Original filename: ie50470a017.pdf
This PDF 1.3 document has been generated by Acrobat Capture 3.0 / Adobe PDF Library 4.0, and has been sent on pdf-archive.com on 30/09/2016 at 22:43, from IP address 37.239.x.x.
The current document download page has been viewed 274 times.
File size: 1.1 MB (8 pages).
Privacy: public file
Download original PDF file
INDUSTRIAL AND E N G I N E E R I N G C H E M I S T R Y
Yield, wt. %
See Figure 12.
Tall Oil Fractionation I"
Yield, wt. %
F a t t i acids,
Rosin acids, lo
See Figure 13.
j?oslr A = , ~ -
Tall Oil Fractionation IIa
F A T w
Figure 13. Tall Oil (Table XI)
In a rosin acid fraction of 73%, while the fatty acid content
dropped to 20%. The saponification number followed this
change by dropping from 175 to 165.
Figure 13 shows a similar tall oil fractionation where an
attempt was made to isolate a fatty acid-rich fraction. In this
case, a 25% cut was removed as an overhead product, after the
removal of a 5 % color body fraction, and a 70% rosin acid-fatty
acid fraction. Analytical results on these fractions are shown
ln Table XI. The fatty acid value in fraction I11 increased to
87%, while the rosin acid content dropped t o 6%. Additional
studies on tall oil are being carried out in attempts to improve the
decolorization of the oil, as well as the fatty acid-rosin acid
Many of the results obtained and pictured in this paper were
the initial runs made on oils, and more extensive studies have been
and are being carried out on the various oils. It appears from
the work which has been conducted thus far that the application
of the Solexol process in the glyceride oil field should bring about
some striking changes in the refining of various fats and oils, and
its commercial acceptance is already definitely established.
(1) Drew, D. A., and Hixson, A. N., Trans.Ant. Znst. Chem. Engra.,
40, 675 (1944).
(2) Hixson, A. W., and Bockelmann, J. B., Ibid., 38,891 (1942).
(3) Hixson, A. W., and Hixson, A. N., Ibid., 37,927 (1941).
(4) Hixson, A. W., and Miller, R., U. S. Patent 2,219,652 (1940).
(5) Ibid., 2,226,129 (1940).
(6) Ibid., 2,247,496 (1041).
(7) Ibid., 2,344,089 (1944).
(8) Ib$d., 2,388,412 (1945).
(9) Larner, H. B., U. S. Patent 2,432,021 (1947).
(10) Schaafsma, A., Ibid., 2,118,454 (1938).
(11) Van Orden, L., Ibid., 2,394,968 (1946).
RBCIIYED February 26, 1948.
Synthetic Drying Oils
Don S. Bolley
National Lead Company, Brooklyn I , N . Y.
'The drying oil chemist has become accustomed to consider synthetic drying oils as oils prepared through chemical treatment of fatty oils. A n extensive and critical literature review of drying oils of this nature was made. These
included dehydrated castor, methods of increasing un*aturation, maleic oils, drying oil esters of polyhydric
alcohols, nonfatty drying oils, and copolymerized drying
ails. Experimental data are given on the comparative
properties of linseed pentaerythritol ester with linseed ail
and soybean pentaery thritol ester with soybean oil.
materials is given in Table I. Drying oil, as used in this discussion, means an oil-like material which dries by oxidation when
exposed in a thin film to the air. This might be considered a
restricted definition; the more general definition is: an oil-like
material which dries when exposed in a thin film. The general
term would include oxidizing resinous solutions dissolved in
thinner which dry by solvent evaporation.
A raw drying oil is one obtained from the seeds or nuts directly
by hydraulic pressing, expelling, or solvent extraction. I t has
had no additional treatment except possibly a storage period to
allow suspended material to settle out. The soybean and edible
HE term, synthetic drying oils, as used by the paint and
varnish chemist, as well as the drying oil chemist, might be
misleading t o those outside of the field. Synthetic drying oil
usually means an oil which has been prepared through chemical
treatment of fatty oils. Less frequently, i t may refer t o drying
materials of an oily nature which have been made from nonfatty
Since the drying oil chemist uses a variety of special terms in
reference to oils, a suggested classification of various drying oil
Table I. Drying Oil Materials
Raw drying oils
Refined drying oils
Modified drying oils
Synthetic fatty drying oils
Synthetic nonfatty drying oils
Copolymerized drying oils
Drying oil resins
INDUSTRIAL AND ENGINEERING CHEMISTRY
oil industry quite often refer to these so-called raw oils as crude
To explain the next classes, it would be best to consider the
composition of an oil. Roughly, it is made up of minor components and triglycerides. The triglycerides consist of materials
having poinB of unsaturation and ester groups. A refined oil
then may be considered one in mhich the minor components have
been partially removed or modified. This includes alkali treatment to remove free fatty acids, bleac3hing to remove coloring
matter and other materials, refrigeration to remove waxes, and
acid treatment to remove mucilaginous materials. A modified
drying oil is one in which the double bonds of the triglyceride arc
affected. This would include heat bodying, blowing, maleic
addition, and isomerization. A synthetic fatt,y drying oil may be
thought of as one in which t'he ester grouping has been altered.
Thus, if a natural drying oil is hydrolyzed to fatty acids and
glycerol, and the fatty acids re-esterified with another polyhydric
alcohol such as pentaerythritol, t,he product is called a synthetic
drying oil. The source of a synthetic, nonfatty drying oil is
derived from substances other than fats, These may be synthetic unsaturated materials, modified petroleum products, etc.
A copolymerized drying oil, as used in this discussion, is considered one in which polymerizable unsaturated materials are
added to fatty materials and the whole intcrpolymerized, usually
using heat. An example would he styrene heated with dehydratcd castor oil.
A drying varnish might be considered a resin dissolved in oil
with or without the presence of suitable solvents, such as an ester
gum linseed varnish.
A drying oil resin is a material mhich will oxidize on exposure
to air, and is of a solid nature in the absence of solvents-for
example, a drying alkyd.
These classifications arc not proposed definitions; it is realized
that many drying oil materials might be considered to fit several
of the classes. Hoxwver, this scheme of classification has proven
useful in the writer's laboratory.
The following discussion will be limited to two types of
modified drying oils (dehydrated castor and maleic), synthetic
fatty and nonfatty drying oils, and copolymerized drying oils.
Many of thc oils discussed are covered by patents.
Dehydrated Castor Oil
The modern type of dehydrated castor oil, under the above
clawification, would be considered a modified oil. However, since
it was originally a synthetic oil, and many people consider it a
synthetic oil, this discussion is believed rightfully to come under
the title of this paper. It was the first synthetic drying oil to
attain general commercial usage and is still of the greatest interest
to the paint and varnish chemist of all the so-called synthetic
drying oils. Dehydrated castor oil was originally made by
Seheiber (SQ)in about 1930 by dehydration of castor oil fatty
acids. The following equations show the reactions involved.
,_ - - - - -
1 - 1- HOH
- _ - - _ _ I
When castor oil is hydrolyzed, the fatty acids arc found to be
composed of about 85% ricinoleic acid. This acid contains a
double bond in the 9,10 position and a hydroxyl group on the 12th
Vol. 41, No. 2
carbon atom. Scheiber found that by subjecting this acid to a
distillation treatment, a more highly unsaturated acid is produced; this could be combined with glycerol to produce a drying
oil which was known as Scheiber oil. The reaction may take
place by the elimination of water from the hydroxyl group on the
12th carbon atom and a hydrogen on the l l t h , or by the hydroxyl
group on the 12th carbon atom and a hydrogen on the 13th. In
the former case, a conjugated acid is produced, while in the lattcr,
a nonconjugated acid results. Scheiber claimed a preponderance
of conjugated acids formed, but later expcrimcnts indicate that.
this is not the case.
Shortly after Scheiber's discovery, it was found that castor oil
could be dehydrated by the use of catalysts without resorting t o
hydrolysis. The type of catalyst (25) generally used for this
purpose is of an acidic nature, such as sulfuric acid, potassium
and sodium acid sulfates, phosphoric acid, phthalic anhydride,
acid earth, etc. The oil and a small amount of catalyst are
charged into a large stainless steel kettle which is heated to 450"
to 475" F. with agitation in a vacuum to facilitate removal of
water. When the reaction is complete, the catalyst is removcd
by filtration. More modern mcthods involve the use of a continuous method of treatment and soluble catalyst.
There has been considerable discussion on the percentagc of
conjugated and nonconjugated bonds formed by the reaction.
The diene determinat>ion using maleic anhydride indicates that
commercial products contain 17 to 26% conjugated and 59 t D
64% nonconjugated acids. In other words, i t is usually coilsidered that there are three nonconjugated groups formed to every
conjugated. Nore modern technique, using the ultraviolet
absorption spectrophotometer, indicates a slightly higher percentage of conjugated acids, perhaps about 2.5 to 1. The dehydration process apparently is never quite complete as the present
dehydrated castor oils have a hydroxyl value of 12 to 20. However, this is a great improvement on the original dchydratcd
castor oil which had a high pcrcentage of hydroxyl groups which
result,ed in sticky films because of the plasticizing action of free
The properties of dehydrated castor oil mighl be consitlered ar
lying between those of linseed and tung. However, somc caution
should be used in applying this generalization. Drying time, rate
of heat polymerization, and resistance to water and alkali by
varnish films are intermediate between linseed oil and t,ung oil.
They produce films which are softer than those of linseed and
much softer than tung, but of superior elasticity than either.
Because of its lack of triply unsaturated acid, the resistance to,
yellowing is outstanding.
In about 1930, a process for making a triply unsaturated con-jugated ester from castor oil by the Miinzel proceas was an-.
nounced ( 4 ) . (The Munzel works are located in Switzerland.)
Castor oil is dehydrated using a freshly precipitated tungstic acid
anhydride. The Munzel process claims this produces nearly
exclusively conjugated dehydrated castor oil. Workers in this
country have found that while tungstic acid anhydride can be
used efficiently to dehydrate castor, about the samc proportion
of conjugated and nonconjugated oils are formed &s from othcr
catalysts. Munzel's conjugated dehydrated castor, known as
Dienol, then is treated with hypochlorous acid. This adds a
hydroxyl group to the 9th carbon atom and a chloride group t o
the 12th. By simult,aneous removal of hydrogen chloride and
mater, a triply unsaturated conjugated est,er is formed. Thc
double bonds during Eormation are said to shift over one carbon
forming Trienol which has the points of unsaturation a t the same
place as eleostearic ester or tung oil. There seems to be considerable doubt as to the actual production of this oil. Some
Trienol was received in this country and found to bo equivalent,
if not identical with, tung oil. All outside attempts to prepare
Trienol have been unsuccessful.
I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY
There are a number of references, mostly patents, on increasing
the unsaturation of fatty oils of low unsaturation through chlorination followed by removal of hydrogen chloride (26, 40).
Theoretically, every double bond so chlorinated would yield two
double bonds in the conjugated position. The difficulty with the
procedure is that usually dark colored resinous materials of poor
drying characteristics are formed rather than the improved oil
one might hope for when working out the reactions on paper.
The writer does not know of any commercial oils made in this
A similar reaction involving hydroxylation and dehydration
seems t o offer greater promise. Light colored hydroxylated oils
can be made in a variety of ways and the methods of catalytic
dehydration used on castor oil may bo applied (13, 36). Again,
as far as the writer is aware, this method for increasing unsaturation and producing conjugation has not been used commercially.
However, it might be pointed out that blowing a n oil with air
forms some hydroxyl groups and on subsequent heat treatment
this hydroxylated oil will be dehydrated and show increased
bodying rate. I t s absorption spectra shows the presence of some
Rlaleic modified drying oils are an interesting and extremely
versatile type of material. They may be considered under either
oils or resins. A paper by K. A. Earhart given at this symposium presents some applications in the field of resin chemistry;
this discussion, therefore, will be limited to the mechanics of the
reaction and the preparation and properties of materials that
would be considered drying oils. The reaction involves addition
of maleic anhydride to the double bonds of a drying oil with subsequent modification of the maleic’s acidic groups.
I n 1928, Diels and Alder (17) announced their famous reaction
on the addition of maleic type materials to conjugated double
bonds. This reaction is known as the Diels-Alder reaction or
diene synthesis. Diels and Alder were originally concerned with
quinones, although it soon became apparent their reaction could
be generalized to many unsaturated materials.
In 1930, Boeseken and Hoevers (5),in France, published an
article concerning the action of ma’eic anhydride on dehydrated
castor oil acids and esters. They used the diene synthesis to
explain the reaction between the conjugated fatty acids of dehydrated castor and maleic anhydride in accordance with the
both analysis before and after the reaction for hydroxyl value, aa
well as the use of dehydrated castor containing few hydroxyl
groups, seemed t o eliminate this postulate. Since maleic in this
concentration does not gel other oils, the study of this phenomenon might shed some light on either the composition of dehydrated castor or the reaction mechanism. It was suggested a t this
symposium (14) that the polymerization of maleic anhydride
with dehydrated castor oil might be analogous to the reaction of
maleic anhydride with cis- and trans-piperylene (16). The ciscis, 9,ll-octadecadienoic acid which probably occurs in dehydrated
castor oil (SO) would be expected t o copolymerize with maleic
anhydride, whereas the cis-trans, trans-cis, or trans-trans would
be expected t o form a stable adduct, nonpolymeric in nature.
Other oils tested with maleic anhydride apparently do not
copolymerize readily, but form adducts. However, recent work
indicates that some copolymerization occurs with even linoleic
Tung oil is composed mainly of glyceride ester,, containing
three conjugated double bonds. Morrell (sd), in 1932, studied
the reaction between maleic and tung oil acids and found that
maleic reacted with one set of conjugated double bonds for CYeleostearic and another for p-eleostearic. Sometime later,
Carleton Ellis obtained several U. S. patents (2f-%S) treating
with the usefulness of the reaction product between maleic and
tung oil. He found an aqueous soluble tung oil compound could
be made by treating the anhydride groups of the combined
maleic with alkali or ammonia. The possibility of preparing a
rubbery-rosinous eompound by treating the adducts with
polyhydric alcohols also was shown. However, little commcrcial
use seems to have been made of the tung maleic adduct.
Edwin T. Clocker, in this country, recognized that maleic
anhydride and similar substances would react with nonconjugated
oils as linseed, soybean, etc., and has obtained a series of U. S.
patents (8-19)on this reaction and its application to the paint
and varnish field. Bevin, in England, has obtained a somewhat
similar set of British patents (2, 3 ) . Root (87) claims the reaction of maleic anhydride with an oil such as linseed can be
facilitated by a peroxide. As a result of these studies, it is now
recognized that maleic anhydride reacts with only conjugated
type materials a t an appreciable rate up to near 200’ C. and with
all types of double bonds that appear in fatty acids a t temperatures above 200’ C. There has been considerable speculation
as to the possible mechanics of the reaction with nonconjugated
systems. Clocker originally assumed that the reaction went in
accordance with the following equation:
By using this reaction, Boeseken and Hoevers concluded that
dehydrated castor fatty acids were made up of 75% conjugated;
this subsequently has been shown to be too high. When one
attempts to make an adduct with 10 parts of maleic anhydride
and 90 parts of dehydrated castor, a gel results (6). This can be
prevented by carrying on the reaction in the presence of a nonconjugated oil such as linseed. The reason for the gel is not
understood. It was originally thought that it might be due to
the hydroxyl groups remaining in the dehydrated castor oil, but
While this explains the resulting properties of the oil, the reaction does not seem plausible because of the assumption of a
butane ring which is difficult to form in this manner. I n addition, it has been shown that when the reaction is applied t o
materials such as oleic esters, the oleic adduct still retains the
unsaturation which would not be explainable by this reaction.
To bring the reaction into line with the original Diels and Alder
concept, i t has been postulated that nonconjugated bonds isomerized under the influence of heat to conjugated bonds which
subsequently reacted as follows:
INDUSTRIAL AND ENGINEERING CHEMISTRY
Other studies have shown that corijugation iesults irom heat
treatment of oils. However, since the addition reaction proceeds a t an appreciable rate a t 200" C., it IS doubtful if such
conjugation can completely explain the reaction mechanism, It
would be totally inadequate to explain the reaction betwerri
maleic and oleic which has only one double bond.
To account for the retention of unsaturation, the folloaing tn o
reactions are proposed:
Of the tTTo, it is believed that the second, involving a hydrogen
shift from the methylene group, more nclarly f i t q modcrn theory
of chemical reactions. Thus, the writer believes the main reaction is probably this reaction with some conjugated reaction a i
previously indicated More work is necessary to definitely
establish these postulates.
The interesting feature of maleic oils is that one has a simple
process for adding on acidic groups in the middle of the oil molecule.
These groups make the oil useful as a grinding medium as well
as give the oil compatibility with ethyl and nitrocellulose. The
acidic groups may be reacted with a variety of substances such ah
alkalies, polyhydric alcohols, and basic dyes to make the oil
useful in water paints, textile sizes, cotton printing, adhesives, or
Vol. 41, No. 2
At present, the three inost important uses of maleic
oils are as water vehicles, improving the drying and heat bodying
properties of nonconjugated fatty drying oils, and the preparation
of resins. The first two v d l be considered here.
Maleic adducts are simple to prepare. The process merely
involves heating maleic anhydride with the oil in a closed vessel,
properly vented, with stirring until completion of the reaction.
Use of copper, iron, or mild steel should be avoided as the maleic
is corrosive. Glass, stainless steel, and aluminum are satisfactory.
For example, a 10% maleic linseed complex may be made by
heat,ing 90 parts of a refined linseed oil with 10 parts of maleic
anhydride with stirring to 200" C. in 1 hour, holding a t this ternperature for 1 hour, raising the temperature to 230' C. in 0.5 hour,
and holding for 2 hours. Testing the cooled adduct' by washing
out any free maleic anhydride will show the reaction to be over
99.7% complete. The time and temperature of heat'irig should
be greater for soybean and other more saturated oils.
Maleic oil adduct,s may be rendered water soluble by iieutralising the anhydride groups with inorganic alkalies, ammonia, or
amines. Ammonia seems to be t,he preferred material. Adducts
cont,aining a relat,ively high percentage of combined maleic anhydride, when neutralized with ammonia, are completely miscible
with water. Water paints and other coating materials cont'aining
water may be made from t,hese vehicles. However, present practices tend toward the use of lower percent,ages of maleic and a half
methylated-half neut,ralized product. Such a vehicle, nhile not
completely soluble, readily emulsifies with water and may be
used in the production of the emulsion type of water paints.
The base vehicle is made from the maleic oil adduct (usually
containing about 10% maleic) by first adding equal maleic molar
quantities of methyl alcohol. This transforms the anhydride
group into a methyl ester plus an acid. The carboxylic acid
group then is neutralized with ammonium hydroxide. These oile
are sometimes referred to a9 solubilized oils.
Water paints properly formulated with the solubilized oils dry
readily to a good film. One might expect them not to have good
wat'er resistant properties. However, on drying, the oils lose
their hydrophilic properties and have been found to have good
washing resistance. The nature of this inversion is not well
understood. J. C. Coman (16)believes it could be explained
by the fact that these oils dry to give polymeric materials and the
presence of a few maleic anhydride groups on the polymer molecules mould not make t,hem water soluble, whercas one group on
a single fatty acid or oil might make it water soluble.
Borrowing a reaction from the alkyd resin chemist, the maleic
oil adduct may be react,ed wit,h polyhydric alcohols such as
glycerol or pentaerythritol to increase its complexity. When
only a small amount of maleic anhydride is used, less than 5%'
the oil is only somewhat hodied and may be used without thinner.
However, if larger quantities of maleic are used and esterified, II
viscous material results; this is claimed by the resin chemist.
As an example, a useful maleic oil may be prepared as follows:
A 10% maleic linseed oil complex is prepared as previously de-.
scribed. This is diluted v,.ith an equal weight of refined linseed
oil, which result,s in essentially a 5% complex. A stoichiometric
quantity of pentaerythritol is a,dded to the oil a t about 190" C.
during the course of 4.5 hours. At the end of this addition, the
mixture is heated under mild vacuum for 2 hours. 4 slight
turbidity usually rcsults, which may be removed by filtration.
The oil is then heat-bodied a t 260" C. to a 22 viscosity in an
atmosphere of carbon dioxide. The resulting oil is of light color
with an acid value of about 10.
I n general, the maleic oils may be used to advantage in paints
and varnishes. They dry faster and heat-body considerably
faster than t,he oil from which they are made. Although of
increased ester content, their water resistance is good. In varnishes, they cook rapidly and easily with most resins with the
exception of limed rosin. They impart improved drying and
hardness t o the varnish films.
INDUSTRIAL AND ENGINEERING CHEMISTRY
Swain (43) suggests a variation in the above described procedure for preparing maleic oils. A diglyceride is first made by
ester interchange with 2 moles of glyceride oil plus 1 mole of
glycerol. The diglyceride is then reacted with a maleic adduct
to obtain a similar type of product. Schwarcman (41, 4%')
suggests preparing oil with free hydroxyl groups by partial ester
interchanging with glycerol, pentaerythritol, or mannitol with
subsequent reaction with maleic anhydride.
There are a number of other compounds such as fumaric, itaconic, and aconitic acid that react similarly, but at a slower rate
than maleic. However, because of the ease with which maleic
anhydride adds on to oil and the price oonsideration, little use
seems to be made of them in the abnormal diene synthesis.
Synthetic Fatty Drying Oils
Drying oil fatty acids may be esterified with various polyhydric
alcohols. The esters so formed are usually referred to as synthetic drying oils. Various alcohols of interest are shown by the
Mannitol and Sorbitol
The folloa ing generalizations are frequently referred to in drying
oil theory and have been confirmed on an experimental basis.
As one might expect from the theory of functionality, the
methyl esters of drying oil fatty acids, while oxidizing at about
the same rate as oil, do not form a solid film. When ethylene
glycol is esterified with fatty drying oil acids, a slow forming soft
film may be obtained from those fatty acids, such as tung and
perilla, that have a large amount of unsaturation. This film,
however, would not be sufficiently tough to find use in the paint
and varnish field. The glycerol ester of fatty acids, of course,
is similar to the glycerides formed in nature. There is one
Interesting difference: the synthetic glyceride oil has a short
induction period, while the natural product, particularly raw
oils, has a pronounced induction period, probably due t o the
natural antioxidants contained in the minor constituents. From
the performance of these three esters, it would be expected that
as complexity of alcohol increases, the rate of drying and heat
polymerization, as well as the toughness of the film should
increase also. This has indeed been found the case and is the
basis for the value of the various synthetic oils made from higher
Erythritol is the next higher polyhydric alcohol with a structure similar to glycerol, Fatty esters of this material have been
found by this laboratory to dry somewhat faster and harder than
the glycerides. However, some difficulty attributable to secondary alcoholic groups has been encountered in this esterification
and the erythritol is fairly expensive. Little work has been done
on this material, and insofar as the writer is aware, no commercial
application of it has been made in the field of synthetic oils. I t
should not be confused with pentaerythritol, which will be discussed in detail later. Mannitol and sorbitol are hexahvdric
alcohols further along in this series. The chemical structures
of the two materials are similar except for spacial arrangements
of the hydroxyl groups which are not incorporated in the formula
given. Their properties are similar and synthetic oils resulting
therefrom, while showing some slight differences, may be considered alike. A great deal of work has been done on the preparation of synthetic oils from these materials since they are of reasonably low cost and are readily available. It has not been found
possible, without the use of special tricks, t o add six fatty groups
to these alcohols. When mannitol and sorbitol are heated to
the temperature necessary to esterify them direct with fatty
acids, they have a tendency to split off a molecule of water and
form an inner ether. This prevents the complete esterification
of the hexahydric alcohols with the fatty acids. It has been
found that about 4.5 moles of fatty acids may be added direct
t o the mannitol and sorbitol, I n addition, since two thirds of
the hydroxy groups are secondary alcohols, this adds to the
difficulty of esterification. It has been found that a low temperature, long heating, and stepwise addition of the alcohol
assist in the preparation of a good product. These synthetic oils
dry more rapidly and with a harder film than the glycerides.
Varnishes may be prepared although they are somewhat dai k and
many of them are not too satisfactory. Our observations have
shown that outdoor paints formulated with esters have stood up
well on over 5 years' exposure. It is hoped that the research
work now in progress will allow more complete esterification of
these two alcohols; this should result in an improved product.
Continuing with increasing complexity of alcohols of this type
leads to polysaccharides and starches. Some work has been done
along this line, although a great deal of difficulty is encountered
when using unsaturated fatty acids for their esterification. As
far as is known, there is no synthetic oil on the market made from
the fatty drying acids and these high polyalcohols.
Pentaerythritol has proved the most useful of all the polyhydric alcohols in the preparation of synthetic oils; the formula
shows that it contains four hydroxy groups, all primary. It is
easy to esterify, and because of its increased functionality, gives
harder drying oils.
A typical pentaerythritol oil has been made in this laboratory
as follows: 1.05 moles of technical pentaerythritol are added t o
4 moles of linseed fatty acids. The mix is heated gradually to
200" C. with stirring and passing carbon dioxide over the mixture.
Conducting under a mild vacuum is preferable. At 200' C. a
homogeneous solution results and i t is held a t this temperature for
2 hours. The temperature is raised to 230" C. in about 0.5
hour and held for 3 hours. At this time the esterification is substantially complete. It is then held at 250' C. until the desired
viscosity is reached. At a viscosity of G, the oil should have an
acid value of less than 5.
Table 11, compiled from data observed in these laboratories,
compares the properties of a pentaerythritol ester of linseed, a
INDUSTRIAL AND ENGINEERING CHEMISTRY
Properties of Oils and Pentaerythritol Esters
Set to touch, hr.
Alcohol benzol, %
I l f
similar bodied linseed oil, a peritaerythritol ester of soybean, and
a similar bodied soybean oil. For comparison, oils of similar
viscosity should be used. While the examples are not identical
in viscosity, i t is believed they are sufficiently close to permit
The color of the synthetic oils is somewhat darker t,han usual.
Other products prepared are only slightly darker t h m refined
linseed or soybean oils. The acid value indicates little reactivit,y wit'h basic pigments; this has been found t o be t>rue. Saponification values of the synthetics are always somewhat below the
natural glycerides. The hydroxyl values of synthetics are always
high, akhough the soybean- shown here is perhaps abnormal.
The heat bodying test on these oils was conducted in a quart
metal beaker and is instructive. The rate of bodying of the linseed pentaerythritol ester is about. twice that of linseed oil, while
the linseed oil and soybean pentaerythritol esters are similar;
soybean oil is slow. The results indicate the synthetic esters to
be more tack-free after 24 hours than the natural glycerides.
Water resistance of the various oil films is somewhat incousistent. These results show that, in general, the pentaerythritol
synthetic oils are somewhat inferior t'o the nat,ural product. The
reverse is often true when the oils are used in varnishes.
Film solubility ( 7 ) is measured by exposing the oil on sand for
3 weeks under a sunlamp in the constant temperature room. The
sand-oil mixt.ure then is extracted with the various liquids for 3
hours in a Soxhlet thimble. The last section in Table I1 shows
the results obtained. This type of data has not been correlated
with actual performance. However, in general, the lesser the
amount extracted, the higher the complexity of the film; this
shows up in hardness, mater resistance, etc. Acetone is perhaps
the best solvent for this purpose although all tell about the same
story. Acetone extracts about one third of the material from a
linseed pentaerythritol film, about one half from both linseed and
soybean pentaerythritol oil films and netirly all of the soybean oil
film. These results would change on aging; the soybean oil film
would become more insoluble, whereas the others would reach a
minimum, and then increase.
Varnishes have been made up with a variety of resins and their
properties thoroughly tested. Paints made from these types of
oils have been exposed for 6 years outdoors. The story on both
varnishes and paints is about the same as would be expected from
the results of the films. Varnishes made from pentaerythritol
esters are, in general, superior t o t,hose of the natural oil. In the
Vol. 41, No. 2
case of paints, the linseed peiitaerythritol showed only slight
improvement over the linseed oil whereas the soybean pentaerythritol ester was a distinct improvement over the soybean oil,
particularly the initial phases, and might be considered similar
to linseed oil in most of its properties. This latter fact is highly
significant since attempts are being made to introduce more and
more soybean oil into paints and varnishes in place of linseed.
Although some caution should be exercised, it is believed the
soybean pentaerythritol ester may be used in most cases interchangeably with linseed oil.
Techiiical pentaerythritol usually contains about 15% dipentaerythritol. Fortunately, this mixture seems t o be bettcr
adapted for the paint and varnish field than the pure and more
expensive explosive grade of C.P. pentaerythritol. Pent,aerythritols conhining larger amounts of dipent,aerythritol and tripentaerythritol are commercially available. synthetic oils prepared using the polypentaerythritols heat-body considerably
faster, while drying is soniewhat,faster. They may be considered
a harder drying oil than the esters made from the technical
Inositol is a cyclic hcxahydric alcohol. At present, it, is
expensive, but since it occurs widely in nature combined with
organic phosphorus a large commercial use should permit its
product,ion on a more economic basis. When 6 moles of fatty
drying acids are esterified with 1 mole of inositol, the resulting
syntheric oil dries to a hard tack-free film. Inositol linseed
synthetic oils have been prepared by this laboratory; they dry
much faster than linseed, although occasionally an oil results
which actually dries slower than linseed. This is believed to be
due t,o various antioxidants eirher present, iri the inosit.ol or
formed during esterification. Additional research worlc should
iron out these difficulties, aiirl if the price of inositol could be
reduced, a new interesting oil would be available.
Synthetic oils similar to those described above may br prepared also by ester interchange and alcoholysie. For example,
a nat,ural glyceride may react with methyl alcohol resulting in the
methyl ester plus glycerol. The methyl ester then may be reacted with mannitol to form the synthetic mannitol oil with the
regeneration of methyl alcohol. Since all these reactions are
equilibrium reactions, removal of one of t,he constituents allows
the reaction to go in the desired direction in accordance with t,he
laws of mass a d o n . The polyhydric alcohol ester--for example,
mannitol--can be made direct from the glyceride by treatnient
with mannitol and removal by distillation of the glycerol. 1Iowever, this is apparently more difficult than the previous method
described. These are all examples of alcoholysis. An example
of ester interchange would be treat,ment of triacetin with the
methyl esters of fatty oils and production of the glyceride during
removal of methyl acetate. Further study is necessary to make
t,his type of rcaction commercially practicable.
Little mention has been made of the various catalysts which
are used in esterificat,ion. These include inorganic alkalies,
alcoholat,es,litharge, and other inorganic oxides, acids, and salts.
hlthough many of these assist, in the rate of esterification, they
are not nccessary for the regular rcaction between alcohol and a.
fatty acid. I n fact, they oftcn give precipitates and gels and
darken the product,. However, they are highly essential in ester
interchange and alcoholysis reactions.
A large number of other polyhydric alcohols exist; many have
made interesting oils and many remain to be tried. I n addition,
research work is continually in progress on the preparation of new
polyhydric alcohols which would be of interest to both the oil and
Tall oil is a by-product of the sulfate pulp industry when pine
wood is pressure-cooked with caustic soda solution and other
chemicals. Tall oil is not an oil, since it is made up of mixtures
of resin acids, fatty acids, and various unsaponifiable materials.
The refined product may be esterified with the various polyhydric alcohols previously mentioned. The most interesting
INDUSTRIAL AND ENGINEERING CHEMISTRY
product, the pentaerythritol ester, will dry in a reasonable time
to a hard tack-free film. This product may be considered similar
to about a 12-gallon ester gum linseed varnish. It may be used
for interior paints, but only small portions should be used for
exterior paints because of its high rosin content. The product
may be improved further by treating with maleic and esterifying,
as discussed under Maleic Treated Oils.
A reverse ester was prepared by reacting fatty alcohols with
polycarboxylic acids. For example, linseed alcohols were made
from linseed methyl esters by sodium reduetion. They also
may be made from a variety of other fatty derivatives such as
the acids, nitriles, and esters by high pressure hydrogenation.
Tricarballylic acid is similar to glycerol except that it has three
carboxyl groups in place of three hydroxyl groups. Tricarballylic acid then was reacted with the linseed alcohols in a
manner similar to the normal esterification, and a linseed tricarballylate obtained. This ester, which has the same functionality as a glyceride, was found t>oform dry films and act similarly
to the glyceride. However, it was noted that during this preparation using sodium, some conjugation took place; this gave the
oil the various properties which would be obtained from a somewhat conjugated glyceride ester of linseed.
An infinite number of possibilities for the preparation of various
synthetic oils exists. Fatty materials can be reacted with other
than polyalcohols such as various polyamines, amides, phosphorous compounds, and silica compounds. Future research
should turn up many interesting compounds of this nature.
Synthetic Nonfatty Drying Oils
For a number of years various synthetic nonfatty drying oils
have been prepared from petroleum. During the cracking process
for the production of gasoline, various unsaturated hydrocarbons
are formed. These are removed, polymerized, and frequently
oxidized. The result gives a material which dries when exposed
t o air as a film. Other methods used are chlorination and dechlorination to increase the unsaturation. These substitute petroleum drying oils which have come to the writer's attention dry
slowly with an initially.soft film and some yellowing. I n addition,
on aging, they check badly. Undoubtedly, they could be used in
small quantities to extend natural drying oils, but the ones examined fall far short of exhibiting the desired properties for a
paint and varnish oil. The writer has been informed that new
petroleum drying oils of greatly improved properties are about to
A large variety of film-forming materials can be prepared from
synthetic unsaturated materials not derived direct from the petroleum unsaturates. These include polymerized acetylene,
polymerized butadiene, and acrylates and other vinyls. During
the last war, the Germans, because of the shortage of natural fats,
expended a great deal of energy in attempting t o develop products as drying oil substitutes. These substitutes now are being
evaluated by the industry, and i t is probable that some will prove
meritorious. In addition, a large number of new and cheap unsaturated products are being introduced commercially. The
polymerization products of these unsaturates have commercial
possibilities in the paint and varnish field. Foremost among
these products are those prepared from acetylene by the new socalled Reppe chemistry.
defined as any unsaturated organic compound capable of forming
chain polymers through its double bonds.
In recent years there has been an increasing interest in these
relatively new fatty oil copolymers as is evidenced by the more
frequent mention of them in the literature, particularly in
patents. At the present time, however, the literature does not
reveal much concerning the structure of the resulting copolymer.
Bearing in mind the slow progress made so far in elucidating the
mechanism of the drying and polymerization of fatty oils, it is
doubtful whether much more than general principles will result
from investigations into the structure of the fatty oil copolymers.
Many different fatty oils have been used as starting materials.
I n this respect, it is apparent that conjugated oils-tung, oiticica,
and dehydrated castor (90,98)-react with active unsaturated
compounds more readily than the nonconjugated type. I n
general, i t has been found necessary t o give the latter type (linseed and soybean) some pretreatment such as partial polymerization (46) or oxidation (18) to obtain homogeneous copolymers.
I n some instances, preformed varnishes (94),oil modified alkyd
resins, or concentrated standoils ( 9 9 ) have served as the fatty
oil portion of the copolymer. The list of unsaturated compounds
which have been reported t o copolymerize with the fatty oil coniponent includes styrene (228,W), a-methylstyrene, acrylic and
methacrylic acids and their esters, vinyl halides (Sf?), vinyl esters
(SS), vinyl ethers (SI),cyclopentadiene ( d 7 ) , acrylonitrile (It?),
butadiene ( I ) , diallyl maleate (19), allyl esters of dimeric fatty
acids (&), acyclic terpenes (B),and furylethylene (54). References to styrene outnumber all other compounds listed above.
Thus, it seems safe to say that it lends itself readily to being the
second component of fatty oil copolymers. Several copolymers
made with styrene have been marketedrecently. Cyclopentadiene
and dicyclopentadiene also are particularly adaptable to copolymer formation.
End products which have resulted from copolymerization of the
two classes of compounds just mentioned range from oily liquids
capable of air drying in the normal manner when exposed as a thin
film through solid polymers. Between these extremes, there are
products which require stoving for conversion to solid films.
Claimed uses for the fatty oil copolymers include: all kinds of
coating compositions such as paints, varnishes, and enamels;
adhesives; plastic compositions such as linoleum; and resins
suitable for use in paints and varnishes. Professed generalized
advantages of the copolymers, depending on composition, are
production of fast drying, tough, flexible, adherent films showing
strong resistance to water and alkali.
At the present time it is not possible to predict just how valuable the fatty oil copolymers will become. Either they will show
to advantage only in certain specialty uses, or they will exhibit
properties that will cause their adoption on a much larger scale.
The latter seems probable when it is considered that development
work has only recently been undertaken in earnest and that new
unsaturated compounds are constantly being made available on a
Ambros, O., and Reindel, H., German Patent 523,033 (Feb. 5,
Copolymerized Drying Oils
Bevan, E. A., British Patents 500,349 and 500,350 (Feb. 6,1939).
Bevan, E. A., and Tervet, J. R., British Patent 500,348 (Feb. 6,
Blom, A. V., Paint Oil Chem. Rev., 101, No. 15,9 (1939).
Boeseken, J., and Hoevers, R., Rec. trau. chim., 49, 1165-8
The term copolymerization has conie to be known as the reaction of two different polymerizable materials with each other
to give a homogeneous polymer built up from both materials.
The resulting copolymer often has more desirable properties than
the polymer of either of the two materials alone. This discussion
will deal with those copolymers which have as one component a
drying or semidrying oil. The second component might be broadly
Bolley, D. S., U. 5. Patent 2,414,712 (Jan. 21, 1947).
Bolley, D. S., and Gallagher, E. C., J. Am. Oil Chemists' SOC,
24, NO. 5 , 146-9 (1947).
Clocker, E. T., U. 8. Patents 2,188,882-90 (Jan. 30, 1940).
Ibid,, 2,262,923 (Nov. 18, 1941).
Ibid.,2,275,843 (Mar. 10, 1942).
Zbid., 2,285,646 (June 9, 1942).
Ibid.,2,286,466 (June 16, 1942).
Colbeth, I. M., Ihid., 2,388,122 (Oot. 30, 1945).
INDUSTRIAL AND ENGINEERING CHEMISTRY
14) Cowan, J. C., IND.ENG.CHEM.,41, 294 (1949).
15) Cowan, J. C., perbonal communication to Don S. Bolley.
16) Craig, J.Am. Chem. Soc., 65, 1006 (1943).
17) Diels, O., andillder, K., Ann., 460, 98 (1928).
18) Dunlap, L. H., U. S. Patent 2,382,213 (Xug. 14, 1945).
19) du Pont de Nemours & Co., E. I., Brit,ish Patent 552,095 (Mar.
(20) Zbid., 556,113 (Sept. 21, 1943).
(21) Ellis, C., U. S. Patent 2,033,131 (Mar. 10, 1936).
(22) Ibid., 2,033,132 (Mar. 10, 1936).
(23) I b i d . , 2,146,671 (Feb. 7 , 1939).
(24) Flint, R. B., and Rothrock, H. S.,Ibid., 2,276,176 (Mar. 10,
Forbes, W. C., and Neville, H. A , , ISD.ENG.CHEM.,32, 555-8
Gardner, H. A , , U. 8. Patent 1,452,553 (1923).
Gerhart, H. L., Zhid., 2,361,018 (Oct. 24, 1914).
Ilemitt, D. H., and Armitage, F., J . Oil & Colour Chemists'
Jordan, O., and Kollek, L., U. S. Patent 2,054,019 (Sept. 8,
Kass, J. P., presented before the Division of Paint, Varnish, and
Plastics Chemistry at the Memphis Section Meeting of the
Memphis, Tenn., 1942.
Assoc., 29, 109 (1946).
Vol. 41, No. 2
Lawler, W. D., Hable, G . J., and Steinle, J. V., U. S. Patent
2,353,910 (July 1944).
(32) Lawson, W. E., and Sandborn, L. J., Ibid., 1,975,959 (Oct. 9,
(33) Mighton, C. J.,Ibid., 2,346,858 (Agiil 18, 1944).
(34) Ibid., 2,401,769 (June 11, 1946).
(35) Milas, N. A., Ibid., 2,267,248 (Dee. 23, 1941).
(36) Morrell, R. S., and Samuels, II., J . Chem. Soc., 1932, p. 2251.
(37) Root, F. B., U.S. Patent 2,374,381 (April 24, 1945).
(38) Rummelsburg, A. L., I b i d . , 2,370,689 (Mar. 6 , 1945).
(39) Scheiber, J., Ibid., 1,979,495 (Nov. 6, 1934).
(40) Scheiber, J., British Patent 316,872 (Nov. 24, 1930).
(41) Schwarcman, A., U. S. Patent 2,412,176 (Der. 3, 1946).
(42) Zbid., 2,412,177 (Dee. 3 , 1946).
(43) Swain, R. A., Ibid., 2,304,288 (Dec. 8 , 1942).
(44) Teeter, H. M.,and Cowan, J. C., Oil h Soap, 22, 177-80 (1945).
(45) Teeter, H. >I., Geerts, M. J., and Cowan, J. C., J . Am. Oil
Chemists' Soc., 25, 158 (1948).
(46) Wakeford, L. E., andHewitt, D. H., U. S. Patent 2,392,710 (Jan.
February 28, 1948
J. C. Cowan
Northern Regional Research Laboratory, Peoria, I l l .
T h e fundamental and practical aspects of the isonierization reactions of the unsaturated acids are discussed.
Of particular interest to the drying oil chemist are a review of the methods of effecting conjugation and evaluating the conjugated oils, and discussions on the drying of
oil films, the relation of isomerization to drying and copolymerization, and the factors responsible for after-tack.
Particular attention is given to the problem of making a
tung oil replacement, to the mechanisms of isomerization, to alkali isomerization, to nickel and iodide catalysts
for isomerization, and to styrene copolymerization.
IKCE early in this century when the process for gasproofing
tung oil was developed, the importance of the isomerization
of conjugated and nonconjugated oils in protective coatings has
steadily increased, Tung oil is easily isomerized and its isomerization has definite commercial importance. When tung oil
is exposed to sunlight or treated with sulfur, selenium, or iodine
a change from a liquid oil to a solid fat is effected. This chznge
of the liquid oil to a solid is the result of the isomerization of t i e
a-eleostearic acid to the @-eleostearicacid. ljeedless t o say, the
producer who is now fortunate enough to have a supply of tung
oil does not desire to have the physical state of his raw mat,erial
changed since he would be unable to handle it in equipment normally available in his American plant (23). American tung oil
when prepwed by extraction is readily isomerized and a heat
treatment is necessary to stabilize it,. Expressed oil benefits
from a similar treatment (66).
When the supply of tung oi. became limited during the Japanese
occupation of China, attempts to prepare replacements for t,ung
oil were made by a large number of investigators. One of the
direct methods of approach which might lead t o a tung oil replacement is the shifting of the unsaturated bonds in nonconjugated oils such as linseed or perilla oils to give conjugated unsatu-
ration. Although some efforts of coinmercial importance have
been made to obtain extracted oils of higher polyunsaturated fatty
acid content, more efforts have been toward isomerizing the oil
t o produce conjugated systems since it was known that the conjugation was primarily reqponsible for the reactivity of tung oil.
In addition to this interest in preparation of conjugated oils
for industrial uce, studies on isomerization reactions havc resulted in a new method for analys's of oils and have extended the
scope of research on the problem of the utilization of vegetable
oils by the preparation of new derivatives.
KOattempt has been made in this paper t o covcr all thc literature on isomerization and conjugation. The many known isomers
of the fatty acids which have been reported in the literature are not
included specifically. This paper deals primarily with the isomerization and conjugation as it is related to the drying oils and
Theoretical Aspects of Isomerization of Fat Acids
The term isomers is generally applied t o those compounds which
have the same molecular formula but which differ in a t least one
of their physical and chemical properties ($6). The term isomerization is applied t o reactions which effect changes among
With the mono-unsaturated fatty acids, the isomeric forms are
limited to the cis-trans isomers of the different positional isomers
with the omega-unsaturated fatty acid existing in only one form.
The different forms of mono-unsaturated fatty acids can be rcpresented as follows: