INTERNATIONAL UNION OF

PURE AND APPLIED CHEMISTRY

APPLIED CHEMISTRY DIVISION

ORGANIC COATINGS SECTION
SUBCOMMITTEE ON ANALYTICAL METHODS

RECOMMENDED METHODS FOR

THE ANALYSIS OF ALKYD
RESINS

LONDON

BUTTERWORTHS

PAC-33-2-K
 FOREWORD
This is the second publication ofthe Subcommittee on Analytical Methods
of the Organic Coatings Section, tbe first being 'Recommended Methods for
the Analysis of Drying Oils'. The trend of the Organic Coatings industry
has been to use oil-modified resins, particularly alkyd resins, in place of the
simpler drying oils and this has brought the need for standard methods for
analysis of these resins. The methods chosen for investigation were those
considered to be of most importance in current practice.
Of the chemical methods, particular attention has been given to the method
for the determination of hydroxyl value, on which there were conflicting
views. In a cooperative exercise, laboratories in five countries obtained
highly satisfactory results by two methods and since there was little to
choose between them both are included.
As envisaged in the earlier publication, increasing use is made of instru-
mental (chromatographic and spectroscopic) methods. In the present state
of development they are of main value for qualitative or semi-quantitative
analysis. The gas chromatographic methods were the subject of a further
cooperative investigation involving laboratories in five countries.
The methods for the determination of phthalic anhydride content and
fatty acids content are essentially equivalent to those of A.S.T.M. 0563-52
and D 1398-58 respectively.
The members of the Subcommittee who took part in the work were
Dr R. Bult (drs F. H. de Ia Court) Netherlands
Dr M. Hochweber Switzerland
Mr H. K. Raaschou Nielsen Denmark
Mr P. Fink-Jensen (Mr G. Christensen) Denmark
Mr B. Remberg Sweden
Mr A. Toussaint Belgium
Dr L. A. O'Neill Great Britain
Thanks are due to Mr A. Toussaint for providing a French translation
and to drs de la Court (T.N.O.) for suggesting the gas chromatographic
methods and providing samples for the cooperative investigation.

L. A. O'NEILL
Chairman of the Subcommittee on Analytical Methods

413
 CONTENTS
1. Acid value 417
2. Saponification value 417
3. Hydroxyl value 418
4. Phthalic anhydride content 421
5. Fatty acids content 422
6. Identification of type of alkyd resin by infrared spectroscopy 425
7. Qualitative analysis of carboxylic acids in alkyd resins by gas
chromatography 428
8. Identification and semi-quantitative determination ofpolyols in
alkyd resins by gas chromatography 431

415
 1. ACID VALUE
1.1 Definition
The acid value is defined as the number of milligrammes of potassium
hydroxide required to neutralize one gramme of the sample under the
conditions specified below.

1.2 Scope
The acid value is a measure of the free carboxyl groups in the alkyd. Free
anhydride groups are also partially measured. The content of anhydride
groups in a finished alkyd is likely to be small but in samples of alkyd of high
acid value taken in the early stage of reaction it may be appreciable. In this
case, the anhydride groups may be determined separately, or the acid and
total anhydride groups determined by an atternative procedure in which the
latter are fully hydrolysed.

1.3 Reagents
Standard potassium hydroxide solution (O.lN) in methanol.
One per cent of phenolphthalein in ethanol (95 per cent). Other indicators
with a colour change in the same pH range (8.3-10) may be used also.
Ethanol-toluene mixture. Equal volumes of ethanol (95 per cent) and
toluene are mixed and neutralized with a methanolic potassium hydroxide
solution (O.lN) in the presence ofphenolphthalein as indicator.

1.4 Procedure
Weigh to the nearest 0.01 g into a conical flask, 1-2 g ofthe sample, accord-
ing to the expected acid value.
Add 50 ml of ethanol-toluene mixture to the sample and shake until solu-
tion is complete.
Titrate with methanolic potassium hydroxide solution in the presence of
pheno1phthalein as indicator.

1.5 Calculation
Calculate the acid value as follows:
Acid value = 56.1 aNfp
a = m1 methanolic potassium hydroxide solution
N = normality of the methanolic potassium hydroxide solution
p = weight of the sample in g.

2. SAPONIFICATION VALUE
2.1 Dermition
The saponification value is defined as the number of milligrammes of
potassium hydroxide required for the saponification of one gramme of the
sample under the conditions specified below.

2.2 Scope
The saponification value is a measure of the sum of the free and hydro~
lysable acids in the alkyd. The determination can be carried out on straight
417
 METHODS FüR THE ANALYSIS OF ALKYD RESINS
and modified alkyds. For simple alkyds of known type it provides a measure
of the oillength.

2.3 Apparatus
Conical flasks of alkali-resistant glass of 200 or 300 ml capacity.
Reflux condensers. The use of interchangeable ground glass joints is
recommended.

2.4 Reagents
Standard hydro.€hloric acid solution (0.5N).
Pot'assium hydroxide solution (0.5N) in ethanol (95 per cent). This solu-
tion must be colourless or not darker than straw yellow.
One per cent of phenolphthalein in ethanol (95 per cent). Other indicators
with a colour change in the same pH range may be used also.

2.5 Procedure
Weigh to the nearest 0.001 g into a conical flask, approximately 1 g of
the sample; dissolve in 5 ml of toluene.
Add exactly 25 ml of ethanolic potassium hydroxide solution to the solu-
tion by means of a burette or another suitable instrument of equal precision,
and 5 inl ofwater. Connect the flask with a condenser and reflux for 1 hour,
shaking from time to time.
Titrate the soap solution, while still warm, with hydrochloric acid solution
in the presence of phenolphthalein as indicator.
Carry out a blank by refluxing and titrating exactly 25 ml of ethanolic
potassium hydroxide solutionunder the same conditions.

2.6 Calculation
Calculate the saponification value as follows:
Saponification value = 56.1 (b - a) Njp
a = ml hydrochloric acid required for the titration of the sample
b = ml hydrochloric acid required for the titration of the blank
N = normality of the hydrochloric acid solution
p = weight of the sample in g.

3. HYDROXYL VALUE
3.1 Dermition
The hydroxyl value is defined as the number of milligrammes of potassium
hydroxide required to neutralize the acetic acid taken up on acetylation of
one gramme of the sample. The hydroxyl content is commonly expressed in
these units, in conformity with the acid and saponification values.
For comparison with hydroxyl contents determined by methods other
than acetylation, e.g. active hydrogen or infrared spectroscopic methods,
the hydroxyl content may be expressed as the percentage of hydroxyl
groups, by multiplying the hydroxyl value by the factor 17/560.

3.2 Scope
Determination of the free hydroxyl groups present in the alkyd, these being
418
 METHODS FOR THE ANALYSIS OF ALKYD RESINS
present as free polyhydric alcohol, partial esters, polyester end groups or
hydroxylated fatty acids. Two equally satisfactory methods are proposed.

3.3 Method 1
3.3.1 Apparatus
Glass-stoppered conical flasks of ab out 250 ml capacity.
Reflux condensers. The use of ground glass joints is required.
Waterbath.

3.3.2 Reagents
Standard potassium hydroxide solution (0.5N) in methanol. Standardize
this solution against potassium acid phthalate.
Acetylating reagent. Dissolve 10 ml of redistilled acetic anhydride in
90 ml of redistilled pyridine. Make sure that the titration of 10 ml of this
reagent requires a volume between 40 and 50 ml of the methanolic potassium
hydroxide solution.
One per cent of phenolphthalein in ethanol (95 per cent).
Potassium acid phthalate, primary standard grade.
Toluene-butanol mixture, 1:2 (v/v). Neutralize this mixture with methan-
olic potassium hydroxide solution in the presence of phenolphthalein as
indicator.

3.3 .3 Procedure
Weigh the sample to the nearest 0.001 g into a dry conical flask. The
weight of the sample is chosen so that 3.0 to 3.5 milliequivalents of hydroxyl
are present; this corresponds in g with 170, divided by the expected hydroxyl
value.
Add exactly 10 ml of acetylating reagent by means of a pipette. Connect
the flask with a condenser and heat on a boiling water bath for 2 hours,
shaking gently in the initial stages of the heating to promote solution of the
sample in the reagent.
Add 10 ml of cold distilled water down the condenser and continue the
heating forafurther 15 minutes.
Cool the flask by immersion in a beaker ofcold water and, while immersed,
add 75 ml of toluene-butanol mixture down the condenser so as to wash
thoroughly the condensing surface of the condenser.
Remove the condenser from the flask and allow the contents of the flask
to cool to room temperature.
Titrate with methanolic potassium hydroxide solution in the presence of
phenolphthalein as indicator.
Carry out a blank with exactly 10 ml of acetylating reagent under the
same conditions.

3.3 .4 Calculation
Calculate the hydroxyl value as follows:

(b - a) N X 56.1
Hydroxyl value = + A.V.
p
419

PAC-33-2-L
 METHODS FOR THE ANALYSIS OF ALKYD RESINS
a = ml methanolic potassium hydroxide solution required for the titra-
tion of the sample
b = m1 methanolic potassium hydroxide solution required for the titra-
tion of the blank
N = normality of the methanolic potassium hydroxide solution
p = weight of the sample in g
A.V. = the acid value of the sample.

3.4 Method II
3.4.1 Apparatus
Glass-stoppered conical flasks of about 250 m1 capacity.
Bath adjusted at 50°C ( ± 1oc).

3.4.2 Reagents
Standard potassium hydroxide solution (0.50N) in methanol. Standardize
this solution against potassium acid phthalate.
Acetylating reagent. Dissalve 4.0 g of p-toluene sulphonic acid
(CH 3 C6 H 4 S03 . H 2 0) in 100 m1 of ethyl acetate (dried over barium oxide and
distilled), preferably by means of a magnetic stirrer. To this solution, add
slowly while stirring 33 ml of distilled acetic anhydride. Make sure that the
titration of 5 m1 of this reagent requires a volume between 40 and 50 ml
of the methanolic potassium hydroxide solution.
One per cent of phenolphthalein in ethanol (95 per cent).
Potassium acid phthalate, primary standard grade.
Toluene-butanol mixture, 1 :2 (vjv). Neutralize this mixture with metha-
nolic potassium hydroxide solution in the presence of phenolphthalein as
indicator.
Pyridine-water mixture, 3:1 (vjv)
Ethylacetate (dried over barium oxide and distilled).

3.4 .3 Procedure
Weigh the sample to the nearest 0.001 g into a dry conical flask. The weight
ofthe sample is chosen so that 5 to 6 milliequivalents ofhydroxyl are present;
this corresponds in g with 280, divided by the expected hydroxyl value.
However, do not use a sample weight of more than 10 g, as would be the
case with resins with hydroxyl value below 28.
Add 5 m1 of ethyl acetate and shake, if necessary under gentle warming,
until the sample is dissolved.
Allow to cool to room temperature, add exactly 5 ml of acetylating re-
agent by means of a pipette and stopper the flask.
Place the flask for twenty minutes in a bath adjusted at 50°C ( ± 1°C) in
such a way that the liquid in the flask is just above the level of the liquid in the
bath and shake every 5 minutes.
Remave the stopper, add 2 ml of waterandshake vigorously.
Add 10 ml of pyridine-water mixture, rinsing the sides of the flask during
the addition.
Allow the flask to stand for 5 minutes at room temperature, add 60 ml
of toluene-butanol mixture and titrate with methanolic potassium hydroxide
solution in the presence of phenolphthalein as indicator.
420
 METHODS FOR THE ANALYSIS OF ALKYD RESINS
Carry out a blank with exactly 5 ml of acetylating reagent under the same
conditions.

3.4.4 Calculation
See method I.

4. PHTHALIC ANHYDRIDE CONTENT

4.1 Scope
In this method the phthalic anhydride content of alkyd resins including
those containing styrene is determined. This method is not applicable for
the determination of phthalic anhydride in alkyd resins containing other
dibasic acids such as maleic or fumaric, or modifying agents such as urea,
melamine, and phenolic resins.

4.2 Apparatus
A 500 ml conical flask fitted with an air-cooled glass reflux condenser
30 in in length.
Waterbath.
A sintered glass filter crucible, fine porosity, of 30 ml capacity.
A desiccator containing concentrated H 2 S04 as the desiccant.
A guard tube filled with soda lime.
Filter flasks.
Crucible holder.

4.3 Reagents
Benzene.
Alcoholic potassium hydroxide solution. Dissolve 66 g of reagent grade
KOH in 1 litre of absolute ethyl alcohol (Note 1). Allow the solution to
stand ovemight protected against C02 absorption. Filter just before use.
Anhydrous analytical reagent grade ether.
Alcohol-benzene wash solution. One volume of absolute ethyl alcohol is
mixed with three volumes of benzene.
Hydrochloric acid (O.lN).
4.4 Procedure
Weigh by difference, from a closed container into the 500 ml conical
flask, a sample of resin or resin solution sufficient to yield from 0.8 to 1.2 g
of potassium alcohol phthalate. Add 150 m1 of benzene, warming slightly
on the steam bath, if necessary, to effect solution. Add 60 m1 of alcoholic
KOH solution, and attach the condenser.
Place the flask in a water bath to a depth approximately equal tothat ofthe
contents of the flask. Warm the bath, maintaining a temperature of 40°C
for 1 hour, then gradually raise the temperature until the alcoholic solution
boils gently. Reflux for 1! hours.
Remove the flask from the bath and wash down the inside of the condenser
with a few millilitres of alcohol-benzene wash solution. Remave the con-
denser, cap the flask with the soda-lime guard tube, and cool by means of
running water or an ice bath.
421
 METHODS FOR THE ANALYSIS OF ALKYD RESINS
When cool, filter immediatefy, and as rapidly as possible, through a
sintered glass crucible that previously has been weighed, using the alcohol-
benzene wash solution for transferring the precipitate and washing the
reaction flask. Wash the precipitate with successive portions of alcohol-
benzene wash solution until a few millitres of washings collected in a second
suction flask are no Ionger alkaline to phenolphtha1ein. (Normally about 75
ml of wash solution are :sufficient.) Do not allow air to be drawn through the
crystals, as they are hygroscopic. Finally pour 25 ml of ether into the crucible
and draw through the precipitate with the aid of suction.
Wipe the outer surface of the crucible with a clean cloth and place in a
gravity convection oven at 60°C for 1 hour (Note 1).
Cool to room temperature in a desiccator, and weigh.

Note 1
The precipitate is the alcoholate C 6 ~ (COOK)2 • (C 2 H 5 0H), and the
alcohol of crystallization will be slowly driven off on prolonged heating.
lt is safe, however, to dry the alcoholate at a temperature up to 60°C for as
long as 1 hour.
Correction for carbonate-coprecipitation of K 2 C03 with the potassium
alcoho1 phthalate may be a source of error. If a correction for K 2 C03 is
desired, proceed as follows:
Dissolve the weighed precipitate in about 50 mi of distilled water that has
been neutralized to phenolphthalein. Add 3 to 4 drops of phenolphthalein
indicator, and if the solution is a1kaline, titrate with 0.1N HCI.

4.5 Calculation
Calcu1ate the percentage of phthalic anhydride in the sample as follows:
K = VN x 0.1382
Phthalic anhydride, per cent
= (P- K) X 0.5136 X IOO
s
K = correction for K 2 C0 3 , in grammes (if determined)
V= millilitres of HCI used for titration
N = normality of H Cl
P = grammes of potassium alcohol phthalate
S = grammes of sample used.

5. FATTY ACIDS CONTENT

5.1 Scope
This method covers the gravimetric determination of the total fatty acid
content of alkyd resins free from modifying agents.

5.2 Apparatus
A 500 m1 conical flask fitted with an air·cooled glass reflux condenser
30 in in length.
Water bath and steam bath.
422
 METHODS FOR THE ANALYSIS OF ALKYD RESINS
A sintered glass filter crucible, fine porosity of 30 ml capacity.
Crucible holder.
A desiccator containing concentrated H 2 S04 as the desiccant.
A guard tube filled with soda lime.
Filter flasks, filter funnel and filter papers.
Beakers, 150 and 400 ml.
Three 500 ml separating funnels.
Vacuum drying oven.
Nitrogen gas.
5.3 Reagents
Anhydrous sodium sulphate.
Universal indicator paper.
Benzene.
Alcoholic potassium hydroxide solution. Dissalve 66 g of reagent grade
KOH in 1litre of absolute ethyl alcohol (Note 1). Allow the solution to stand
overnight protected against C02 absorption. Filter just before use.
Ether-anhydrous analytical reagent grade ether.
Alcohol-benzene wash solution. Mix one volume of absolute ethyl
alcohol with three volumes of benzene.
Hydrochiarie acid (sp.gr. = 1.19).
Note 1 ,
The alcohol may be denatured but must be absolute.
5.4 Procedure
Weigh by difference, from a closed container into the 500 m1 conical
flask, a sample of resin or resin solution sufficient to yield from 0.8 to 1.2 g
of potassium alcohol phthalate. Add 150 m1 of benzene, warming slightly on
the steam bath, if necessary, to effect solution. Add 60 m1 of alcoholic KOH
solution and attach the condenser.
Place the flask in a water bath to a depth approximately equal to that of
the contents of the flask. Warm the bath, maintaining a temperature of
40°C for 1 hour then gradually raise the temperature until the alcoholic
solution boils gently. Reflux for 1! hours.
Remave the flask from the bath, and wash down the inside of the condenser
with a few millilitres of alcohol-benzene wash solution. Remave the con-
denser, cap the flask with the soda-lime guard tube, and cool by means of
running water or an ice bath.
When cool, filter immediately, and as rapidly as possible, through a
sintered-glass crucible that previously has been weighed, using the alcohol-
benzene wash solution for transferring the precipitate and washing the
reaction flask. Wash the precipitate with successive portions of alcohol-
benzene wash solution until a few millilitres of washings collected in a second
suction flask are no Ionger alkaline to phenolphthalein. (Normally about
75 m1 of wash solution are sufficient.) Do not allow air to be drawn through
the crystals, as they are hygroscopic. Finally pour 25 m1 of ether into the
crucible and draw through the precipitate with the aid of suction.
Transfer the combined filtrate and washings to the 400 ml beaker with
the aid of 25 m1 of water from a wash bottle.
423
 METHODS FOR THE ANALYSIS OF ALKYD RESINS

Concentrate on the steam bath to a volume of approximately 25 ml under

a blanket of nitrogen to prevent oxidation of the fatty acids using a hood.
Transfer to a 500 ml separating funnel with the aid of water from a wash
bottle, dilute with water to approximately 300 ml, and add 10 ml of alcohol.
Extract the unsaponifiable and volatile thinners with successive 50 ml
portians of ether (not less than three, or until a colourless ether extract is
obtained), combining the ether extracts in the first separating funnel and
using the other two funnels for the successive extractions (Note 2). Finally,
wash the combined ether extracts with three 15 ml portians of water, adding
the water washes to the main aqueous phase. Discard the combined ether
extracts.

Note2
If the layers do not separate easily, carefully draw otf the lower, clear
aqueous layer and add 2 to 3 ml of alcohol, by means of a pipette, to the
ether emulsion phases in the separating funnel. Swirl gently to break the
emulsion, and continue to draw otf the lower layer. This procedure for
breaking the emulsion may be repeated on subsequent extractions, if neces-
sary.
Acidify the aqueous phase to a pH of approximately 2 by slowly adding
HCI, cooling under running tap water. When the mixture has cooled to
room temperature, extract the fatty acids with successive 25 ml portians
(not less than three) of ether until a colourless ether extract is obtained,
combining the ether extracts in the first separating funnel and using the other
two funnels for the successive extractions. Wash the combined ether exgacts
with successive I 0 ml portians of water until free of mineral acid when tested
with an indicator paper. Discard the aqueous phase.
Dry the combined ether extracts in the separating funnel by the addition
of successive small quantities of anhydrous Na2 S04 •

Note3
The free water will have been removed when, by the addition of a small
quantity of Na2 S04 and gentle swirling, the excess Na2 S04 will be seen to
disperse as a freely moving powder.
Filter the dried ether extract through rapid, low ash paper, portionwise
into the 150 ml beaker, containing a small boiling stone and previously
weighed to the nearest 1 mg. The ether extract should be decanted from the
top opening of the separating funnel. Evaparate the ether portionwise by
placing the beaker and its contents on the steam bath, in a hood. Cover the
filter funnel with a watch glass during the portionwise evaporation steps
and apply a blanket of nitrogen over the beaker during evaporation. Re-
move the last portians of fatty acids from the N a 2 S04 by washing with
successive small portians of ether until a colourless extract is obtained.
Remave the final traces of fatty acids from the filter paper by using several
successive small portians of ether.
Complete the evaporation of the fatty acid ether solution on the steam
bath, while maintaining a nitrogen atmosphere over the acids. Remave the
final traces of ether by heating for successive 20 minute periods in a vacuum
oven at 60°C until minimum weight is obtained. After each heating period,
424
 METHODS FOR THE ANALYSIS OFALKYD RESINS
allow the beaker and contents to ·cool in a desiccator and then weigh to the
nearest milligramme.

5.5 Calculation
Calculate the total fatty acids content as follows:
A--B
Total fatty acids, per cent = --w x 100

A = weight ofbeaker plus residue

B = weight of beaker \
W = weight of sample taken.

6. IDENTIFICATION OF TYPE OF ALKYD RESIN BY

INFRARED SPECTROSCOPY
The infrared spectrum provides information on the type of alkyd resin
under examination. The main constituents from which the resin is made
and any modifying agents can often be recognized. The evidence depends
on the presence and intensity of particular bands in the spectrum. In some
cases, the bands are sufficiently characteristic to enable components to be
identified with certainty; in others the interpretation is more doubtful, and
the bands for some modifying agents may be masked by those of the parent
alkyd.
In some cases, where the nature of the components of the alkyd resin are
known, quantitative estimations of oillength or of the proportion of modify-
ing agents may be made.

6.1 Preparation of sample

The spectrum of the alkyd resin may be obtained by any of the standard
methods of infrared spectroscopy. The simplest method is to examine the
resin in film form; foraliquid resin this can be done by pressing the sample
between two sodium chloride plates and for a solid by evaporation from
solution on to a sodium chloride plate or potassium bromide disc (taking
care to remove residual solvent under vacuum or in a current of inert gas
at 50°C).
The resin may also be examined in solution (in a solvent, the spectrum
of which causes minimum interference, such as CS2 or CC14 ), as a mull with
an inert liquid (liquid paraffm or a fluorinated hydrocarbon), or as a pressed
disc with inorganic powder (potassium bromide).
lf a wedge-shaped film of the sample is prepared this facilitates the choice
of the most suitable film thickness.
In all cases resin films should be compared among themselves or with
known standards at similar film thicknesses.
These procedures will give the spectrum of the original alkyd, but if the
alkyd is of an oxidizing type and the film is allowed to dry in air, chemical
changes occur and the spectrum alters. Account must be taken of this in
comparing the spectrum of a dried alkyd fthn with that of the original alkyd.
A special technique for obtaining the spectnun of a film already attached
425
 METHODS FOR THE ANALYSIS OF ALKYD RESINS
to a substrate is by attenuated total reflectance (a.t.r.) spectroscopy. In
interpretation it must be remernbered that the spectrum obtained is that of
the upper layers of the film only and there are small differences between
the spectra obtained by a.t.r. and transmission methods.

6.2 Interpretation of spectra

The spectrum obtained will show general features characteristic. of most
alkyd resins and special features which assist in the identification of particu-
lar components.
The general bands are in the following regions:
Band region ( {Jm) Causalive group
2.8 to 2 ·95 Hydroxyl
3.3 to 3.4 Hydrocarbon
5.7 to 5.9 Esterandacid
6 to 6.75 Aromatic ring
6.75 to 7.5 Aliphatic hydrocarbon
7.5 to 10 Ester
10 to 10.5 Trans unsaturation
10.5 to 14·5 Aromatic ring
The extent to which particular bands give an indication of particular
components is considered below for each group, namely polybasic acids,
monobasic acids, polyhydric alcohols and modifying agents.

6.3 Polybasic acids

Clear distinction can be made between o-phthalic and isophthalic acid
groups from the following bands: ·
o-phthalic 6.2, 6.3, 7.9, 8.9, 9.3, 13.5, 14.2 ~m
isophthalic 6.2, 7.7, 8.2, 8.6, 9.1, 13.7 ~m
lndications of the presence of other polybasic acids can sometimes be
obtained by the band pattem in certain regions. Thus trimellitic anhydride
gives a band at 12.6 J..lffi.
No bands generally reliable for the recognition of maleic anhydride modi-
fication of an alkyd have been found, but bands at 3.1, 6.05 and 12.9 ~m
are sometimes indicative. Free anhydride groups, whether from residual
phthalic anhydride or maleic anhydride give bands at 5,4 and 5.65 ~m.

6.4 Monobasic acids

The different fatty acids in an alkyd resin cannot readily be distinguished
from theinfrared spectra. The most useful region is 10 to 10.5 J.liD which is
indicative of trans unsaturation. This can be characteristic for tung oil or
dehydrated castor oil but for non-conjugated natural oils, such as linseed or
soya, it indicates only trans unsaturation developed during the resin manu-
facture.
In this regiön, unsaturated oils of this type will show a band at 10. 3 ~m
characteristic of monoene or non-conjugakd trans unsaturation.
Tung oil and dehydrated castor oil acids will however show bands of con-
jugated trans unsaturation at 10.1 ~m.
426
 METHODS FüR THE ANALYSIS OF ALKYD RESINS
In alkyds based on saturated fatty acids these trans bands will be completely
absent.
Oils of hydroxylated fatty acids (castor oil) will show a strong hydroxyl
band at 2.9 J..Lffi but this is not distinguishable from that due to the unesteri-
fied hydroxyls of the polyhydric alcohol. Of other monobasic acids, benzoic
acid, in conjunction with o-phthalic acid, gives a band at 14.05 J..lffi com-
pared to one at 14.2 J..LID in the presence of the o-phthalic acid alone and
also at 9.7 J..Lm.

6.5 Polyhydric alcohols

Assistance in the recognition of the polyhydric alcohol component is
sometimes obtained from the i.r. spectrum.
Glycerol may often be distinguished from trimethylol ethane (TME),
trimethylol propane (TMP), pentaerythritol and mixtures of these polyals
with glycerol, by the shape of the band at 7.25 J..Lffi. This is sharp and very
5mall with glycerol, but broader when pentaerythritol is present and con-
siderably Ionger when trimethylol ethane and trimethylol propane (TME
and TMP) are present.
Furthermore, in spectra of alkyds containing pentaerythritol, TMP and
TME the CH 2 -bands and CH 3 -bands at 6.9 J..Lm have been shifted towards
lower wayelengths with the result that the benzene ring at 6. 7 J..LID, which is
shown as a distinct sharp peak in the spectrum of the glycerol alkyd, appears
only as a shoulder in the spectra ofpentaerythritol, TMP and TME.
Finally, the spectrum of a glycerol esterified alkyd is characterized by
an absorption at 9.1 J..Lm (which partly eliminates the minimum at 9.15 J..lffi)
which is characteristic of spectra of alkyds containing pentaerythritol, TMP
and TME.
When mixtures of glycerol and pentaerythritol are present in alkyds the
spectra are difficult or impossible to distinguish from spectra of alkyds
containing pentaerythritol as the only polyol.

6.6 Modifying agents

Modification of the alkyd with styrene or vinyltoluene can be recognized.
In o-phthalic based alkyds styrene causes a shift of the 14.2 J..Lffi band to
14.3 J..Lffi with increase of intensity; the band at 6.2 J..Lffi becomes stronger
than the 6.3 J..Lffi band, instead of weaker as in unmodified alkyds; vinyl-
toluene also increases the intensity of the 14.2 J..Lffi band and gives a band
at 12.8 J..Lm. New bands are introduced at 8.6 J..lffi and 12.3 J..lffi.
Modifications with rosin or with phenolic resin do not cause specific
changes in the spectrum, to enable their recognition.
Modifications with acrylate or methacrylate esters may sometimes be
recognized by bands at 8.3, 8.7 and 10.1 J.Lm.
The presence of urea-formaldehyde condensate in an alkyd resin can be
detected by a band at 6.1 J..Lffi and ofmelamine formaldehyde by a sharp band
at 12.25 J..Lffi.
Silicone modified alkyds show a broad absorption at 8.5 to 10 J.Lm.

6. 7 Quantitative analysis
An estimation ofthe oillength ofthe alkyd may be obtained by determining
427
 METHODS FOR THE ANALYSIS OF ALKYD RESINS

the ratio:
C-H absorbance at 3.4 J.!ID
C 0 absorbanc~at 5. 75 J.!ID
and reading the oillength from a calibration graph. Results are more accurate
if a graph specific for the type of oil in the alkyd is used.
Estimation of the styrene and vinyltoluene contents of modified alkyds
may be1obtained from the ratios of absorbances at 14.2 J.!ID and 12.4 Jlffi
respectively to that at 5. 75 J.tm.

7. QUALITATIVE ANALYSIS OF CARBOXYLIC ACIDS IN ALKYD

RESINS BY GAS CHROMATOGRAPHY
7.1 Scope
This method covers the qualitative determination of the carboxylic acids
in alkyd resins, but may be also used for analysing polyesters.

7.2 Principle of the method

The acids in the resin are converted directly to the methyl esters with
Iithium methoxide in methanol.
The esters so formed are extracted with methylene chloride and submitted
to gas chromatography on a polar and on a non-polar column.
The method has been worked out for a series of carboxylic acids, and the
retention times relative to triacetin are given in Table 1. These retention
times have to be considered as a guide to a satisfactory analysis, since varia-
tions with respect to apparatus, substrates, stationary phases, etc. may Iead
to deviations.

7.3 Reagents and auxiliaries

Methanol, absolute
Lithium
Sulphuric acid, 3M
Methylene chloride, pure
Saturated sodium chloride solution
Toluene, pure
Chloroform, pure
Magnesium sulphate, free of water
Triacetin.

7.4 Apparatus
Any linear programmed temperature chromatograph should be suitable,
if adaptable to the operating conditions described below (Note 1).

Chromatography on apolar column

1. Column: stationary phase diethylene glycol succinate, Carbowax
20M
solid support Chromosorb W (sil.)
length 2m
diam. (internal) 4.6 mm
428
 METHODS FOR THE ANALYSIS OF ALKYD RESINS

2. lnjection port temperature 320 oc

3. Column temperature 125-220 oc (rising 4 °C/min)
4. Carriergas He (flow rate 60 ml/min) (Note 2)

Chromatography on a non-polar column

1. Column: stationary phase silicone grease DC II
solid support Chromosorb W (sil.)
length 2m
diam. (internal) 4.6 mm
2. Injection port temperature 320 oc
3. Column temperature 75-250 oc (rising 4 °C/min)
4. Carriergas He (flow rate 60 ml/min) (Note 2)

7.5 Column preparation

Polar column
Dissalve the diethylene glycol succinate (Note 3) in toluene and add the
solution to the Chromosorb W (Note 3) in a porcelain dishin such a way
that the weight ratio, stationary phase: solid support, is 10: 90.
The solid support particles must be under the surface of the solution.
Remave the solvent on a water bath or with a stream of warm air while
stirring the mixture gently with a spatula. After removing the solvent,
heat in an oven for one night at 105°C.
Prepare in the same way as above a stationary phase of Carbowax 20M
(Note 3) on Chromosorb W. Use chloroform instead of toluene as a solvent.
About 7.5-8 g of stationary phase and solid support is needed to fill a
column of 2 m length and 4.6 mm (intemal) diameter.
Pack the column subsequently in the following way:
Bring a piece of glass-wool into a 'U' shaped stainless steel column of the
prescribed dimensions, dividing it approximately into two equal parts. One
part is filled through a funnel with the polyester-containing phase, by gravity.
The sides of the column are vibrated to assist the flow of the packing material.
The other part is filled in the same way with the carbowax-containing phase.
Mount the column so that sample passes first through the polyester section.

Non-polar column
The stationary phase of the non-polar column is made in the same way as
the polar column.
In this case the weight ratio, silicone grease: Chromosorb W (Note 3)
is as before, 10:90.
Use chloroform as a solvent.
Packing of the column must be carried out by plugging one end of the
column with a piece of glass-wool or a stop-cork and filling and vibrating
as described before.

Note 1
If the apparatus is equipped with a hot wire type of detector it may be
necessary sometimes, when only very small amounts of acids are present,
to use a high er liquid phase :solid support ratio, e.g. 20 :80. This is due to the
lower sensitivity of this type of detector compared with flame ionization.
429
 METHODS FüR THE ANALYSIS OF ALKYD RESINS

This may Iead to deviations from the retention times given in Table 1.

Note 2
The use of helium as an elution gas is not necessary but the use of other
gases may Iead to different retention times (Table 1).

Note 3
Diethylene glycol succinate, Carbowax 20M ( = polyethyleneglycol, m.wt.
20000) and silicone grease DC 11, are all commercially available. Chromo-
sorb W (60-80 mesh; sil.) is a diatomaceous earth, specially treated and
silanized. 1t is also commercially available.

7.6 Procedure
7.6.1 Sampiepreparation
In a round-bottomed flask weigh 0.3 g of alkyd resin, free of solvent,
and add 15 ml of 0.5M Iithium methoxide in methanol (Note4). Dissalve
the resin by heating on a water bath, and reflux for a few minutes. Then add
5 ml of 3M H 2 S0 4 , and transfer the contents of the flask to a separating
funnel; dilute with water to 50 ml, add 35 ml ofmethylene chloride and shake.
After separation of the methylene chloride layer, repeat the extraction with a
further 35 ml of methylene chloride and combine the extracts. Wash this
solution three times with 15 ml of saturated sodium chloride solution, dry
over magnesium sulphate and distil ofT the methylene chloride in excess.

7.6.2 Chromatography
The recovered methyl esters are submitted directly to gas chromatography
on polyester-carbowax as weil as on silicone grease, following the operating
conditions described in 7 .1.
A drop of triacetin is added as in ternal standard.
A chromatogram without triacetin has alsotobe obtained, under the same
conditions.
The amount of sample to be brought on the columns depends on the relative
amounts of the acids in the sample and its concentration.

Note4
The Iithium methoxide is prepared by adding small pieces of metallic
Iithium to absolute methai1ol, which is cooled in an ice bath, until a 0.5M
solution is obtained.

7.6.3 ldentification
ldentification is carried out with reference to the relative retention times
listed in Table 1.
430
 METHODS FOR THE ANALYSIS OF ALKYD RESINS

Table 1. Relative retention times for methylesters of carboxylic acids

Methyl ester of acid Polyester carbowax Silicone grease

column column
(triacetin = 1 ; (triacetin = 1 ;
ret. time = 11 min.) ret. time = 16 min~

succinic 0.26 0.38

benzoic 0.27 0.50
maleic 0.36
p-methyl benzoic 0.42 0.75
lauric 0.50 1.34
adipic 0.54 0.82
p-t-butyl benzoic 0.67 1·18
hexahydrophthalic 0.73 1.07
tetrahydrophthalic 0.89 1.08
p-hydroxybenzoic 0.97 1.05
azelaic 1·00 1.36
endomethylene tetrahydrophthalic 1.08 1.19
palmitic 1.15 2.04
sebacic 1.21 1.52
terephthalic 1.22 1.30
isophthalic 1.31 1.33
o-phthalic 1.31 1.20
stearic 1.53 2.34
oleic 1.53 2.30
linoleic 1.58 2.30
citric 1.65 1.22
linolenic 1.70 2.30
trimellitic 2.37 1.83

As it is not possible to standardize the method completely with respect

to all possible variables, these retention values have to be considered as a
guide. The real values must be determined by each operator under his con-
ditions.

Precision and accuracy

The method described has been investigated in collaborative work by five
laboratories. The constituents of two alkyds were determined correctly in
most cases. Quantities of fatty acids as low as 5 per cent could be detected.

8. IDENTIFICATION AND SEMI-QUANTITATIVE DETERMINATION

OF POLYOI.S IN ALKYD RESINS BY GAS CHROMATOGRAPHY

8.1 Scope
This method covers the qualitative and semi-quantitative determination
of the following polyals in alkyd resins :ethylene glycol, glycerol, trimethylol
ethane, trimethylol propane, pentaerythritol and dipentaerythritoL
In principle it can be used for other polyals and for analysing polyesters.

8.2 Principle of the method

The alkyd resins are subjected to aminolysis with ß-phenylethylamine and
the liberated polyals acetylated. The polyol acetates are separated by
431
 METHODS FüR THE ANALYSIS OF ALKYD RESINS
gas chromatography and identified from the relative retention time with
reference to the second of the peaks given by the reagent. The amount is
determined with the help of an internal standard.
The relative retention times given in Table 2 have to be considered as a
guide to satisfactory analysis since variations with respect to apparatus,
substrates, stationary phases, etc. may lead to deviations.

8.3 Reagents and auxiliaries

ß-phenylethylamine, pure.
Acetic anhydride, pure (distilled before use).
Various polyols, gas chromatographically pure.

8.4 Apparatus
Any linear programmed temperature gas chromatograph should be
suitable, if adaptable to the operating conditions described below (Note 1).

8.5.1 Analysis of glycerol, ethylene glycol, trimethylol ethane, trimethylol

propane and pentaerythritol
1. Column: stationary phase Carbowax 20M
solid support Chromosorb W (60---80 mesh, sil.)
length 2.40m
diam. (internal) 4.6mm
2. lnjection port temperature 320°C
3. Column temperature 70-225°C (rising 8°Cjmin)
4. Carrier gas He (flow rate 65 mljmin) (Note 2)

8.5.2 Analysis of dipentaerythritol

1. Column: stationary phase silicone gum ruhher SE 30
solid support Chromosorb W (60---80 mesh, sil.)
length 1.20.m
diam. (internal) 4.6mm
2. lnjection port temperature 330°C
3. Column temperature 150-290°C (raising 6°Cjmin)
4. Carrier gas He (flow rate 65 mljmin) (Note 2)
8.6 Column preparation
8.6.1 Carbowax column
Dissalve the Carbowax 20M (Note 3) in chloroform and add the solution to
the Chromosorb W (Note 3) in a porcelain dishin such a way, that the weight
ratio, stationary phase: solid support, is 10: 90.
The solid support particles must be under the surface of the solution.
Remave the solvent on a water bath or with a stream of warm air while
stirring the mixture gently with a spatula. After evaporation of the solvent,
heat in an oven for one night at 105°C.
About 9-9.5 g of stationary phase solid support is needed to fill a column
of 2.40 m length and 4.6 mm (inside) diameter.
Packing of the column must be carried out by plugging one end of the
column with a piece of glass-wool or a stop-cork and filling through a funnel
by gravity.
432
 METHODS FOR THE ANALYSIS OF ALKYD RJ?SINS

The sides of the column are vibrated to assist the flow of the packing
material.

8.6.2 Silicone gum ruhher column

Prepare in the same way as described above, a silicone gum ruhher SE 30
(Note 3)/chromosorb W column.
Use toluene instead of chloroform as a solvent.

Note 1
If the apparatus is equipped with a bot wire type of detector it may be
necessary sometimes, when only very small amounts of polyols are present,
to use a higher liquid phase :solid support ratio, e.g. 20 :80. This is due to the
lower sensitivity of this type of detector compared with flame ionization.
This may lead to deviations from the retention times given in Tahle 2.

Note 2
The use of helium as an elution gas is not necessary. The use of other
gases may lead to different retention times (Tahle 2).

Note3
Silicone gum ruhher GESE-30 (methyl silicone gum rubber) and Carho-
wax 20M (polyethylene glycol, m.wt. 20000), are bothcommercially availahle.
Chromosorb W (60-80 mesh; sil.) is a diatomaceous earth, specially treated
and silanized. It is also commercially availahle.

8.7 Procedure
8.7.1 Sampiepreparation
Add to 0.5 to 1 g ofthe alkyd resin, free of solvent, about 0.1 g of an internal
Standard (Note4) both weighed to thenearestmilligramme. To themixtureadd
4 ml of ß-phenylethylamine and heat in a round-bottom flask with ground-
glass joint for Thours (Note 5) under reflux.
Add to the cooled mixture 25 ml of acetic anhydride, and heat for 2.5 hours
under reflux.
After cooling, analyse the mixture of polyol acetates as such by gas
chromatography.

8.7.2 Chromatography .
The polyol acetates are suhmitted to gas chromatography following the
operating conditions descrihed in 8.5.1 (in the case of dipenta, 8.5.2).
The amount of sample to he brought on the columns depends. on the relative
amounts of the polyols present in the sample and its concentration.

Note4
As an internal standard a polyol not present in the alkyd must be used.
Therefore in the first place sample preparation has tobe carried out without
internal standard in order to identify the polyols in the alkyd. In the case of
dipentaerythritol, pentaerythritol must be used as internal standard.
Because in most cases both are present, this can be done by determining
433
 METHODS FOR THE ANALYSIS OF ALKYD RESINS

pentaerythritol on the carbowax column and using the value found to

calculate the dipentaerythritol from the result on silicone gum ruhher.

Note 5
In the case of dipentaerythritol this is 4 hours.

8. 7.3 Identification
ldentification is carried out with reference to the relative retention times
listed in Table 2 (Note 4).
Table 2 a. Relative retention times of polyol acetates on a
column of carbowax 20M

(second reagent peak 1.00 (Note 6); retention time =

=
22 min)
1,2 propanediol 0.46
ethylene glycol 0.48
neopentyl glycol 0.52
diethylene glycol 0.69
glycerol 0.73
trimethylol ethane 0.81
trimethylol propane 0.85
triethylene glycol 0.91
1st reagent peak 0.96
2nd reagent peak 1.00
pentaerythritol 1.09

Table 2b. Relativeretention time of dipentaerythritol ace-

tate on a SE 30 column

(penta = 1.00; retentiontime = 9 min)

pentaerythritol 1.00
dipentaerythritol 2.50

As it is not possible to standardize the method completely with respect

to all possible variables, these retention values have to be considered as a
guide. The real values have to be determined by each operator under his
own conditions.

Note 6
Ifidentification required cf. De La Court, Farbe und Lack, 15, 218 (1969).

8. 7.4 Calculation
The weight percentage of polyol in the non-volatile resin is calculated
from the formula:

Per centpolyol = fx X Ox X g1 X 100

01 gm
fx = conversion factor for the polyol in relation to the internal standard
Ox = area under the Chromatographie peak of the determined polyol
434
 METHODS FOR THE ANALYSIS OF ALKYD RESINS

01 = area under the chromatographic peak of the internal standard

g1 = weight of the internal standard
gm = weight of the non-volatile alkyd resin.

Precision and accuracy

The method described has been investigated in collaborative work by five
laboratories.
The constituents of two alkyds were identified correctly in most cases.
Most laboratories found quantitative values lying within 10-15 per cent of
the amount of polyol actually present.
The method is therefore considered as semi-quantitative.

435