyl
phenol treated with 28m. in a solution of chloroform, gave a
product having two crystalline forms, one being colorless
needles melting at 44°, but which was unstable and readily
passed to the more stable form on standing, the stable form
being rhombic crystals M. 57°. The acetyl derivative of this
di bran benzyl phenol formed mono-clinic prisms, M.P. 55°.
{6 concludes that the formula or this canpound is
<:>—¢<:2«~
because when it is treated with NaNOg,

a bromine atom is readily replaced with a 1.103 as is the case
when ordinary brom phenols are treated with IIaNOg, under the
same conditions. Zincke and halter also describe tri-brom
and ponta brom-derivat ives of p-bf-nzyl phenol. It is to be
noted, that the di brom derivative of o-benzyl phenol, announced
by Paterno and Fileti (Gazz. chin. ital. a, 121-129, 251-254)
M.P. 175°, does not agree with any compound prepared by Zincke
and Walter. ‘

The action of A1013. as a catalyst, in condensixg
various substances has been studied extensively since 1876,
when Friedel and Crafts worked with and alcohol and hydro-
carbons. (“ompt. rend. 84;, 1592) At this time they worked
only with aliphatic coupounds, and even state that AlCl. will
not condense aromatic compounds. Custavaan (Ber. 15, 157,1880)
(Bull. Soc. Chem. 33, 325, 1884) believed that an unstable
organic aluminum compound was formed, which broke down giving

11101.. In 1881, Lierz and heith, (Ber. 3.1, 187), report having

prepared di phenyl ether from the action of A101. on phenol.
1his would point to a dehydrating action of the A1013.
Friedel and Urafts (Ann. chim. rhys.,;, 489, 1884), advanced
the following theory of the action of an aromatic campound on
an alkyl halide in the presence of A1013, giving an alkylated
benzene ring.

Cl
0.1255 + A1013 ——-> I101 + Alibi
£0.35

01 '
AjCc1 + R'Cl -—e A101. + 8' CeRs
\CeRs regenerated

In 1901, Jaubert, (Comp. Rand. 152, 841) prepared
aniline and p-toluidine from.the action or HgNOH.H01 with
the hydrocarbon in the presence of A101., or Zn013. This
action.must be strictly a dehydration reaction.

Buttgenbach (J. Prakt. chem..;g§, 355 (1923) ),
believes that an unstable Al-organiccompound is first pro-
duced, which breaks up, giving A101; and the final condeised
product. I

In 1916,(Huston and Friedeman, J. A. 0.5. §§, 2527)
published the results of some work with A101. as a catalyst“
in the condensation of benzene and benzyl alcohol. They ob-
tained di phenyl methane, some p-dibenzyl benzene, a trace
of the corresponding ortho isomer, and anthracenc as a by-
product. There appeared to be a dehydrating action of the
A101. in this reaction. The proportion.of reagents used,
and the temperature, affected the yields of different products.

Huston published a brief report, (Sci. _5_§,
206-1920), of the condensation of benzyl alcohol and phm 01
in the presence of A101” obtaining p-benzyl phenol as
a product.

In 1918, Huston and Friedeman, (3'. 11.0.8. 32,
785-93) published work on the condensati on of benzene
with secondary alcohols. They used phenyl methyl carbinol,
phenyl ethyl carbinol and di phenyl carbinol in this
study. They found that the methyl and ethyl group seemed
to hinder the condensation, the ethyl having the greatest
retarding effect, while the presence cf the phenyl group
did not hinder the condensation.

Huston, (J. A. 0.8. 39., 2775-9, (1924) studied
the action of AlCls in the condensing of phenol and benzyl
alcohol, to give p-benzyl phenol. Phenol + A101. in petroleum
ether gave a viscous mass, probably Al phenolate, from which
the phenol was quantitatively recovered on addition of 801.
With phenol and benzyl alcohol in petroleum ether, with 0.5
mol. of A101. at 20-300, he obtained 43-59% p-benzyl phenol.
The yield was not increased by the addition of larger amounts
of A1013. Similarly, phemrl methyl ether gave 46% p-benzyl
phenyl methyl ether, and phmyl ethyl ether gave 57% of
p-benzyl phenyl ethyl ether. Benzyl chloride and phenol in
the presence of 19101;, gave 56% of p-benzyl phenol. Thi I
would bar the possibility that the action of benzyl alc olnl
and phenol with A101. is a reaction in which the benzyl aloof-
hol is first converted to benzyl chloride, and then the react-

ion being nerely an example of Friedel and Crafts reaction.

10

This is further barred by the fact, that when benzyl alco-
hol and A101. react below 55°, 96% of the benzyl alcohol
is recovered, while if the reaction occurs above 40°,
25-30% of benzyl chloride is recovered after a vigorous
reaction. It will be noted that the above condensation
was carried on between 20-500, which eliminates the poss-
ibility of the format ion of benzyl chloride in appreciable
amounts.

In 1923, L. Claisen (Z. Angew. Chem..§§, 478—9)
published some work on the carbon-alkylation of phenols in
which he gives a method of preparing the ortho alleated
phenol, to the practical exclusion of the para isomer. This
is done by treating the Na phenolate with the alkyl halide,
in the presence of a non-dissociatbng :medium, such.as
CsH.,or toluene, which gives almost entirely the ortho
alkylated phenol, instead of’the ether, as obtained with
the use of alcohol or other dissociating medium, as a sol-
vent. He characterizes several ortho alkyl phenols, among
them being ortho benzyl phenol, 1. 21°, B 512°, which he
reports is accompanied by p-benzyl phenol, M. 84, B 52) -20.
In an article published by Claisea, Kremers, Roth and -
Lietze, (Ann. 442, 210-45, 1925),ortho benzyl phenol is re-
ported to exist in two Terms, — the labile form.m.zl and
the stable fomn, M 52°. Other ortho alkylated phenols are
also described. Claisen's work is also reported by Claisen
& Tietzke (Ber. 588, 275 (1925) ). in this article, they
show that when the reaction between the sodium.phenokate

and benzyl chloride takes place in a non-dissociating medium,

such as benzene or toluene, that ortho alkylation takes

place, to the almost entire exclusion of the ether.
Claisen explains this reaction, on the basis

of Michael's theory (J. pr. 52, 486; 216,189) of tie

reaction between silver cyanide and methyl iodide. Accord-

ing to this theory, Claisen's reaction takes place as

f0 110178: 0

a "U. ’m
- . 'fi
$21”?ch O": I’M—7 0") 6'

K. von Auwers, Wagoner and Bohr (Chem. Zentr.
1926, I, 2547) carried on experiments to eXplain the re-
action of salts of keto-enols and alkyl halide. ihey sum
up three possibilities:

l. ”The initial formation of addition products
with subsequent splitting" (Michael).

2. "The initial formation of normal O-derivahves
with-rearrangement into the C-derivative".

5. "Separation.of metal as metallic halide, form-

ation of free hate and cool radicals,

 

/
-0+0/>-0-C
0 ‘\ n l\
o o

\\
and, with the slight reactivity of the alkyl group, partial
or complete rearrangement.of‘the enol to keto radical, and
finally union of the radicals”. (Wislicenus).
The course of the alkylation depends on the nature

of the enol-heto, and also the alhylization agent, as well

12

as on the type of medium used. saturated alkyl halides
promote the formation of 0-derivatives, while unsaturated
alkyl halides such as allyl and benzyl halides promote the
formation of C-derivatives. These facts are best explained
by the first hypothesh, which it will be noted agrees with
Claisen's idea. It is hard to see why an ether should be
rearranged to a C-derivative in benzene,more readily than
in alcohol, as the second hypothesis would indicate. Lenard-
ing to the hypothesis, it would seem that allyl and benzyl
halnies should form.et1ers because of their greater react-
ivity, but this is not the case. Allyl and benzyl radiaals
hold oxygen only loosely, whereas they form.stable compounds
with carbon because of their slight valence requirements.

some work has been carried.out on the rearrange-
ment of phenyl benzyl ether to give benzyhated phenols. J.
Von.A1phen (Rec. trav. chim.,g§, 799-812,1927) heated benzyl
phenyl other with ZHClg to 160° for'one hour, and found that
it was converted to 4-hydroxydiphenylmethane, M. 84° and
also a dyestuff. The above named compound was also obtained
when 10 g. phenol, 14 g. of benzyl chloride and a small piece
of ZHCIg, were heated together at 100°.

t. F. bhort, (J. Chem. Soc. 1928, 528) ale) reports
a small note on the rearrangement of benzyl phenyl ether.
The ether was heated with ZnCla.to 2250 or to 1800 in a
stream of H01 and he obtained a mixture of‘phenol, 0-benzyl
phenol, p-benzyl phenol and high boiling produ:ts. He re-
marks that the rearrangement follows a course similar to tie

Hofmann rearrangement of alkyl anilines. The p-OH dipheiyl

13

methane so obtained melted at 840 - 84.50, and was later
oxidized to p-methoxy benzo phenone M. 61-620. The 0-
hydroxy diphenyl methane reported melted at 540 and no
proof of structure is given. No further details are given
in this note.

There is also in existence a patent (U. S. 1,
580, 055), issued to W. KrOpp, W. Schremz and W. Schule-
man in which they prepare an 0-benzy1 phenol, M. 52°, by
treating the phenol with benzyl chloride and hater treat-
ing the resulting mixture with b‘a(0H).-, and heating. 'Jhe
cooled liquid is filtered, and the 0-benzyl phenol liberated
from.the filtrate by acidulation. Inquiry has failed to
produce any of this product to be checked with the 0-benzyl
phenol, later reported in this paper.

l4

EXperimental

This work was undertaken with two objects in
mind; -

1. To study the condensation of phenol and benzyl
alcohol in the presence of’anhydrous AlCla, wdth.emphasie on
the effect of varying preportions of phenol and AlCla, on
the yield of the various products produced.

2. To determine definitely, whether substitution in
benzylated phenols took place in the phenol ring, or in the
ring of the benzyl radical.

Part I
Duplicate determinations were made with the follow-
ing proportions of phenol, benzyl alcohol and AlCl.:
1. l moi. phenol; 1 mol. benzyl alcohol; 1/2 mol AlCl.

2.2 n n i n ,. " 1/2 a n
3.3 .. n 1 .. .. n 1/2 .. ..
4. 1 n n 1 .. .. .. 1 .. ..
5.2 n w l n " " l n .
5.3 w i .. " " i " ..

The phenol, 52 3., and benzyl alcohol, 55 g., were
suspended in 100 cc. of petroleum.ether in a bottle which had
three Openings in the top. In one Opening was fitted a reflux
condenser to return all evaporating petroleum.ether during the
course of the reaction, In the second and center Opening was
fitted a mechanical stirrer so arranged with a mercury seal
that no gasses could escape around the stirring rod during the

process. This stirring rod was hooked up with an.slectric

MAIL-a

15

motor in such a way that a steady flow of power was assured.
The third hole of the bottle was kept corked, except when
A101. was added through it from time to time. 33 g. AlCls,
finely pulverized, were added in small portions over the course
of about an hour, until the entire amount had been added.
When the AlCls is added, some heat is evolved, and if added
too fast, the temperature rises too high. The temperature-
should be maintained below 300 - 55°, as Huston has atom
that at this temperature very little benzyl alcohol is con-
verted to benzyl chloride by AlCla. At higher temperatures,-
above 40°, however, 25-30% of the alcohol is converted to the
chloride, and the reaction then becomes a Friedel and Crafts
reaction rather than one of dehydration. (J. Am. Chem. Soc.
‘29, 2775-9, 1924). A thermometer may be inserted in the cork
if desired, but it is difficult to read, since the mass soon
becomes amber colored, and smears the thermometer. After one
or two runs, the Operator can tell quite definitely, shouhi
the temperature rise above 30°. A cooling bath is unnecessary_
if the A1013 is added slowly, because the evaporation of the
petroleum.ether will carry off all excess heat generated.

After all the A101. has been added, the mass is
stirred for another hour, at which time it is quite viscous
and cloudy amber in color. It is allowed to stand for at
least twenty-four hours, and then the Jelly-like mass is de-
composed with ice, acidified with HCl, and extracted with
ether. -

The other is distilled from.the ether extract, and _
the residue heated at atmospheric pressure to about 220°, to

remove all phenol and benzyl alcohol remaining in the mixture.

16

It is then distilled in vacuo at 5-4 mm. and the fraction
collected from.l$0° to 175°. Also a fraction is collected
from.l75° to 220° at 5-4 mm. This is a viscous oil, pro-
bably a di benzyl phenol which seldom crystallizes, although
one sample has crystallized some on long standing. Above
2200 at 3-4 mm. no distillate is obtained, the residue being
a thick tarry mass, which will coke on continued heating.
This residue has not been investigated. All condensations as
above outlined were carried out in the same apparatus as des-
cribed above, and in essentially the same manner, save for
the varying amounts of‘ingredients.

The weights of the fractions from 150~1750 at 3-4

mm. follow. 1/2 mol. wts. were used in the ratios givel

below:
1/2 mol.AlCls l mol.AlCla
1 mol. Benzyl alcohol; 1 mol. Phenol 27 g. 23 g.
l ” " " 2 ” " 45 " 45 "
l " " " 3 " " 49 " 55 "

t will be noticed at once that the increased.amount
of A1013 causes little or no increase in the fraction from
130-1750, but that the increased preportion of phenol doas
cause a decided increase, especially the increase from 1 to
2 mole of phenol per mol of benzyl alcohol.

ihese samples crystallized partially, and developed
a reddish quinone color on standing, as any phenol does. Closer
fractionation indicated that there were two compounds present,
whose boiling points were rather close together. These two com-

pounds are ortho benzyl phenol and para benzyl phenol. 0-

17

benzyl phenol did not crystallize, whereas the p—benzyl
phenol did largely crystallize out. These fractions were

then filtered by auction, and a heavy 011 thus obtained in the
filtrate was fractionated a number of times, and the major
part of it proved to be ortho benzyl phenol, b.p. 158-900,

3 mm. The crystals remaining were redistilled and then.re-
crystallized and proved to be p-benzyl phenol, b.p. l53-4°

at 3 mm., m.p. 84°. The purified p-benzyl phenol does not
color on standing, but remains as a white crystalline compound.
The ortho benzyl phenol does develOp the characteristic red
color on standing.

It is impossible, b; any known.meth3d, to quant-
itatively separate a mixture of ortho and para benzyl phenol.
Accordingly the exact amounts of each isomer obtained in each
case are not known, but approximate amounts are of interest.
These figures are as accurate as can be obtained by filtration,
and subsequent close fractionation of the isomers, but at best
are only approximate. The approximate figures follow:

‘ 1/2 mol.AlCla l mol.nlCla
at. 330- 731»- wt. 25 0- ,3?-
l mol.henzyl alcohol;l mol phenol 27 g. 30% 70% 25 g. 40% 60%
1 " " " 2 " " 45 " 40% 60% 45 " 405 60p
1 " fl " 3 " " 49 " 55% 45% 53 " 50; 50%

These figures show merely that the amount of the ortho
isomer produced is greatly increased, as the amount of phenol.
is increased . Tith 3 mole of phenol to l of benzyl alcohol,

the amount of ortho isomer Iroduced is practically equal to

18

the amount of para isomer.

The ortho benzyl phenol Spoken above proved
identical with that obtained by Claisen's method of pro-
ducing ortho alkylated phenols. Claisen's ortho compound
is prepared by preparing sodium phenols to in toluene sol-
uticn, and later adding molecular quantities of benzyl
chloride and refluxing on an oil bath for about five hours
at 160° and allowing to stand for about twelve hours. In
a typical preparation, forty-we've: grams of phenol were
dissolved in 150 cc. of toluene, and 11.5 g. of Na added in
small pieces. This mass was refluxed on an oil bath for about
four to five hours, with occasional addition of more toluene
as it was needed to cover the reacting mass. A total of 100
cc. were added in 611. 63.6 grams of benzyl chloride were
then added, and allowed to stand for one hour. The reaction
mass was then refluxed on an oil bath, at 1500 to 160°, fa'
about five hours. ri‘he mass was then extracted with 2023 LaOH
solution, the lower dark red layer being reserved. This ex-
tract was then acidified with H01, and extracted with ether.
0n the third fractionation, 16 grams of prtho ~benzyl phenol,
b.p. 134-1400 at 2-3 mm. were obtained. This is about 17% of
the theoretical. {puch a yield was obtained twice, while
attempting to duplicate a previous yield of 4823, apparently
under the same conditions. This yield, however, could mt be
duplicated.

The ortho-benzyl phenol, so prepared, is a liguid
at ordinary temperatures, b.p. 1558-90 at 5 mm. It vri ll solidify

if kept in a freezing mixture for some time, but readily melts

19

on being brought to room.temperature, ZOO-21°C. Claisen
describes this form.as the labile form which changes on
standing to a stable form with m.p. 52°. I have samples
that were prepared a year ago and which have not changed
at all. It seems rather doubtful that they will change.

I have not obtained any sample, as yet, with a m.p. of 52°.
U. S. Pat. 1,580,053 describes this product obtained from
treating phenol with benzyl chloride. However, I have not
tried this particular method for the preparation, and the
owners of the patent are apparently unable to supply me
with any or this material.

The ortho benzyl phenol, so prepared by Claisen's
method, gives a di brom derivative, which crystallizes from
alcohol and water in fi‘e white silky crystals, m.p. 900-

919. The ortho isomer from the A1013 condensation, gives the
same d1 brom.derivative. Analysis for bromine gave the follow-
ing results: '
Calculated for CaHSoCH3.CaHzBr30H’ 46.75% Br.
Found 46.82% 46.80%
Av. 46.81% Br.

c c L Ill'lu «t nil-I! 1.6.! .m J .1 I I In t . t 1.. J... I I
a 6 4i

. a. (Fir LIE

In..." .

 

20

Part II

As previously stated, this part of the work was
designed to determine definitely where bromine substitution
took place in benzyhated phenols, in the phenol ring or
in the benzyl ring. Zincke and Walter, (Ann. pgg, 567-585
(1904) ), assumed that this substitution took place in the
phenol ring because of'the ease with which haNOa wouli react
with the compound, substituting N02 for Br, as was also trs
case with simple brom phenols.‘ Other than this assumption,
they produced no positive evidence.

To study this question, two emperiments were carried
on:

1. The ortho benzyl phenol prepared in Part I by
both Claisenis method, which produces ortho alkylated compounds,
and the ortho isomer from the A101. condensation, were bromin-
ated. A di-bram.derivative formed very readily, but fUrther
bromination was difficult. This would be orpected if the two
bromine atoms went in the 2,4 positions of the pha.ol ring.:

2. The next step was to take 2, 4 di-brom phenol,

K. 55-560, (Eastman) and convert it to the sodium salt, 81d
condense the resulting compound with benzyl chloriie, accord-
ing to Claisen's method. The resulting compound proved a: be
identical in every way with the compound resulting from.the
direct bromination of the ortho benzyl phenol. In tlis case
the benzyl group must be in the ortho or 6 position and the
structure must be as follows:
raw ' éO 7414”. 7/”
. sfi / 7‘fvé’ °cez W-
{51

 

21

Since the product of the bromination of ortho benzyl phenol
is identical with the above compound, bromination in benzyl
phenols takes place in the phenol ring, rather than in the
benzyl group attached.

The di bfom.derivative of p—benzyl phenol is like-
‘wise easily formed, but further bromination is difficult.

This compound was easily purified by distillation but was
crystallized with difficulty, until material was available
for seeding. After eight days the thick liquid did crystallize,
and on first crystallization, the crystals melted at 42~5o.
These same crystals, after standing for three days were
melted again, and melted at 56-70. These crystals had not
been touched or treated in any way, but apparently a change
took place as described by Zincke and Walter. (10c. cit.).
Since this time, the only form.that has been obtained, is the
apparently stable form, I. 56-70. It is possible that this
is true, because all samples have crystallized as a result of
seeding with the stable form. The compound crystallizes

in prismatic needles, M. 56-70.

The same product is produced on brcmdnation of p-
benzyl phenol, whether CHCla or C83 is used as a solvent,
thereby confirming Zincke and Walter’s statemmt, tint Paterno
and I"ileti (Gazz. chim. ital. é: 121—129, 251-254) rqmrted
erroneous work, when they stated that bromination in CS; scl-
ution, produced an amorphous substance, M. 175°. It is indeed

”difficult to know what they may have had in hand".

The above compounds were analyzed for bromine dan-

tent, by the Parr Bomb method, as reported by J. F. Kemp and

v..... 3|.

.1 it! 'Iwutymrihnlfiuhnw

l..— “m-lu.r.slplwul.. a

 

tr "0 !r

H. J. Broderson, (J. Ame Chem. Soc. g9, 2069).

. “-
Theory for <:::>§v<:;;jkl 46.75% Bromine.
Found I 46.28% II 46.6% Av. 46.44% Bromine.
a)
Theory for “'090 46.75;?) Bromine

31
Found I 46.84% II 46.80% Av. 46.82% bromine

The benzoates of ortho and para benzyl phenol,
were prepared and analyzed. The benzyl phenol in eithar
case was dissolved in 5% KOH, and equimolecular quantities
of benzoyl chloride added with stirring.

The benzoate of ortho-benzyl phenol has never
crystallized. The compound boils at 201-202 at 2 mm.
Analysis gave the following results.

-é’
Theory for odd 0 83.2971} C 5.59% H

Found I II Average
0 85.56% 82.95% 85.15%
H 5.58% 5.55% 5.565%

The benzoate of para-benzyl phenol crystallized well, H.P.

84-50. Analysis gave the following results:

. a
Theory for (3.90er 85.29%0 5.59% H
Found I II Average
0 85.25% 85.40% 85.55%

H 5.595% 5.62% 5.61%

An attempt was made to prepare the acetates

of both para and ortho benzyl phenol. 10 grams of benzyl
phenol were added to 17 cc. of acetic anhydride and 4.5 g.

of sodium.acetate added. The molecular ratios are l : 3 z 1
respectively. The mixture was then refluxed eight hours,
although later it was found that the same result was ob taimd
with two hours of refluxing. The mass was neutralized with
Nasco. and extracted with ether. Then purified by redistillat—
ion the para benzyl phenol product boiled at 151r2° at 2 mm.
This compound crystallized in the ice box and the crystals
melt at 31-20. The corresponding ortho benzyl phenol com!
pound boiled at 140-1410 at 2 mm.,and has never crystalltzed.
On analysis, these compounds, both isomeric, gave the fbllowing
result:

I II III IV’ V VI AV.
0 77.96% 78.31% 77.76% 77.61% 78.13% 78.01% 77.96%
H 6.08% 5.98% 6.01% 5.93% 6.02% 6.07% 6.004%

Theory for to C - 79.6%
é<§oflc
Q ' /C\ H " 6.24%

It is evident that the products obtained above are
not the acetates of para or ortho benzyl phenol. The compounds
are very stable to moisture of the air. Paterno and Fileti
(Gazz. chim. ital.1§, 121-129, 251-254) report the acetate of
para benzyl phenol as being a pale yellow liquid, nhich.is
easily hydrolyzed by.moisture of the air, and which never cry-
stallized. Their compound boiled at 317°, density at 16°,
~l.1043. The compound which I prepared with acetic anhydride
acting on para benzyl phenol, was a colorless liquid, crystalliz-

84

ed in the ice box to icy appearing crystals, M 31-2°, and
is very stable to atmospheric moisture.

Time has not permitted complete investigatiOn of
these compounds, but the work is being continued, and it is

heped that a report in the future will be possible.

25

Summary

The work reported in this thesis may be summarized
as follows:

1. The condensation of benzyl alcohol and phaiol,
in the presence of A1013, has been studied with reference to
the varying proportions of phenol and A101. added to the benzyl—
alcohol. In this particular condensation at least, the increase
in amount of phenol, increases the preportion of the ortho benzyl
phenol produced with the para benzyl phenol. 1 mol. of A101.
does not give higher yields than 1/2 mol. of A101. used as a
catalyst.

2. It has been.definitely shown, that bromination of
ortho benzyl phenol, takes place in the phenol ring rather than »
in the benzyl radical. L\r

3. 2-benzy1, 4-6odi brom.phenol has been prepared and
described for*the first time. It was prepared by Rennie, (Jaur.
Chem. Soc.‘gg, 406) as a by—product of the study of para benzyl
phenol, but its structure was not determined.

4. The benzoate of ortho benzyl phenol has been prev
pared and described.

5. Benzyl phenols have been treated with acetic anhy-
dride, and products Obtained which apparently are not acetates,
but the exact nature of’the compounds has not yet been determined.

This work is being continued,in the hOpe of being able to make

a later report.

N STA E UNIVERSITY LIBRARIES

HI Tillyllllllilllll I

1293 3145 2620

 

WI
3